FEDERAL COURT OF AUSTRALIA

SNF (Australia) Pty Limited v BASF Australia Ltd [2019] FCA 425

File number(s):

VID 211 of 2016

VID 1318 of 2016

Judge(s):

BEACH J

Date of judgment:

27 March 2019

Catchwords:

PATENTS – appeals from decisions of the delegate – rehearing de novo – opposition to grant of patent applications – mining processes – treatment of waste material slurries – disposal of tailings – secondary flocculation – co-disposal of coarse and fine solids as a homogeneous mixture – thickener underflow – process for improving rigidification of material whilst retaining fluidity – asserted lack of inventive step – asserted lack of novelty – asserted secret use – standard of proof – grounds of each appeal not made out – appeals dismissed

Legislation:

Patents Act 1990 (Cth) ss 7(1), 7(2), 7(3), 9(a) to (d), 18(1)(d), 60, 105(1A)

Cases cited:

Aktiebolaget Hässle v Alphapharm Pty Ltd (2002) 212 CLR 411

Apotex Pty Ltd v Sanofi-Aventis (2008) 78 IPR 485

Artcraft Urban Group Pty Ltd v Streetworx Pty Ltd (2016) 245 FCR 485

Aspirating IP Ltd v Vision Systems Ltd (2010) 88 IPR 52

AstraZeneca AB v Apotex Pty Ltd (2014) 226 FCR 324

AstraZeneca AB v Apotex Pty Ltd (2015) 257 CLR 356

Austal Ships Pty Ltd v Stena Rederi Aktiebolag (2005) 66 IPR 420

Australian Medic-Care Company Ltd v Hamilton Pharmaceutical Pty Ltd (2009) 261 ALR 501

Azuko Pty Ltd v Old Digger Pty Ltd (2001) 52 IPR 75

BlueScope Steel Limited v Dongkuk Steel Mill Co (2017) 135 IPR 1; [2017] FCA 1537

Bradken Resources Pty Ltd v Lynx Engineering Consultants Pty Ltd (2012) 210 FCR 21

Bristol-Myers Squibb Company v Apotex Pty Ltd (No 5) (2013) 104 IPR 23

British Acoustic Films Ld v Nettlefold Productions (1936) 53 RPC 221

Commissioner of Patents v Emperor Sports Pty Ltd (2006) 149 FCR 386

Coretell Pty Ltd v Australia Mud Company Pty Ltd (2017) 250 FCR 155

Delnorth Pty Ltd v Commissioner of Patents (2013) 100 IPR 175

Delnorth Pty Ltd v Dura-Post (Aust) Pty Ltd (2008) 78 IPR 463

Encompass Corporation Pty Ltd v InfoTrack Pty Ltd (2018) 130 IPR 387

Fieldturf Tarkett Inc v Tigerturf International Limited (2014) 317 ALR 153

Flexible Steel Lacing Co v Beltreco Ltd (2000) 49 IPR 331

Garford Pty Ltd v DYWIDAG Systems International Pty Ltd (2015) 110 IPR 30

General Tire & Rubber Co v The Firestone Tyre & Rubber Co Ltd (1971) 1A IPR 121

George C Warner Laboratories Pty Ltd v Chemspray Pty Ltd (1967) 1B IPR 649

Gilead Sciences Pty Ltd v Idenix Pharmaceuticals LLC (2016) 117 IPR 252; [2016] FCA 169

Graham Hart (1971) Pty Ltd v SW Hart & Co Pty Ltd (1978) 141 CLR 305

Grove Hill Pty Ltd v Great Western Corporation Pty Ltd (2002) 55 IPR 257

Lundbeck A/S v Alphapharm Pty Ltd (2009) 177 FCR 151

Hill v Evans (1862) 1A IPR 1

ICI Chemicals & Polymers Ltd v Lubrizol Corp Inc (1999) 45 IPR 577

Idenix Pharmaceuticals LLC v Gilead Sciences Pty Ltd (2017) 134 IPR 1

In the matter of Klabers Patent (1906) 23 RPC 461

Insta Image Pty Ltd v KD Kanopy Australasia Pty Ltd (2008) 239 FCR 117

JMVB Enterprises Pty Ltd v Camoflag Pty Ltd (2006) 154 FCR 348

KD Kanopy Australasia Pty Ltd v Insta Image Pty Ltd (2007) 71 IPR 615

Lockwood Security Products Pty Ltd v Doric Products Pty Ltd (No 2) (2007) 235 CLR 173

Longbottom v Shaw (1891) 8 RPC 333

Kirin-Amgen Inc v Hoechst Marion Roussel Ltd (2004) 64 IPR 444; [2004] UKHL 46

Martin Engineering Co v Trison Holdings Pty Ltd (1989) 14 IPR 330

Meat & Livestock Australia Limited v Cargill, Inc (2018) 354 ALR 95; (2018) 129 IPR 278; [2018] FCA 51

Meyers Taylor Pty Ltd v Vicarr Industries Ltd (1977) 137 CLR 228

Minnesota Mining & Manufacturing Co v Tyco Electronics Pty Ltd (2002) 56 IPR 248

Minnesota Mining and Manufacturing Co v Beiersdorf (Australia) Ltd (1980) 144 CLR 253

Neurizon Pty Ltd v Jupiters Ltd (2004) 62 IPR 569

Nicaro Holdings Pty Ltd v Martin Engineering Co (1990) 91 ALR 513

Olin Mathieson Chemical Corporation v Biorex Laboratories Ltd [1970] RPC 157

Palmer v Dunlop Perdriau Rubber Co Ltd (1937) 59 CLR 30

Product Management Group Pty Ltd v Blue Gentian LLC (2015) 240 FCR 85

Ramset Fasteners (Aust) Pty Ltd v Advanced Building Systems Pty Ltd (1999) 164 ALR 239

Sachtler GmbH and Co KG (formerly Sachtler AG) v RE Miller Pty Ltd (2005) 221 ALR 373; [2005] FCA 788

SNF (Australia) Pty Ltd v Ciba Speciality Chemicals Water Treatments Ltd (2011) 92 IPR 46

SNF (Australia) Pty Ltd v Ciba Specialty Chemicals Water Treatments Limited (2015) 114 IPR 231

SNF (Australia) Pty Ltd v Ciba Specialty Chemicals Water Treatments Limited [2016] APO 8

SNF (Australia) Pty Ltd v Ciba Specialty Chemicals Water Treatments Limited [2016] APO 72

SNF (Australia) Pty Ltd v Ciba Speciality Chemicals Water Treatments Ltd (2012) 204 FCR 325

SNF (Australia) Pty Ltd v Ciba Speciality Chemicals Water Treatments Ltd [2016] FCAFC 88

SNF Inc. v Ciba Specialty Chemicals Water Treatments Limited (2015) 133 CPR (4th) 259; [2015] FC 997

The Wellcome Foundation Ltd v VR Laboratories (Aust) Pty Ltd (1981) 148 CLR 262

Welcome Real-Time SA v Catuity Inc (2001) 113 FCR 110

Date of hearing:

31 July 2017, 1 to 4, 7 to 11 and 14 to 18 August 2017, 7 to 10 May 2018, 1 and 2 November 2018

Registry:

Victoria

Division:

General Division

National Practice Area:

Intellectual Property

Sub-area:

Patents and associated Statutes

Category:

Catchwords

Number of paragraphs:

1784

Counsel for the Appellant/Cross-Respondent:

Mr B Caine QC and Mr B Fitzpatrick

Solicitor for the Appellant/Cross-Respondent:

K & L Gates

Counsel for the Respondent/Cross-Appellant:

Mr D Shavin QC, Mr P Creighton-Selvay and Ms R White

Solicitor for the Respondent/Cross-Appellant:

Gilbert + Tobin

ORDERS

VID 211 of 2016

BETWEEN:

SNF (AUSTRALIA) PTY LIMITED

Appellant

AND:

BASF AUSTRALIA LTD

Respondent

AND BETWEEN:

BASF AUSTRALIA LTD

Cross-Appellant

AND:

SNF (AUSTRALIA) PTY LIMITED

Cross-Respondent

JUDGE:

BEACH J

DATE OF ORDER:

27 March 2019

THE COURT ORDERS THAT:

1.    The appellant’s appeal be dismissed.

2.    Australian patent application no. AU 2004203785 (as amended) proceed to grant.

3.    The appellant pay the respondent’s costs of and incidental to the appeal.

4.    There be a stay on the operation of orders 2 and 3 for 21 days subject to further order.

5.    The respondent’s cross-appeal be dismissed with no order as to costs.

6.    Any previous orders made under ss 37AF and 37AG of the Federal Court of Australia Act 1976 (Cth) be varied, if necessary, so as to permit publication of the Court’s reasons.

7.    Liberty to apply.

Note:    Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

ORDERS

VID 1318 of 2016

BETWEEN:

SNF (AUSTRALIA) PTY LIMITED

Appellant

AND:

BASF AUSTRALIA LTD

Respondent

AND BETWEEN:

BASF AUSTRALIA LTD

Cross-Appellant

AND:

SNF (AUSTRALIA) PTY LIMITED

Cross-Respondent

JUDGE:

BEACH J

DATE OF ORDER:

27 March 2019

THE COURT ORDERS THAT:

1.    The appellant’s appeal be dismissed.

2.    Australian patent application no. AU 2013204568 (as amended) proceed to grant.

3.    The appellant pay the respondent’s costs of and incidental to the appeal.

4.    There be a stay on the operation of orders 2 and 3 for 21 days subject to further order.

5.    The respondent’s cross-appeal be dismissed with no order as to costs.

6.    Any previous orders made under ss 37AF and 37AG of the Federal Court of Australia Act 1976 (Cth) be varied, if necessary, so as to permit publication of the Court’s reasons.

7.    Liberty to apply.

Note:    Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

REASONS FOR JUDGMENT

BEACH J:

1    The appellant (SNF) has appealed two decisions of a delegate of the Commissioner of Patents made in respect of Australian patent applications no. AU 2004203785 (the 785 application) and no. AU 2013204568 (the 568 application); I will refer to both as the opposed applications. The appeals have been brought under60(4) of the Patents Act 1990 (Cth) (the Act), with each appeal by way of a rehearing de novo.

2    Each of the opposed applications now stands in the name of BASF Australia Limited (BASF), having previously been assigned by Ciba Specialty Chemicals Water Treatments Limited (Ciba) to BASF.

3    Each of the opposed applications seeks the grant of a standard patent for an invention(s) entitled “Treatment of aqueous suspensions”. The opposed applications are derived from an international patent application filed on 7 January 2004 no. PCT/EP2004/000042; this application was published on 22 July 2004 as international publication no. WO/2004/060819 (the PCT application). The opposed applications each claim priority from a UK patent GB 0310419.7 filed on 7 May 2003 (the priority date).

4    The decisions which are the subject of the appeals before me are:

(a)    a decision of the delegate Dr Steven Barker made on 16 February 2016 (the 785 decision); see SNF (Australia) Pty Ltd v Ciba Specialty Chemicals Water Treatments Limited [2016] APO 8; and

(b)    a decision of that delegate made on 18 October 2016 (the 568 decision); see SNF (Australia) Pty Ltd v Ciba Specialty Chemicals Water Treatments Limited [2016] APO 72.

5    The opposed applications are governed by the provisions of the Act as they stood before the “Raising the Bar” amendments (Intellectual Property Laws Amendment (Raising the Bar) Act 2012 (Cth)). Accordingly, the pre-amendment versions of provisions such as s 7 concerning novelty and inventive step continue to apply.

6    In the 785 decision, the delegate:

(a)    decided that the invention defined in each of claims (as accepted) 1 to 11, 14, 16, 17, 19, 21 to 27, 29 and 30 of the 785 application lacked an inventive step solely in light of international publication no. WO 01/92167 (the Gallagher patent) filed by Ciba and the inventors for which were inter-alia Mr Michael Gallagher and Mr Stephen Adkins, and were therefore invalid;

(b)    decided that the invention defined in each of claims (as accepted) 12, 13, 15, 18, 20 and 28 of the 785 application had not been shown to lack an inventive step and were not invalid; and

(c)    allowed the patent applicant a period of 60 days within which to propose amendments to the 785 application, indicating that the deficiency in relation to claims 1 to 11, 14, 16, 17, 19, 21 to 27, 29 and 30 could be overcome by amendment to include a co-disposal integer.

7    Pursuant to s 105(1A) of the Act and under orders made by me on 21 July 2016, the claims of the 785 application were so amended. The effect of the amendment was to limit the scope of all of the amended claims to a co-disposal process by including the requirement that the “process comprises co-disposal of coarse and fine solids as a homogenous mixture”. That amendment was sufficient to address the delegate’s concerns.

8    In the 568 decision, the delegate:

(a)    decided that the invention defined in claims 2 and 3 (as accepted) and the claims appended to claims 2 and 3 lacked an inventive step in light of the Gallagher patent;

(b)    decided that the invention defined in claim 1 (as accepted) had not been shown to lack an inventive step; and

(c)    allowed the patent applicant a period of 60 days within which to propose amendments to the 568 application to overcome the deficiency in relation to claims 2 and 3 concerning the Gallagher patent.

9    Again, pursuant to s 105(1A) of the Act and under the orders made by me on 21 December 2016, the claims of the 568 application were amended. The effect of the amendment was to limit the scope of all of the amended claims to a co-disposal process by making a similar amendment as for the 785 application.

10    As noted, the opposed applications are derived from the PCT application filed on 7 January 2004. Ciba had previously also divided out five innovation patents from the PCT application, which were the subject of proceedings before Kenny J discussed below. It had also divided out various standard patent applications from the PCT application. Two of those applications are the opposed applications which are the subject of the present appeals. Another application (AU 2012216282) has been the subject of an opposition proceeding in which a decision was handed down by the delegate on 18 April 2018. SNF has filed a notice of appeal in respect of that decision which appeal I am also presently case managing. There have also been other applications for standard patents derived from the PCT application that have been withdrawn, lapsed or refused; there is yet a further application awaiting examination. For present purposes I need say nothing more about them.

11    As I have said, the parties were involved in Federal Court proceedings which concerned, inter-alia, the validity of five innovation patents which were filed as divisional applications of the 785 application.

12    In 2008, SNF commenced revocation proceedings in Federal Court proceedings number VID 447 of 2008 (the 2008 proceedings) concerning the validity of innovation patents nos. 2006100744, 2006100944, 2007100377, 2007100834 and 2008100396 (the innovation patents). Ciba cross-claimed for infringement.

13    The innovation patents were held to be valid by Kenny J and Ciba succeeded in its cross-claim for infringement (SNF (Australia) Pty Ltd v Ciba Speciality Chemicals Water Treatments Ltd (2011) 92 IPR 46).

14    SNF appealed that decision to the Full Federal Court, where a majority dismissed the appeal (SNF (Australia) Pty Ltd v Ciba Speciality Chemicals Water Treatments Ltd (2012) 204 FCR 325). SNF then sought special leave to appeal, which leave was refused.

15    Subsequently, in April 2014 SNF filed an interlocutory application seeking to re-open the 2008 proceedings and her Honour’s judgment therein. That application was dismissed (SNF (Australia) Pty Ltd v Ciba Specialty Chemicals Water Treatments Limited (2015) 114 IPR 231). SNF then sought leave to appeal that decision, but the Full Federal Court refused such leave (SNF (Australia) Pty Ltd v Ciba Speciality Chemicals Water Treatments Ltd [2016] FCAFC 88).

16    Further, in November 2011 a related entity of SNF, SNF Inc., commenced proceedings against Ciba in the Canadian Federal Court seeking a declaration that a patent related to the opposed applications (the Canadian patent), which claimed essentially the same process as the opposed applications, was invalid. The Canadian patent was also derived from the PCT application. On 24 August 2015, Phelan J determined that the Canadian patent was invalid (SNF Inc. v Ciba Specialty Chemicals Water Treatments Limited (2015) 133 CPR (4th) 259; [2015] FC 997).

17    I would note at this point that many of the witnesses who gave evidence before me on these appeals have previously given evidence in these other proceedings. Moreover, some aspects of Kenny J’s reasoning discusses issues that I am also concerned with. Nevertheless, I must deal with the issues on the evidence before me. Further, some of Kenny J’s analysis must be understood in light of the fact that she was dealing with innovation patents whereas I am concerned with applications for standard patents.

18    In general, SNF contends before me that:

(a)    the invention so far as claimed in any claim of the opposed applications does not involve an inventive step;

(b)    the invention so far as claimed in any claim of the opposed applications was secretly used in the patent area before the priority date of that claim by or on behalf of or with the authority of the patent applicant or nominated person or the patent applicant’s or nominated person’s predecessor in title to the invention; and

(c)    all of the claims of the opposed applications are not novel except for:

(i)    claims 5, 18, 21 and 23 for the 785 application; and

(ii)    claims 15, 27 and 30 of the 568 application.

19    BASF has cross-appealed those parts of the 785 decision and 568 decision in which the delegate held that certain claims of the 785 application being claims 1 to 11, 14, 16, 17, 19, 21 to 27, 29 and 30 and certain claims of the 568 application being claims 2 and 3 and the appended claims lacked an inventive step in light of the Gallagher patent. That was the only ground upon which the delegate held any claims of the opposed applications to be invalid. But notwithstanding its cross-appeal, BASF has amended the claims of the 785 application and the 568 application to overcome the findings of the delegate by narrowing the relevant claims of the opposed applications to include a “co-disposal” integer. In the 785 decision the delegate concluded that it has not been shown that the claims relating to co-disposal lack inventive step” (at [325]). In the 568 decision the delegate held that claim 1 of the 568 application features the “significant new step” of co-disposal (at [21]), held that the Gallagher patent suggested “a preference for disposal of tailings that are substantially composed of fine particles” (at [39]), and said that “I am satisfied that the evidence shows that it was known to combine fine and coarse particles. However, the evidence does not show that it would have been a matter of routine to do this in the context of the problem” (at [56]). Given the amendments ordered by me and my decision on SNF’s appeals, it is not necessary to say anything further on BASF’s cross appeals.

20    BASF also contends by its notices of contention that the delegate erred in finding that the person skilled in the art could as at the priority date have been reasonably expected to have ascertained the relevant prior art relied on by SNF. I will deal with this later when I discuss the s 7(3) question.

21    Now SNF bears the onus in relation to establishing each ground of opposition and corresponding ground of appeal before me. The authorities establish that for the opposition to be upheld it must be clear that the patent, if granted, would not be valid (Meat & Livestock Australia Limited v Cargill, Inc (2018) 354 ALR 95; (2018) 129 IPR 278; [2018] FCA 51 at [11] (MLA (No 1))).

22    Further, in MLA (No 1) at [11], I agreed with the views expressed by Bennett J in Austal Ships Pty Ltd v Stena Rederi Aktiebolag (2005) 66 IPR 420 (Austal Ships), that where there are conflicting sets of expert opinions on a principal question such as lack of inventive step, unless one set of views can be rejected on proper grounds, the opponent will not have discharged its legal burden to the requisite degree. But the higher standard of satisfaction does not apply to findings of underlying primary facts. As Bennett J explained in Austal Ships at [12]:

I can accept that a lower standard may apply to proof of evidence such as whether a document has been published or, indeed, whether a prior art vessel was well-known. I do not accept that it properly applies to the factual question that itself is the test for obviousness or lack of inventive step. Where the factual question is itself the legal test, as set out in s 7(3) of the Act, it seems to me that it should be determined at the higher standard. That means that where there are two opposing expert views that are conclusive on obviousness, both presented bona fide by witnesses of accepted expertise, unless one set of views can be rejected on proper grounds, the legal burden to establish a ground of opposition is not discharged; the court cannot be practically certain that obviousness or lack of inventive step is established.

23    Further, in Aspirating IP Ltd v Vision Systems Ltd (2010) 88 IPR 52, Besanko J cited the above passage in Austal Ships and said at [35] that:

The primary facts are to be established on the balance of probabilities, but the ultimate facts the facts leading directly to a conclusion of a lack of novelty or a conclusion of obviousness must be proved to the level of practical certainty.

24    Now there has been limited judicial consideration of particular matters which must be proved to a level of being “clear” as opposed to those that can be established on the balance of probabilities. But decisions of various delegates of the Commissioner of Patents provide some guidance on where the division between the two standards has been drawn in practice.

25    For example, delegates have found that the lower standard of proof is applicable as to whether a matter formed part of the common general knowledge, whether a document has been published and whether in the context of secret use an invalidating sale occurred before the priority date. Contrastingly, delegates have found that the higher standard of proof is applicable as to whether a person skilled in the art would achieve the same results as the invention so far as claimed using slightly different techniques and altered conditions, whether the claims of the invention so far as claimed lack utility and whether the claims of the invention lack novelty.

26    In my view, the following matters are required to be established by SNF at the lower standard of on the balance of probabilities:

(a)    the materials routinely referred to by a person skilled in the art;

(b)    whether various matters formed part of the common general knowledge at the priority date and what about them and their use was common general knowledge at the priority date including, in the present context, the use of co-disposal processes, the use of belt presses and centrifuges, the use of processes involving the addition of flocculant to thickener underflow in the outlet pipe from the thickener (secondary dosing), the use of flocculants in treating tailings, and the variables routinely adjusted in their use, whether a treatment process which worked effectively on one mineral type would also work effectively on other mineral types, and the use of processes involving the deposition of thickener underflow onto a slope, wall or floor in order to allow tailings to become beached and water to flow out to a lower point for re-use (tailings beaching); I note that it is not now asserted that the use of secondary dosing in tailings beaching (SDITB) was part of common general knowledge at the priority date;

(c)    whether relevant prior art being the Backer & Busch papers, the Condolios patent, the Gallagher patent and the Pearson patent (I will precisely identify these later) would have been ascertained, understood and regarded as relevant by a person skilled in the art at the priority date, although there is an argument for saying that the higher standard should apply;

(d)    what is disclosed in the Backer & Busch papers, the Condolios patent, the Gallagher patent and the Pearson patent, although there is an argument for saying that the higher standard should apply;

(e)    what use was made of the claimed invention by Ciba at Yarraman, Sandalwood and Ardlethan (I will identify these locations later) before the priority date and what was the nature of that use; and

(f)    whether the Cable Sands documents (I will describe these later) would have been treated by a person skilled in the art as a single source of information, although there is an argument for saying that the higher standard should apply.

27    On questions of construction, I am also prepared to apply and have applied the lower standard albeit that there is some force in BASF’s submissions that the higher standard should apply.

28    But in my view the following matters are required to be established to the higher standard:

(a)    whether any of the claims of the opposed applications were obvious in light of common general knowledge alone;

(b)    whether any of the claims of the opposed applications were obvious in light of common general knowledge combined with any of the prior art relied on by SNF;

(c)    whether the claimed invention was secretly used by Ciba prior to the priority date; and

(d)    whether any of the claims of the opposed applications lacked novelty in light of the disclosures in the Cable Sands documents.

29    In summary and for the reasons that follow, I have determined to dismiss each of SNF’s appeals.

30    For convenience, I have divided my reasons into the following sections:

(a)    Technical background ([32] to [141]);

(b)    The opposed applications ([142] to [220]);

(c)    The witnesses ([221] to [261]);

(d)    The person skilled in the art ([262] to [279]);

(e)    Construction of the claims ([280] to [372]);

(f)    Common general knowledge ([373] to [501]);

(g)    Obviousness – legal principles ([502] to [536]);

(h)    Obviousness in light of common general knowledge alone ([537] to [1080]);

(i)    Common general knowledge plus s 7(3) documents ([1081] to [1410]);

(j)    Secret use ([1411] to [1720]);

(k)    Lack of novelty ([1721] to [1782]); and

(l)    Conclusion ([1783] to [1784]).

31    In these reasons, in terms of the respondent to the appeals I will generally refer to BASF, save for when I need to refer to Ciba directly in discussing contemporaneous conduct and dealings before the priority date.

TECHNICAL BACKGROUND

32    For the most part, the following matters that I have described in this section were common general knowledge as at the priority date and do not appear to be in dispute. It is useful to set these out in order to appreciate the later discussion of the relevant technical issues. In part, I have drawn upon the helpful discussion of some of these matters given by Dr Ross de Kretser, an expert witness called by SNF. Before getting into the detail, let me set out a mini glossary.

33    Terms or expressions used in the relevant field include the following:

(a)    beach angle” refers to the slope or angle of a deposit (relative to underlying material) which forms a stack or heaped geometry;

(b)    beneficiation” is a process of concentrating the value(s) in an ore by separation from waste material;

(c)    bimodal distribution of particle sizes means there are two distinct sizes of particles within the solid material. Often these are referred to as “fines” and “coarse” fractions;

(d)    co-disposal” means, in simple terms, the disposal of a combination of both coarse and fine tailings particles. In some cases this will involve a clearly bi-modal size distribution which has both coarse and fine particles;

(e)    flocculants” are high molecular weight polymers that induce aggregation of solid particles. This occurs by individual flocculant molecules co-adsorbing on two or more solid particles and binding them together. Due to the increase in their size, the bound solids (often called “aggregates”, “floccules” or “flocs”) have a faster settling rate, but a much lower density, than the individual solids particles;

(f)    intrinsic viscosity” is a measure of the capability of a polymer in solution to enhance the viscosity of the solution. It is related to the volume per unit mass occupied by flocculant molecules in solution and is measured in the unit “dL/g”, which is decilitres per gram;

(g)    rheology” is the study of deformation and flow of fluid matter;

(h)    spirals” are equipment used in mineral processing industry to beneficiate an ore by means of the forces imposed on particles as they flow under gravity in a spiralling motion;

(i)    a “stack” or a “heaped geometry” is a deposit of material which is characterised by being larger at the bottom than at the top;

(j)    tailings” are the end product of a mineral processing operation. Tailings are what remain of an ore after the value(s) (e.g. alumina, coal, copper, diamond, gold, nickel, uranium, zinc) have been extracted during the processing operation; and

(k)    yield stress” means, in simple terms, a measure of the minimum force required to be applied to an object to make it deform (i.e. begin to behave as a fluid).

34    Mineral processing is the processing of mined materials to separate the valuable component of the ore such as alumina, coal, copper, diamond, gold, nickel, uranium, zinc from the waste material i.e. the tailings. Such processing usually produces tailings that have the following characteristics:

(a)    First, they are in slurry form, comprising process water and particulate matter.

(b)    Second, the particulate matter is comprised of large amounts of quartz/silica, various types of clays, and other minerals depending on the host rock for the ore.

(c)    Third, the particle size of the particulate matter depends on the level of grinding required for mineral extraction which in many cases is related to the type of mineral and nature of the deposit.

(d)    Fourth, in many operations, multiple tailings streams are generated with different characteristic particle size distributions. For example, in mineral sands processing, slimes (fine) and sand (coarse) streams are generated from different parts of the processing circuit. Sand streams are also generated in alumina refining in addition to the fine red mud tailings. In coal processing a coarse reject stream is commonly generated in addition to the fine tailings.

35    Tailings slurries vary in their consistency from having a low solids concentration, and accordingly, being very liquid, to having a higher solids concentration and being more viscous. The solids in the tailings may consist of fine particles, coarse particles or a mixture of both. The fine particles are often referred to as “clay” or “fines”, and the coarse particles are often referred to as “sand”.

36    The particle size of the particulate matter strongly depends on the level of grinding required for mineral extraction which in many cases is related to the type of mineral and nature of the deposit. For example, typical tailings particle sizes can be as follows:

(a)    lead zinc: top size 80 to 500 microns, 80% passing 30 to 150 microns;

(b)    gold-silver-platinum: top size 80 to 500 microns, 80% passing 30 to 150 microns;

(c)    coal: top size 100 to 1500 microns; 80% passing 10 to 900 microns;

(d)    bauxite (red mud): top size 100 microns, 80% passing 5 to 10 microns;

(e)    mineral sands – slimes: top size 75 microns, 80% passing 10 to 50 microns; and

(f)    mineral sands – sand: top size 1000 microns.

37    Mineral tailings typically have a range of particle sizes, generally within the range of 0.5-400 microns, depending on the ore mineralogy and processing regime.

38    Coal, mineral sands, copper, gold, lead, zinc, red mud, silver, uranium and platinum tailings often comprise a mixture of coarse and fine particles. Phosphate and uranium tailings usually only comprise fine particles.

39    Tailings with a bimodal distribution of particle sizes were common at the priority date. The bimodal size distribution of tailings could arise due to either the nature of the treatment process and the mineral being mined, or by deliberate addition of coarser material to a finer tailings stream. Tailings with a bimodal distribution of coarse and fine particle sizes is typical in the mineral sands industry. In contrast, in other industries such as coal tailings, the size distribution is more gradual across a scale of sizes between coarse and fine particles.

40    Let me now address the question of thickeners. A thickener is essentially a large tank which is continuously fed with tailings slurry, typically through a central feeding arrangement called a feedwell, and which has sufficient volume to allow solid particles to settle to the bottom of the thickener and produce a clear overflow of suspending fluid. That overflow is typically recycled to the plant for re-use, or in some cases may contain the valuable product which is processed elsewhere in the plant.

41    The settled tailings or what is referred to as “thickener underflow” are then pumped at relatively low solids concentrations to the deposition area in pipelines or less commonly in open launders or channels.

42    A schematic of a thickener, illustrating key features is set out below:

43    As at the priority date, two of the types of thickeners being used were conventional thickeners and paste thickeners. Conventional and paste thickeners are similar pieces of equipment that are designed and operated to produce underflows of different consistencies. The aim of paste thickeners is to increase water recovery and produce a denser underflow compared to conventional thickeners. Paste thickeners are typically used in relation to alumina tailings which contain caustic soda. However, the underflow from both thickeners is generally transported to a deposition area via a tailings pipeline or launder.

44    Thickener underflow is typically disposed of by pumping the underflow through a conduit to a disposal area where the material is allowed to stand, with a view to land reclamation and water recycling. Typically thickener underflow has a solids concentration of between 15% to 80% by weight, being the percentage of solids in the tailings, the rest being water.

45    Let me now turn to discuss flocculants. Water soluble flocculants are used to increase the yield stress of tailings thereby thickening the tailings in order to improve the dewatering and disposal of tailings, irrespective of the mineral type.

46    Flocculants cause tailings to aggregate into flocs which on deposition under certain conditions start to stick together to form a structure that is permeable and allows for further dewatering. With dewatering, the yield stress of the material increases. The addition of the flocculants co-immobilizes the coarse and fine particles in the tailings on deposition. Typically, the higher the flocculant dose, the greater the flocculation of the tailings.

47    Commercially available flocculants including polyacrylamide which are commonly used in the treatment of tailings are either positively charged (cationic), negatively charged (anionic) or uncharged (non-ionic). The choice of charge of the flocculant is governed by the qualities of the tailings to be treated.

48    Flocculant manufacturers make flocculants with different molecular weights which react differently depending on the characteristics of the tailings. The higher the molecular weight, the longer the flocculant chain. The molecular weight of a flocculant is related to its intrinsic viscosity. The higher the molecular weight of the flocculant, the higher its intrinsic viscosity. Almost all commercially available flocculants have an intrinsic viscosity of greater than 5dl/G.

49    Flocculants with a high molecular weight are more effective for “bridging” or “connecting” the particles in the tailings as part of the flocculation process. Flocculants with a higher molecular weight have longer and more cross-linked arms which make them more effective at connecting with more particles in the tailings.

50    Flocculants formed from ethylenically unsaturated water-soluble monomer or a blend of monomers were well known as at the priority date.

51    Commercially available flocculants as at the priority date could be formed from:

(a)    monomer(s) selected from the group consisting of (meth)acrylic acid, allyl sulphonic acid and 2-acrylamido-2-methyl propane sulphonic acid as the free acids or salts thereof, optionally in combination with non-ionic co-monomers, selected from the group consisting of (meth)acrylamide, hydroxy alkyl esters of (meth)acrylic acid and N-vinyl pyrrolidone;

(b)    monomer(s) selected from the group consisting of (meth) acrylamide, hydroxy alkyl esters of (meth) acrylic acid and N-vinyl pyrrolidone; and

(c)    monomer(s) selected from the group consisting of dimethyl amino ethyl (meth) acrylate methyl chloride, (DMAEA.MeCI) quat, diallyl dimethyl ammonium chloride (DADMAC), trimethyl amino propyl (meth) acrylamide chloride (ATPAC) optionally in combination with non-ionic co-monomers, selected from the group consisting of (meth) acrylamide, hydroxy alkyl esters of (meth) acrylic acid and N-vinyl pyrrolidone.

52    Dr de Kretser gave evidence that flocculants added to tailings in aqueous solution achieve the peak level of flocculation within 10 to 20 seconds after addition to tailings, or in even shorter time periods. After 1.5 minutes after addition, depending on the level of agitation, the level of flocculation (yield stress) would be significantly lower than that peak level. Other evidence was to the effect that an aqueous solution will typically take a few seconds to mix and flocculate the slurry.

53    As I have said, a flocculant (or polymer) is a substance, most commonly synthetic, which is added to slurries to produce flocculation. Flocculation is the aggregation of suspended particles into larger clumps called floccs which are then able to more readily separate from the suspension or slurry. Let me at this point distinguish between a coagulant and a flocculant.

54    Coagulation facilitates the destabilisation and then elimination of colloidal particles, which are insoluble particles suspended in water. In other words, it facilitates the aggregation of insoluble particles suspended in a solution, where they would otherwise be evenly distributed. Coagulants do so by neutralising or “screening” the like charges of suspended particles, preventing repulsion between the particles, and facilitating either attractive forces (for example van der Waals forces) or bonding between the particles. Coagulation can form bonded flocs of particles, however, these flocs are less dense and less extensive than flocs created by flocculation.

55    Flocculation tends to be used where colloidal particles are already destabilised (i.e. the particles are not evenly dispersed in a solution, whether by coagulation or otherwise). Flocculants assemble the destabilised colloidal particles into aggregates. Flocculants, being long polymer chains, fix the destabilised particles and aggregates along the polymer chain, generally via ionic or hydrogen bonding. Properties of the flocculant used will depend on the nature of the solution. Anionic (negatively charged) flocculants “donate” negative charges to the suspended particles and cationic (positively charged) flocculants “donate” positive charges to the suspended particles.

56    The following discussion with Dr de Krester is also relevant:

MR SHAVIN: His Honour asked a question of me when we were opening the case as to the difference between a coagulant and a flocculant. Could you perhaps explain to his Honour what the difference is?---Typically a coagulant is a reagent which is – has the purpose of creating small flocs. So it would – I mean, probably the best thing is to consider it in contrast to a flocculant, which is a large, long chain polymer, which has the objective of grabbing particles and pulling them together into a more open floc. A coagulant is typically creates flocs, but they’re denser, less extensive. So there’s – so coagulation versus flocculation is typically characterised by the extent and size of the particle or the floc that you form, but also the types of reagents differ in the ones that are used to coagulate versus flocculate. So coagulation could be via changing the charge between particles, so they just – so simply the particles attract each other.

HIS HONOUR: So if you’ve got particles that have the same charge and would otherwise repel if you added a mineral salt or something - - -?---That’s – that’s correct.

- - - you would neutralise the charge, which will allow them to aggregate and create the small flocs?---Yes. And so some cases coagulation can be induced just by the concentration of sale in a suspending fluid.

Yes?---And what that actually does – you can still have like charge particles, but the repulsion between them is screened out because you have a lot of other ions.

Yes?---And they get sufficiently close that there’s – I’m not sure if you’re familiar with Van der Waals forces.

Yes, I am. Yes?---So that attractive force can actually – they can approach close enough that they – they reach what – what’s called an attractive minimum. So that’s one method of coagulation. There are also more, I guess, physical coagulants, in the sense that a reagent is added which precipitates and actually sticks, essentially, particles together. And there’s also the case of organic coagulants, which are polymers, but they’re smaller, shorter chain polymers, so they typically act maybe of around 100,000 molecular weight, compared to bridging polymers, which might be upwards of a million molecular weight.

So for a coagulant in some cases it can bond with the particles, but with all flocculants there is that bonding with the interaction of the polymer?---Yes. Yes. With flocculants there’s typically a physical absorption component to it.

Yes?---Coagulants can - - -

May be?---Can be electrostatic or physical.

Yes?---And typically coagulants – or coagulants are commonly used in conjunction with flocculants, but they can also be used in isolation as well.

57    Let me return to flocculants and elaborate further. As I have said, flocculants are composed of long chains of monomers that have been chemically bonded. When activated, flocculants have an elongated chain-like structure containing sites which can bind to solid particles when added to a suspension of particles in fluid. When a flocculant in aqueous solution mixes with a suspension, the flocculant chains adsorb or bind to the particles, aggregating them into flocs. Due to their increase in effective size, the flocs settle faster under gravity than the primary particles alone, increasing dewatering rates.

58    Flocculants have a wide range of chemistries and characteristics and the choice of the particular chemistry and characteristic of the flocculant is made on the basis of achieving the desired aggregation of particles in the system in question.

59    Conventional flocculants available before the priority date were typically a combination of polyacrylamide and polyacrylate monomers in a range of molecular weights. As I have said, molecular weight is determined by the length of the flocculant chain and affects the solids capture ability of the flocculant. Longer flocculant chains (having a higher molecular weight) have more active sites with which to trap solid particles. Higher molecular weight flocculants were also more viscous. Polyacrylamide was a well known form of flocculant that had a high molecular weight.

60    Further, as I have said, flocculants can be anionic, cationic or non-ionic, but the majority of flocculants used in mineral processes before the priority date were either anionic or non-ionic.

61    Flocculant was most commonly delivered to site in dry powder or emulsion form due to the lower transport cost and its more stable storage characteristics. Flocculants were rarely delivered to site in liquid (aqueous) form.

62    Irrespective of the physical form in which a flocculant was delivered, its use in practice in many cases required the flocculant to be converted into aqueous solution.

63    In dry powder and (to a lesser extent) emulsion form, the chains in a flocculant will be tightly coiled. In order for a flocculant to perform effectively, the flocculant chains must be dissolved in water and fully uncoiled so as to maximise the available binding sites. The preconditioning to achieve this state depended on the physical form of the flocculant. In particular the following may be observed:

(a)    Powder flocculant was added, sometimes with a wetting agent, to process water in a mixing device before storage in conditioning tanks to allow time for the powder to dissolve and for the flocculant chains to uncoil.

(b)    Emulsion flocculant was supplied as a water-in-oil emulsion i.e. droplets of concentrated flocculant solution in an oil carrier. Prior to use, the emulsion was inverted in a mixing device to release the concentrated flocculant solution into process water. The inverted flocculant solution was then transferred to conditioning tanks, again to allow time for the flocculant chains to uncoil.

(c)    Liquid (aqueous) flocculant could be supplied in a fully dissolved state such that it could be used directly from the source tanks on delivery. This avoided the need for a flocculant preparation plant, but increased transport cost per unit weight of flocculant, effectively limiting its economic viability to low tonnage / low dosage operations. The supplied flocculant could be more concentrated than required by the application to reduce transport costs, and could be diluted before addition as required.

64    In most cases, the activated flocculant was then added to the tailings from the conditioning tanks with a level of dilution dictated by the application. Typically, for thickening applications, this dilution was by a factor of 10.

65    Further, different forms of water soluble flocculant required different activation times in water. In particular, dry powder flocculant required the longest activation time, flocculant in an emulsion required a shorter activation time and flocculant added in aqueous solution required no activation time at all.

66    If flocculant was added to tailings in a fully activated solution, it would begin flocculating within a few seconds.

67    As I say, when the flocculant was supplied in dry powder form it was mixed and aged in a mixing tank prior to addition to the thickener or thickener underflow to allow the flocculant to unravel and convert to an aqueous solution.

68    And as I say, when the flocculant was supplied in emulsified form, the normal practice would be to invert or “flip” the emulsion and turn it into an aqueous solution prior to addition to the thickener or thickener underflow.

69    The expertise and equipment necessary to convert flocculant from either dry powder or emulsified form to aqueous solution were present at most mine sites. Flocculant was commonly added in aqueous solution to thickeners, belt presses, and centrifuges. After preparation of the flocculant into aqueous form, the prepared flocculant was generally diluted prior to addition to thickeners, belt presses, and centrifuges. Dilution was required to reduce flocculant viscosity and enhance dispersion or mixing of the flocculant solution into the tailings. This facilitated efficient flocculant-particle interaction.

70    Let me now say something about dose point. A key factor to consider in the operation of thickeners, belt press filters and centrifuges is the location at which the flocculant was added. It was important to take the dose point into account to enhance the performance of the equipment and therefore to improve the characteristics of the tailings when discharged in the deposition area.

71    It was necessary to determine the appropriate point at which there was sufficient time after the flocculant was added for it to fully mix and react with the tailings, but not so much time as to allow shear forces to break down the flocs that were formed prior to the point where the material reaches deposition.

72    As to dosage, the following may be noted. The dose used to treat a mineral suspension in a thickener would have a significant influence on the strength of the tailings material in that it would have a higher yield strength.

73    Generally speaking, provided there was adequate mixing, the greater the dose of flocculant the greater the flocculation.

74    The size of flocs formed after flocculant addition was affected by the dosage. It was known that if larger flocs were desired, with a concomitant increase in dewatering rate, a larger dosage of flocculant would typically be required.

75    The addition of flocculants to tailings slurries can significantly change their rheology and dewatering properties. This can result in a slurry that has a higher viscosity, higher yield stress, higher compressive strength and increased permeability due to an increase in the effective particle size caused by the aggregation of fine particles in the slurry. The resultant change in rheology and dewaterability was a function of the dose of the flocculant. Larger doses of flocculant could result in a larger improvement in permeability, but with a stiffer particle network (i.e. increased yield stress at a fixed solids concentration).

76    The appropriate flocculant dosage for a particular application varied significantly based on the following factors which were idiosyncratic to each mine site:

(a)    desired performance e.g. settling rate, settled density, shear sensitivity / robustness;

(b)    flocculant type and process water chemistry;

(c)    combination with other conditioning reagents e.g. inorganic coagulants, dewatering aids;

(d)    solids particle size distribution;

(e)    solids particle composition / mineralogy;

(f)    solids concentration at point of addition;

(g)    shear conditions at and after point of addition; and

(h)    number of different points of floc addition.

77    At this point it is useful to say something about a number of fundamental material properties which govern the flow and dewatering of tailings irrespective of the method of treatment or transport of the tailings, again drawing from some of the evidence given by Dr de Krester.

78    Let me begin with rheology. The rheology of the slurry refers to its viscosity and flow characteristics.

79    The viscosity is a measure of the consistency of a slurry which governs the amount of energy required to make the slurry flow from one point to another. A low viscosity material such as water will flow readily whereas a more viscous material such as mud (or honey) requires more energy to make it flow.

80    In the case of suspensions of particles in liquid, in addition to having elevated viscosities, the slurries can exhibit a yield stress. Yield stress describes a stress level which must be exceeded before flow can be initiated. For example, toothpaste will not flow unless a sufficient force is applied to it - that amount of force is the yield stress of the toothpaste. Yield stress is measured in pascals (Pa) (i.e. force, in newtons, per square metre). The presence of a yield stress is an indication that an interconnected network exists between the particles within a material. This interconnection could be via attractive forces sticking particles together in the slurry, or simply due to the interlocking of particles as the solids concentration increases to the point where they can no longer move freely relative to one another.

81    Let me now turn to dewatering. The primary physical properties that govern the dewatering of any slurry are its permeability and its compressive behaviour.

82    The permeability of a slurry is governed by the friction generated when solid and liquid particles move relative to each other within the slurry. At low solids concentration (i.e. settling), the solid particles can be considered as moving within a liquid continuum. At high solids concentrations, the liquid can be considered as moving within a solid continuum, such as would be the case in filtration or consolidation within a tailings storage facility (TSF). Therefore, permeability governs the rate at which liquid can be expressed or separated from the solid particles in a dewatering operation.

83    A schematic illustrating the range of solids concentrations and inter-particle structures over which the concept of permeability relates to is set out below:

84    The top section of this diagram shows flocculant acting on colloidal particles to form a floc, and ultimately a network of flocs within the material. Once the network has been formed, the application of a stress (for example by the weight of overburden generated by the addition of further material above) compresses the network causing the release of fluid (i.e. further dewatering of the deposited material). Permeability affects the rate of dewatering of the material across the entire spectrum of states depicted.

85    In simple terms, a highly permeable slurry will release water faster than a less permeable slurry; this is the case in thickeners, filters or on TSF deposition.

86    Let me elaborate on compressive behaviour. As the solids concentration of a slurry is increased, a point is reached where the solid particles within the slurry become interconnected. At this point, a solid particle network exists which has an integral strength which can withstand application of an external force (the yield stress). This condition could be generated either through dewatering of a slurry (e.g. sedimentation) or through the modification of inter-particle forces and structure within the slurry, or the addition of more solid particles to the slurry.

87    The compressive behaviour of a slurry refers to the ability of this network of particles to rearrange under the influence of an external force, thereby expressing liquid from between the particles. To facilitate compressive yielding of this network of particles at a particular solids concentration, an external force must be applied that exceeds the integral strength of the network (this is sometimes termed the “compressive yield stress”). At this point, the network will compress to a higher solids concentration with a higher integral strength. Ultimately, the compressive behaviour of a solid-liquid mixture governs the ultimate amount of liquid which can be expressed or removed from the mixture in a dewatering operation.

88    The diagram below illustrates the typical trend in network strength (compressive yield stress or shear yield stress) with increasing solids concentration. This shows that as the solids concentration increases, a point is reached where the strength of the material rapidly increases due to the interaction between the particles within the mixture.

89    There are various factors which significantly affect the rheology and dewatering characteristics of tailings slurry, both within dewatering equipment in a mineral processing plant (i.e. thickeners, filters, centrifuges) as well as within the TSF (i.e. water release upon deposition and long term consolidation after placement).

90    First, a higher solids concentration will usually result in a slurry with a higher viscosity, higher yield stress and a higher compressive strength. This is related to an increase in the number of collisions or interactions between particles under flow or in dewatering.

91    In terms of permeability, a higher solids concentration will result in a lower permeability. This is because the more solid particles in the slurry, the less available space for liquid to flow between those particles.

92    Viscosity, yield stress, compressive strength and permeability are all strong functions of slurry solids concentration.

93    Second, for the same material, a finer particle size distribution will generally have a higher viscosity, higher yield stress, higher compressive strength but lower permeability than a coarser particle size distribution. The lower permeability is due to the more tortuous flow path between or by the particles in the finer suspension. However, the finer particles increase the strength of the network (more particle interactions per unit volume) which in turn produces a higher yield stress.

94    In addition, the relative amounts of coarser and finer material within a size distribution can affect the packing of the particles in the slurry, and therefore affect the viscosity, yield stress, compressive strength and permeability of the slurry.

95    Third, the composition of dissolved ions within the process water influences the interaction between fine particles in the slurry and can result in attractive or repulsive forces between them. This can result in a slurry with a higher viscosity, yield stress and compressive strength due to a stronger inter-particle network (attractive forces). Conversely, a slurry with lower viscosity, no yield stress or compressive strength can be generated with repulsive forces between the particles. In terms of permeability, attractive forces typically promote fine particle aggregation, with a resultant increase in permeability at a given solids concentration.

96    Fourth, the type of minerals present within the tailings particles can have a strong influence on slurry rheology and dewatering. In particular, clay minerals were well known sources of increased viscosity in tailings slurries due to their plate-like structure and high surface area to particle volume ratio. Shear yield stress varies for a range of mineral tailings slurries with varying mineralogy, water chemistry, particle size and shape.

97    Fifth, the addition of chemical modifiers to tailings slurries can significantly change their rheology and dewatering properties, largely based on whether they promote attraction or repulsion between the particles.

98    This can either result in a slurry that has a higher viscosity, higher yield stress, higher compressive strength (with attraction), or is less viscous, has a lower yield stress and a lower compressive strength (in the case of repulsion). In terms of permeability, higher levels of attraction between particles typically results in increased permeability due to an increase in the effective particle size caused by the aggregation of fine particles in the slurry. This effect was the primary objective of the use of polymer flocculants in slurry dewatering (either within a thickener, filter, centrifuge or on deposition in a TSF). In the case of flocculant addition in particular, the resultant change in rheology and dewaterability was a strong function of the dose of polymer. It was known that larger doses of flocculant could result in a larger improvement in permeability, but with a stiffer particle network (i.e., increased yield stress at a fixed solids concentration).

99    The changes in both permeability and network strength with increasing levels of flocculation are illustrated in the graphs set out below. In these graphs, permeability and network strength behaviour for three states of flocculation are conceptually depicted. State A represents a material that is unflocculated. State B represents a material that is moderately flocculated. And State C represents a material that is heavily flocculated. As can be seen in the graph on the left, the more flocculated that a material is, the higher its shear or compressive strength at a fixed solids concentration. As can be seen in the graph on the right, the more flocculated the material is, the higher the permeability for a given solids concentration. Typically, a permeability graph as depicted on the right is logarithmic, such that the changes induced by flocculation as illustrated can result in improvements of factors of 10 (or greater) over unflocculated material.

100    Sixth, slurries can exhibit changes in their rheology and de-watering behaviour as a function of the level of shearing they experience. The presence of shear dependency in a slurry can be related to the chemistry, mineralogy, particle shape or presence of additives in the slurry.

101    In particular, for slurries subject to polymer flocculation, the flocculated network generated by the flocculant, is highly shear sensitive such that exposing the slurry to shear can degrade the flocculated network. This degradation is irreversible (without the addition of further flocculant) and would typically lead to reductions in viscosity, yield stress, compressive strength and a significant reduction in permeability.

102    Let me say something further about yield stress at this point. Flocculants increase the yield stress of a mineral slurry. Yield stress is a measure of the strength of the slurry. That is, the ability of the slurry to withstand the application of applied force such as shear.

103    The yield stress of the slurry is critical to both the pumping of the slurry to a deposition area (such as a tailings dam) and the behaviour of the slurry on deposition.

104    Typically, the greater the amount of flocculant used, the greater the yield stress of the slurry at a given concentration, as demonstrated in the diagram set out below. As the concentration of solids in the slurry is increased the slurry becomes less fluid-like and starts to become paste-like (like toothpaste or peanut butter) and ultimately, at high concentration, becomes solid-like (a cake).

105    Pumping a slurry with a higher yield stress will require more energy to pump than a slurry with a lower yield stress. However, as discussed below, pumping the slurry also exposes the slurry to shear forces, which causes flocs to break down and for the yield stress to decrease.

106    Let me now say something about pumpability and shear thinning. The distance between the thickener or treatment plant and the deposition area could vary from tens of metres to many kilometres depending on the particular mine. Therefore the time the thickened tailings were in transit from the thickener or treatment plant to the deposition area could vary from seconds to hours.

107    A common speed at which slurries were pumped was between 1 to 2 metres per second, depending on the material being pumped. Therefore, if the material was being pumped 300 metres the transit time was between 2.5 and 5 minutes. Whereas if the material was being pumped 10 kilometres, the transit time could be in excess of 2.5 hours.

108    However, the pumping of the tailings, particularly over long distances, could cause flocs formed in the tailings to break up and for the tailings to become more fluid with a decreased viscosity (often referred to as “shear thinning”). This shear thinning was a function of the mineralogy of the tailings, turbulence in the outlet pipe, the time that the floc structure was subjected to that turbulence and/or the sensitivity of the floc structure to such shear forces.

109    The distance the slurry is pumped (and therefore transit time) will also influence the amount of shear thinning that occurs. The distance (and therefore transit time) from the mine site to the point of deposition varied according to the mine and mineralogy involved. The distances involved can be measured in kilometres. The transit time from the thickener to the point of deposition could be very short and take only a few minutes or 60 minutes or longer depending on the setup of the mine site.

110    Shear thinning would typically lead to reductions in viscosity, yield stress, compressive strength and a significant reduction in permeability of the tailings on deposition in the deposition area.

111    Consequently, even if the slurry had a yield stress that was suitable for stacking on deposition when it exited the thickener, that yield stress would inevitably be significantly reduced when the slurry was pumped to the deposition area by virtue of the flocs being broken down by shear thinning.

112    Let me say something about holding vessels. One practice undertaken prior to the priority date was to temporarily divert some of the flocculated tailings to a holding vessel before pumping the diverted tailings to the deposition area.

113    Diverting some tailings to a holding vessel was undertaken for a variety of operational and logistical reasons including generation of sufficient material to achieve the required coarse/fine blending requirements or for the addition and conditioning of chemicals or other additives to the tailings which required time. Furthermore, plant logistics may have required a certain volume of tailings to be deposited at a particular time or over a particular timeframe requiring storage or accumulation of tailings in the interim.

114    Further, one could subject liquor released from deposited tailings to further processing to reclaim or re-use any valuable dissolved materials.

115    At this point let me elaborate further on the question of dewatering and various devices.

116    Dewatering screens were a class of dewatering device where solids were retained on a perforated screen, whilst liquid drained through the apertures, enabling dewatering. The aperture size was selected based on the size of the particles in the feed slurry. The drainage was either driven by gravity (a gravity drainage screen) or an acceleration force provided by a vibrating screen (a vibrating dewatering screen).

117    If the feed slurry contained free solids that were smaller than the screen aperture size, the water released from the screen was dirty and would typically require further treatment. A dewatering screen required rapidly dewatering, free-draining solids and as such was typically employed in coarse tailings dewatering.

118    Alternatively, fine particle systems were treated on gravity drainage screens after application of dosages of polymer flocculant. This created a strong, granular, freely draining floc structure that released water rapidly on the screen (typically similar to the structure developed for in line flocculation and belt press filtration). Inclination of the screen resulted in solids movement down the screen, and ploughs were often employed to gently shear/densify the flocs and enhance water release. Alternatively, a screen could be in the form of an inclined rotating cylinder, whereby a gentle tumbling action promoted floc densification and water release. Gravity drainage also formed the critical initial dewatering phase of operation of a belt press filter.

119    Adoption of dewatering screens was characterised by:

(a)    low technical complexity and capital cost, although vibrating screens would often only be one component of the full tailings management system;

(b)    consideration of additional filtration for fines present in water released;

(c)    potential requirements for conveying and trucking to the TSF;

(d)    reducing costs for TSF development and management with increasing ultimate placed solids concentration; and

(e)    improving water efficiency and environmental credentials proportional with the ultimate placed solids concentration.

120    Further, cyclones were commonly used in mineral processing to fractionate particles in a slurry into a coarse and fine fraction. Use of dewatering cyclones was commonly employed at operations where predominantly sand-sized tailings were generated e.g. in the mineral sands industry. A cyclone essentially was a cylindrical vessel with a tapered conical base in which a slurry was tangentially injected such that a swirling flow was generated. Due to centrifugal force, coarser particles would migrate faster to the outside of the vessel such that they moved downwards and were collected in an underflow stream at the base of the vessel. Fine particles moved in a flow pattern upwards and out of the centre of the cyclone in an overflow stream. Of course, anyone who has a modicum of physics education would know that the concept of “centrifugal force” is a fictitious outward force which is only apparent to observers in a rotating reference frame (e.g. a child on a playground roundabout). Rather, from an inertial reference frame, as materials enter the cyclone, the cyclone wall exerts a centripetal force towards the inside of the cyclone, which forces the slurry to undergo circular motion. Any object moving in circular motion is the subject of a centripetal force towards the centre. Think of it this way. An object in circular motion is accelerating. That is because its velocity is changing. Now its speed may be constant but its direction is changing. As velocity, a vector, is a function of speed and direction, an object in circular motion is changing velocity i.e. accelerating. And that acceleration, also a vector, is towards the centre. Force, also a vector, is a product of mass times acceleration. As the acceleration component (a vector) is towards the centre, with mass as a scalar concept, the force (centripetal force) is towards the centre. Now in the present context as the centripetal force applied is insufficient to keep the larger particles in circular motion, the radius of circular motion for such particles will increase until they migrate towards the outside of the vessel.

121    The natural behaviour of a cyclone was such that the water preferentially reported to the overflow with the fines, and the underflow contained coarse particles at a significantly higher solids concentration than in the feed. This natural dewatering effect could be enhanced by either the operating parameters or cyclone design, to promote enhanced dewatering of the cyclone underflow, such that a stackable, coarse output could be achieved.

122    Some ancillary treatment approach was required to deal with the dilute cyclone overflow stream containing fine material e.g. subsequent thickening or settling pond impoundment.

123    Adoption of cyclones was characterised by:

(a)    low technical complexity and capital cost, however they would typically only be one component of the full tailings management system;

(b)    the requirement of additional thickening / clarification for fines present in overflow;

(c)    reducing costs for TSF development and management with increasing ultimate placed solids concentration; and

(d)    improving water efficiency and environmental credentials proportional with the ultimate placed solids concentration.

124    Further, centrifuges were a dewatering technique whereby a slurry was placed within a rotating vessel which could be either solid (a solid bowl centrifuge) or with some kind of filter medium or screen at the base (basket or screen bowl centrifuge).

125    A wide range of centrifuge designs existed with differing rotational speeds, mechanical methods for solids discharge (e.g. via a helical scroll, scraping blade or plunger) and sometimes with additional vibration to enhance solids dewatering.

126    In general, centrifuges were less commonly used for tailings dewatering than other technologies. Vibrating basket centrifuges were most commonly used in fine coal dewatering, and solid bowl centrifuges were relatively commonly used in dewatering of industrial and municipal water and wastewater sludges, typically after heavy polymer flocculant dosage.

127    Adoption of centrifuges was characterised by:

(a)    moderate technical complexity and capital cost;

(b)    reducing costs for TSF development and management with increasing solids concentration; and

(c)    improving water efficiency and environmental credentials proportional with the dewatered solids concentration.

128    Further, filtration was a dewatering technique where slurry was forced against a semi-permeable filter medium which resulted in retention of slurry solids on the medium as a filter cake, with passage clarified liquid (filtrate) through the medium.

129    A large number of different filtration technologies existed at the priority date such that it is impractical to provide an exhaustive list, but the broadest classification of filter types can be made based on the driving force for separation, which can be as a positive pressure or via application of a vacuum.

130    Pressure filters typically used for tailings dewatering at the priority date were most commonly recessed plate-type pressure filters, although pressure variants of disc and drum filters were also available.

131    The belt press filter was also used for tailings dewatering and involved conditioning of thickener underflow with large doses of flocculant to produce freely dewatering slurry, with a “cottage cheese” structure. This then went through a gravity drainage phase followed by filtration under high pressure between two belts.

132    Successful operation of a belt press filter required the use of sufficiently high doses of flocculant to feed to produce the necessary freely dewatering “cottage cheese” like structure. This was necessary so that the material could be squeezed between the belts rather than being displaced by them.

133    Let me now say something about tailings beaching.

134    Conceptually, beaches can form in two ways. For a homogenous slurry, where solid particles do not rapidly settle out, a beach forms as the slurry stops moving at some point after deposition due to its viscous nature overcoming the level of energy driving the spreading of the slurry. Alternatively, a beach could form due to material settling out from the slurry as it travelled outwards from the deposition point. In some cases, where a wide range of particle sizes were present within a slurry, differential settling rates could exist, with coarser particle settling faster than the finer ones. This could then result in coarser particles being deposited closer to the deposition point, with finer material travelling progressively further down the beach. In extreme cases, the beach may be formed predominantly of coarse material with fines migrating to the decant pond. In situations where separation of the particles from the deposited slurry was the driving force for beach formation, the rate of separation was driven by the particle settling rate (i.e. a measure of permeability), which in turn controlled the beach angle.

135    A schematic illustrating the impact of particle size on separation/settling of solids during deposition is set out below. This illustrates how larger, or denser particles or flocs will tend to be deposited closer to the exit of a discharge pipe than finer particles.

136    A flocculated structure at the point of deposition could enhance beach formation by either:

(a)    binding fines to coarse particles to prevent segregation on the beach; or

(b)    promoting more rapid separation of solids from the carrier fluid.

137    In the case of thickened or paste thickened tailings discharged into a deposition area, it was desirable for the deposited tailings to form a sloped beach in the deposition area.

138    For tailings deposited with perimeter discharge, development of a beach was desirable as it promoted run off of liquid expressed from the tailings into a decant pond for easy recycling. In addition, the integrity of the TSF embankment was also enhanced through the migration of liquid away from the area adjacent to the embankment. TSF embankments were typically not constructed as water retaining facilities due to a higher cost of construction (except in circumstances where a water cover over deposited tailings was required) and as such, the presence of large volumes of water against these structures could lead to structural failure.

139    A further advantage of the development of a beach is that it allows the construction of further embankment walls on top of the beached material when the mine operator wishes to raise the dam wall. This method of construction is termed an “upstream raise”. Adopting this construction approach significantly reduces the cost of raising the dam wall, as the amount of construction material required is significantly reduced. For this to occur, the beached material needs a sufficient level of strength to support the overlying dam wall structure. As can be seen from the diagram below, the relative amount of material required for an “upstream raise” is significantly less than for a “downstream raise” or a “centreline raise”.

140    For the tailings deposited with central discharge, it was also desirable for the development of a sloped beach, which in this case would be a conical shape. The advantages of a sloped beach in such a case are to promote run off of rainfall to maintain the tailings in a dry state, and to allow storage of the maximum amount of tailings within the conical shape without either having to build a high confining embankment or to occupy a large area.

141    Generally, beaching of the tailings allowed for the storage of a greater volume of tailings for a given area of TSF, which was one of the goals of tailings management processes.

THE OPPOSED APPLICATIONS

(a)    Background

142    Let me address some background concerning the disposal of tailings.

143    Generally speaking, until about the mid to late 1950s water was not scarce and there were no restrictions on the dumping of wastes or use of tailings dams. Therefore a mine operator could usually obtain water from any easily available source and dump the tailings whenever convenient. This often meant using water from a river and dumping the wastes back into the river. Typically, mine operators simply deposited the tailings into pits, dams or rivers without further treatment or consideration of the environmental impact of the tailings, and without concern for recovering water from the tailings.

144    Prior to the 1980’s the most commonly used technique for dewatering of tailings involved the use of thickeners to recover some of the water from the tailings. To enhance the separation rate of the solid and liquid within a thickener, the generally accepted practice was to add a flocculant to the feed entering the thickener which aggregated the particles in the thickener and increased their settling rate. As the solids separated from the water in the thickener, clear water was taken from the top of the thickener. The solids removed from the bottom of the thickener were usually pumped in a slurried form to a deposition area. The deposition area could have been an existing mined-out pit, or a purpose built tailings storage facility.

145    A common method for disposing of thickener underflow was to deposit it onto a slope, wall or floor of a deposition area in order to beach the tailings, so that released water flowed to a lower point and could be pumped back to the plant for re-use (tailings beaching). A typical approach to tailings beaching was to deposit thickener underflow onto a slightly sloping surface, which enabled the material to build up on the slope with released water flowing to a lower point where it could be recovered for re-use in the mining process.

146    In the 1980s and 1990s in Australia, mine operators were faced with challenges in relation to environmental issues, water efficiency and limited land availability, and increased regulatory controls. Those factors and good economic reasons forced mine operators to look at ways to improve their tailings disposal processes.

147    As a result of these pressures, in the period leading up to the priority date, the mineral processing industry was focused on improving ways of disposing of tailings. Some objectives in the management of the treatment of tailings were to:

(a)    minimise the amount of land that was taken up by the storage of tailings to create as small a footprint as possible;

(b)    obtain a stable and trafficable deposit;

(c)    maximise recovery of relatively clean water so that the water could be re-used in the plant; and

(d)    rehabilitate the deposited tailings within an acceptable timeframe.

148    Further and in this context, the particular choice of tailings dewatering technique employed for a given operation was strongly influenced by the particular circumstances existing at any mine including:

(a)    land availability and topography;

(b)    water availability;

(c)    environmental restrictions;

(d)    access to power;

(e)    distance to deposition area from plant; and

(f)    available manpower.

149    At the priority date the most commonly used tailings dewatering techniques which involved the use of mechanical dewatering equipment included the use of thickeners/clarifiers, dewatering screens, cyclones, centrifuges, and filtration methods such as belt press filters, all of which involved the use of flocculants.

150    Co-disposal was also a technique used in some cases both to dispose of coarse waste materials and to assist in obtaining a stable deposit of tailings in the deposition area, including in tailings beaching. However, there were well known difficulties involved with co-disposal.

(b)    The invention described in the 785 application

151    The field of the invention is described in the 785 application (p 1 lines 5 to 10) in the following terms:

The present invention relates to the treatment of mineral material, especially waste mineral slurries. The invention is particularly suitable for the disposal of tailings and other waste material resulting from mineral processing and beneficiation processes, including the co-disposal of coarse and fine solids, as a homogenous mixture.

152    The 785 application notes that in some cases the tailings could be back filled into mines. In cases where it was not possible to dispose of the waste in an emptied mine, it was common practice to dispose of the waste material by pumping the aqueous slurry to lagoons, heaps or stacks and allowing it to dewater gradually through the actions of sedimentation, drainage and evaporation. Page 1 lines 26 to 30 states:

For other applications it may not be possible to dispose of the waste in a mine. In these instances, it is common practice to dispose of this material by pumping the aqueous slurry to lagoons, heaps or stacks and allowing it to dewater gradually through the actions of sedimentation, drainage and evaporation.

153    The 785 application notes the environmental pressures to minimise the allocation of new land for disposal and to more effectively use the existing waste areas. A related environmental pressure was the efficient re-use of water in the mining process. Page 2 lines 1 to 4 states:

There is a great deal of environmental pressure to minimise the allocation of new land for disposal purposes and to more effectively use the existing waste areas. One method is to load multiple layers of waste onto an area to thus form higher stacks of waste.

154    The 785 application notes that it is desirable to have a treatment which provides a more rapid release of water from the deposited material and that the concentrated solids are held in a manner that prevents both segregation of any coarse and fine fractions and contamination of the released water, whilst minimizing the impact on the environment. Page 3 line 29 to page 4 line 2 states:

It would therefore be desirable to provide treatment which provides more rapid release of water from the suspension of solids. In addition it will be desirable to enable the concentrated solids to be held in a convenient manner that prevents both segregation of any coarse and fine fractions, and prevents contamination of the released water whilst at the same time minimises the impact on the environment.

155    The 785 application recognises that one method of reducing the area of deposition was to form stacks of “rigidified waste”. One “difficulty” identified in the 785 application was to ensure that the new waste flowed over previously “rigidified” waste, remained within the waste area boundaries, formed a stack and consolidated to support multiple layers without collapsing or overflowing. Page 2 lines 3 to 8 states:

One method is to load multiple layers of waste onto an area to thus form higher stacks of waste. However, this presents a difficulty of ensuring that the waste material can only flow over the surface of previously rigidified waste within acceptable boundaries, is allowed to rigidify to form a stack, and that the waste is sufficiently consolidated to support multiple layers of rigidified material, without the risk of collapse or slip.

156    The 785 application describes the common use of flocculants to assist in the disposal of tailings through the process of flocculation in a thickener so as to give higher density to the underflow and assist in the recovery of water. Page 2 lines 16 to 18 states:

These solids are often concentrated by a flocculation process in a thickener to give a higher density underflow and to recover some of the process water.

157    The 785 application also recognises that SDITB had been used to improve the compaction of the fine waste material and clarity of the recovered water. But the 785 application asserts that such processes applying flocculants at “conventional doses” had produced little or no benefit. Page 3, lines 21 to 27 states:

Attempts have been made to overcome all the above problems by treating the feed to the tailings dam using a coagulant or a flocculant to enhance the rate of sedimentation and/or improve the clarity of the released water. However, this has been unsuccessful as these treatments have been applied at conventional doses and this has brought about little or no benefit in either rate of compaction of the fine waste material or clarity of the recovered water.

158    Dr Farrow acknowledged that the reference to “feed to the tailings dam” in the 785 application encompassed thickener underflow. As he said, the most common way of feeding materials to a tailings dam was where the material had previously gone through a thickener.

159    The 785 application refers to EP-O-388-108 (the Moody patent). This is a patent with G Moody as the inventor filed by Allied Colloids Ltd (subsequently acquired by Ciba) on 12 March 1990, although asserting an earlier priority date. The Moody patent discloses a process of secondary dosing involving the addition of a water insoluble flocculant and “allowing the material to stand and then allowing it to rigidify and become a stackable solid”. The 785 application notes that this prior art process disadvantageously requires high doses of flocculant “in order to achieve a sufficiently rigidified material”. Page 5 lines 5 to 14 states:

EP-A-388108 [sic] describes adding a water-absorbent, water-insoluble polymer to a material comprising an aqueous liquid with dispersed particulate solids, such as red mud, prior to pumping and then pumping the material, allowing the material to stand and then allowing it to rigidify and become a stackable solid. The polymer absorbs the aqueous liquid of the slurry which aids the binding of the particulate solids and thus solidification of the material. However this process has the disadvantage that it requires high doses of absorbent polymer in order to achieve adequate solidification. In order to achieve a sufficiently rigidified material it is often necessary to use doses as high as 10 to 20 kilograms per tonne of mud.

160    The 785 application (p 5 line 23) also refers to a prior art process which is described as WO-A-96/05146 (the Pearson patent). This is a patent filed by Cytec in 1995. Cytec was then a major flocculant supplier. The Pearson patent discloses a process of SDITB which involved the addition of a water soluble flocculant “dispersed in a continuous oil phase with the slurry” (p 5 line 25). This is a reference to a water-in-oil emulsion. The 785 application asserts that it is a requirement of the Pearson patent that the flocculant is not inverted and added to the slurry as an aqueous solution. At page 5 lines 25 to 29 it is stated:

Preference is given to diluting the emulsion polymer with a diluent, and which is preferably in a hydrocarbon liquid or gas and which will not invert the emulsion. Therefore it is a requirement of the process that the polymer is not added in to the slurry as an aqueous solution.

161    The 785 application also refers to the Gallagher patent that I mentioned in my introduction filed by Ciba in 2001. The Gallagher patent discloses a process of SDITB in which the treated tailings are pumped as a fluid and then allowed to stand and rigidify. The rigidification is achieved by introducing into the tailings a water soluble flocculant in the form of a dry powder (particles) which has an intrinsic viscosity of at least 3 dl/g into the suspension.

162    The 785 application describes the importance of using a water soluble flocculant in a powder form in the Gallagher patent as follows (p 6 lines 9 to 11):

The importance of using particles of water soluble polymer is emphasised and it is stated that the use of aqueous solutions of the dissolved polymer would be ineffective.

163    Now the Gallagher patent states (pages 3 and 14) that:

We have surprisingly found that the presence of water soluble polymers applied in the form of particles actually enables the material to remain fluid and pumpable during the pumping stage but results in rapid loss of fluidity and rigidification on standing.

It is surprising that the process according to the invention forms a product which rigidifies far better than alternative treatments, for instance the use of water swellable, water swellable polymers or pre-formed solutions of water soluble polymers.

164    Contrastingly, the 785 application states (p 7 lines 26 to 29):

We have unexpectedly found that the addition of the aqueous solution of polymer to the material does not cause instant rigidification or substantially any settling of the solids prior to standing.

165    It is well apparent that the form in which the flocculant is added to the tailings, namely, a water soluble flocculant in aqueous solution, is an essential feature of the claimed invention.

166    The 785 application (p 6 lines 19 to 23) describes an objective of the claimed invention as finding a more suitable method of treating coarse and/or fine particulate waste material from mineral sands, alumina or other mineral processing operations in order to provide a better release of fluids and a more effective means of disposing of the concentrated solids. The 785 application notes that the claimed invention may be used on tailings slurries that consist of only fine particles or a mixture of fine and course particles (p 8 lines 6 and 7).

167    The 785 application discloses that the claimed invention may be performed in relation to a wide range of materials and mineralogies, including gold slimes, mineral sands, coal, red mud, nickel, silver and iron ore. Further, the only mineral type which is the subject of a specific claim is mineral sands. Further, the only field trials in the Examples are in relation to mineral sands. Page 8 lines 22 to 27 states:

The material particles are usually inorganic and/or usually a mineral. Typically the material may be derived from or contain filter cake, tailings, thickener underflows, or unthickened plant waste streams, for instance other mineral tailings or slimes, including phosphate, diamond, gold slimes, mineral sands, tails from zinc, lead, copper, silver, uranium, nickel, iron ore processing, coal, or red mud.

168    The 785 application states that the most effective point of addition of the flocculant depends upon the substrate and the distance from the thickener to the deposition area. If the outlet pipe is relatively short, with the transit time being correspondingly short, then it may be advantageous to add the flocculant close to the thickener. Conversely, if the deposition area is remote from the thickener, with the transit time being correspondingly long, then it is desirable to add the flocculant closer to or at the discharge point. Page 14 lines 10 to 17 states:

The most effective point of addition will depend upon the substrate and the distance from the thickener to the deposition area. If the conduit is relatively short any may be advantageous to dose the polymer solution close to where the material flows from the thickener. On the other hand, where the deposition area is significantly remote from the thickener in may be desirable to introduce the polymer solution closer to the outlet. In some instances in may be convenient to introduce the polymer solution into the material as it exits the outlet.

169    The 785 application states (p 14 lines 5 to 10) that:

A suitable and effective rigidifying amount of the water-soluble polymer solution can be mixed with the material prior to a pumping stage Alternatively, the polymer solution can be introduced and mixed with the material during a pumping stage or subsequently.

170    The 785 application also states that, in one form of the claimed invention, the treated material is transferred to a settling area, such as a tailings dam or a lagoon (p 16 line 32 to p 17 line 1).

171    The 785 application states that (p 13 lines 19 to 22):

The aqueous polymer solution may be added in any suitable concentration. It may be desirable to employ a relatively concentrated solution, for instance up to 10% or more based on weight of polymer in order to minimise the amount of water introduced into the material.

172    The 785 application then indicates that it will usually be desirable to add the flocculant solutions at lower concentrations to facilitate the distribution of the flocculant throughout the material (p 13 lines 22 to 25).

173    Suitable dose ranges for performing the claimed invention are identified as ranging from 10 grams to 10,000 grams per tonne (g/t) of material solids with preferred ranges of 30 to 3,000 g/t and more preferred ranges from 60 to 200 or 400 g/t (p 8 lines 16 to 20).

174    The release of liquor is a preferred feature of the claimed invention, such liquor containing “significantly less solids” and being suitable for recycling in the process (p 16 lines 9 to 18). The 785 application therefore makes clear that it is not necessary for the effective working of the claimed invention that all solids in the treated tailings be captured in the stacked material.

175    The 785 application contains 13 examples. Examples 1 to 8 involve slump tests on various different selections of flocculants on different mineral slurries. Examples 9 to 11 are laboratory evaluations of various flocculants. Examples 12 and 13 include a plant evaluation using tails from a mineral sands process.

176    Example 1 involves slump tests using a tailings slurry obtained from a mineral sands operation. The results of the slump tests are provided in Table 3. The 785 application then notes that: “The increased rigidification of the mineral tailings through the addition of the water soluble polymer is evident by the reduced slump radius and increased stacking height compared to the untreated material” (p 21 lines 2 to 6).

177    Examples 1 to 11 involve a comparison with thickener underflow that was not treated in any way to assess the increased rigidification achieved by the addition of water soluble flocculant in aqueous solution (see the results reported in Tables 2 to 19).

178    In Example 12, laboratory slump tests were conducted on a combined fine and coarse tailings from a mineral sands operation. Following these tests, a plant evaluation was carried out. The 785 application states that “[b]ased on the above laboratory evaluation, a dosing point close (20 metres or 11 seconds) to the discharge point was chosen to minimize shearing…” (p 35 lines 15 to 17). Figure 5 (untreated) and Figure 6 (treated) show the discharge of the mineral sands tailings the subject of the plant evaluation.

179    Example 13 also involves laboratory and plant evaluations on a combined fine and coarse tailings from a mineral sands operation. The 785 application states that “[t]he combined waste material is pumped to a series of pits that are filled sequentially and re-vegetated afterwards” (p 38 lines 8 to 10).

180    In Example 13, in one case product was added before the small centrifugal pump. The result was an improved slump angle and much greater release of water from the slurry (p 38 lines 18 to 21). The dosing point was then modified to be added directly after the centrifugal pump. The dosage was reduced, with this alternative position achieving similar results to the first dosing point (p 39 lines 2 to 4).

(c)    The claims of the 785 application

181    The 785 application (as amended) makes 29 claims. These claims are expressed as follows:

1.    A process of improving rigidification of a material whilst retaining the fluidity of the material during transfer in which the material comprises an aqueous liquid with dispersed particulate solids that is transferred as a fluid to a deposition area, then allowed to stand and rigidify, by combining with the material during transfer an effective rigidifying amount of an aqueous solution of a water-soluble polymer having an intrinsic viscosity of at least 5 dl/g (measured in IM NaCI at 25℃), in which the process comprises co-disposal of coarse and fine solids as a homogenous mixture.

2.    A process according to claim 1 in which the water soluble polymer has an intrinsic viscosity of at least 5 dl/g and is formed from ethylenically unsaturated water-soluble monomer or blend of monomers.

3.    A process according to claim 1 or claim 2 in which the water soluble polymer is anionic.

4.    A process according to claim 3 in which the polymer is formed from monomer(s) selected from the group consisting of (meth)acrylic acid, allyl sulphonic acid and 2-acrylamido-2-methyl propane sulphonic acid as the free acids or salts thereof, optionally in combination with non-ionic co-monomers, selected from the group consisting of (meth)acrylamide, hydroxy alkyl esters of (meth)acrylic acid and N-vinyl pyrrolidone.

5.    A process according to claim 1 or claim 2 in which the water soluble polymer is non-ionic.

6.    A process according to claim 5 in which the polymer is formed from monomer(s) selected from the group consisting of (meth) acrylamide, hydroxyl alkyl esters of (meth) acrylic acid and N-vinyl pyrrolidone.

7.    A process according to claim 1 or claim 2 in which the water soluble polymer is cationic.

8.    A process according to claim 7 in which the polymer is formed from monomer(s) selected from the group consisting of dimethyl amino ethyl (meth) acrylate-methyl chloride, (DMAEA.MeCI) quat, diallyl dimethyl ammonium chloride (DADMAC), trimethyl amino propyl (meth) acrylamide chloride (ATPAC) optionally in combination with non-ionic co-monomers, selected from the group consisting of (meth) acrylamide, hydroxy alkyl esters of (meth) acrylic acid and N-vinyl pyrrolidone.

9.    A process according to any one of claims 1 to 8 in which the dispersed particulate solids are mineral.

10.    A process according to any one of claims 1 to 9 in which the process comprises the disposal of mineral slurry residues from a mineral processing operation.

11.    A process according to any one of claims 1 to 10 in which the process provides a heaped geometry.

12.    A process according to any one of claims 1 to 11, in which the material is derived from the tailings from a mineral sands process.

13.    A process according to any one of claims 1 to 12 in which the dispersed particulate solids have particle sizes less than 100 microns, in which preferably at least 80% of the particles have sizes less than 25 microns.

14.    A process according to any one of claims 1 to 13 in which the dispersed particulate solids has a bimodal distribution of particle sizes comprising a fine fraction and a coarse fraction, in which the fine fraction peak is substantially less than 25 microns and the coarse fraction peak is substantially greater than 75 microns.

15.    A process according to any one of claims 1 to 14 in which the material has a solids content in the range 15% to 80% by weight, preferably in the range 40% or 50% to 70% by weight.

16.    A process according to any one of claims 1 to 15 comprising flocculating an aqueous suspension of solids in a vessel to form a supernatant layer comprising an aqueous liquor and an underflow layer comprising thickened solids forming the material, separating the supernatant layer from the underflow, wherein the underflow containing the particulate material flows from the vessel and, in which the material is then pumped to a deposition area where it is allowed to stand and rigidify, and wherein the effective rigidifying amount of the aqueous solution of the water-soluble polymer is mixed with the material after flocculating the suspension and before the material is allowed to stand.

17.    A process according to claim 16 in which wet or dry coarse particles are added to the underflow from the vessel either before or during the addition of an effective rigidifying amount of the water soluble polymer.

18.    A process according to claim 16 or 17 in which the material is transferred to a holding vessel prior to being pumped to the deposition area.

19.    A process according to any of claims 1 to 18 in which the process provides a heaped geometry and co-immobilises the fine and coarse fractions of the solids in the material and water released has a higher driving force to separate it from the material by virtue of hydraulic gravity drainage.

20.    A process according to any of claims 1 to 19 in which the material is pumped to an outlet, where it is allowed to flow over the surface of previously rigidified material, wherein the material is allowed to stand and rigidify to form a stack.

21.    A process according to any one of claims 1 to 20 in which the effective rigidifying amount of the aqueous solution of the water-soluble polymer is mixed with the material prior to a pumping stage.

22.    A process according to any of claims 1 to 20 in which the effective rigidifying amount of the aqueous solution of the water-soluble polymer is mixed with the material during or subsequent to a pumping stage.

23.    A process according to any claims 1 to 21 in which the effective rigidifying amount of the aqueous solution of the water-soluble polymer is mixed with the material as it exits the outlet.

24.    A process according to any of claims 1 to 23 in which the material is dewatered during rigidification, releasing liquor.

25.    A process according to claim 24 in which the liquor is recycled to a mineral processing operation.

26.    A process according to claim 24 or 25 in which the clarity of the liquor is improved by the addition of an aqueous solution of water-soluble polymer.

27.    A process according to any of claims 24 to 26 in which the liquor contains dissolved valuable materials and, in which the liquor is subjected to further processing to reclaim or re-use the valuable materials.

28.    Use of an aqueous solution of a water-soluble polymer having an intrinsic viscosity of at least 5 dl/g (measured in 1 M NaCI at 25), in a process in which material comprising aqueous liquid with dispersed particulate solids is transferred as a fluid to a deposition area, then allowed to rigidify, for the purpose of improving rigidification whilst retaining fluidity of the material during transfer, in which the process comprises co-disposal of coarse and fine solids as a homogenous mixture.

29.    A process of improving rigidification of a material whilst retaining the fluidity of the material during transfer according to claim 1 substantially as hereinbefore described with reference to the examples.

182    A number of observations may be made in relation to the scope of claim 1 of the 785 application which contains the following integers:

A process of improving rigidification of a material whilst retaining the fluidity of the material during transfer

in which the material comprises an aqueous liquid with dispersed particulate solids that is transferred as a fluid to a deposition area, then allowed to stand and rigidify,

by combining with the material during transfer

an effective rigidifying amount of

an aqueous solution of a water-soluble polymer having an intrinsic viscosity of at least 5 dl/g (measured in IM NaCI at 25),

in which the process comprises co-disposal of coarse and fine solids as a homogenous mixture.

183    First, the process of claim 1 is not limited by the material having any particular solids content. The only limitation as to the solids content of the material is introduced in claim 15 which limits the process to materials having a solids content of 15 to 80%.

184    Second, the process of claim 1 is not limited by reference to a dose range for the flocculant. The claim requires an “effective rigidifying amount”. This refers to the amount of high molecular weight flocculant solution needed to cause the material to “rigidify”. Although the claim does not specify what this amount might be, the 785 application states that a “suitable” dose will range from 10 grams to 10,000 g/t of material solids.

185    Third, claim 1 requires that the aqueous solution of a water-soluble flocculant be added “during transfer”. It does not appear to be in dispute that this term refers to the addition of flocculant to the thickener underflow and encompasses the flocculant being added at any point between the point at which the underflow is discharged from the thickener to the point at which the treated material is discharged into the deposition area.

186    Fourth, claim 1 requires the deliberate combining of coarse and fine materials.

187    Fifth, claim 1 encompasses the flocculant being added either before, during or after the coarse and fine materials are combined in the thickener underflow.

188    Finally, the process of claim 1 requires that the treated material be allowed to “stand and rigidify” in the deposition area. It does not exclude subsequent removal of the rigidified material or the material standing and rigidifying on a physical structure in the deposition area.

189    Claims 2 to 8 depend on claim 1 and introduce limitations concerning the chemistry of the water soluble flocculant.

190    Claims 9 and 10 limit the material being treated to minerals.

191    Claim 11 refers to a process which provides a heaped geometry.

192    Claims 12 to 15 limit the material being treated by reference to mineralogy (mineral sands), particle size, bimodal distribution of particle size and solids content.

193    Claim 16 is a dependent claim which requires that the tailings being treated are underflow from a thickener.

194    Claim 17 introduces the limitation of adding wet or dry coarse particles to the underflow before or after the addition of the water soluble flocculant.

195    Claim 18 introduces the limitation of the material being transferred to a “holding vessel” before being pumped to the deposition area.

196    Claim 19 limits the process of claim 1 by further requiring that the process provide a “heaped geometry” and co-immobilisation of the fine and coarse fractions of the solid material, with the water released having a higher driving force to separate it from the material by virtue of hydraulic gravity drainage.

197    Claim 20 requires that the material when deposited is allowed to flow over the surface of previously rigidified material and is allowed to stand and rigidify to form a stack.

198    Claims 21 to 23 introduce limitations concerning the mixing of the flocculant with the material prior to, during or subsequent to pumping or as the material exits the outlet.

199    Claim 24 requires the material to be dewatered during rigidification, releasing liquor.

200    Claims 25 to 27 introduce limitations concerning the liquor which is released from the material being treated.

201    Claim 28 is an independent claim which is not relevantly different in scope to claim 1. It claims the use of an aqueous solution of a water soluble flocculant having an intrinsic viscosity of at least 5 dl/g (measured in 1M NaCL at 25℃), in a process in which material comprising aqueous liquid with dispersed particulate solids is transferred as a fluid to a deposition area, then allowed to rigidify, for the purpose of improving rigidification whilst retaining fluidity of the material during transfer, in which the process comprises co-disposal of coarse and fine solids as a homogenous mixture.

202    Claim 29 is an omnibus claim for the process according to claim 1, limited to the examples in the 785 application.

(d)    The claims of the 568 application

203    The body of the 568 application is substantially similar to the body of the 785 application and contains no significant differences of relevance to the present discussion. Accordingly, it is not necessary to set it out. The 568 application (as amended) makes 35 claims. It is not necessary to set them out save for claim 1 which stipulates:

1.    A process of improving rigidification of a material whilst retaining the fluidity of the material during transfer, in which the material comprises an aqueous liquid with dispersed particulate solids that is transferred as a fluid to a deposition area, then allowed to stand and rigidify, the process comprising:

combining aqueous suspensions of fine and coarse particulates for the purpose of co-disposal to form the material;

mixing of the aqueous suspensions into a homogeneous slurry; and

during or after mixing of the aqueous suspensions, combining with the material during transfer an effective rigidifying amount of an aqueous solution of a water-soluble polymer having an intrinsic viscosity of at least 5 dl/g (measured in 1 M NaCl at 25°C).

204    Claim 1 accordingly contains the following integers:

a process of improving rigidification of a material whilst retaining the fluidity of the material during transfer;

in which the material comprises an aqueous liquid with disbursed particulate solids that is transferred as a fluid to a deposition area, then allowed to stand and rigidify,

the process comprising:

combining aqueous suspensions of fine and course particulates for the purpose of co-disposal to form the material;

mixing of the aqueous suspensions into a homogeneous slurry; and

during or after mixing of the aqueous suspensions, combining with the material during transfer an effective rigidifying amount of an aqueous solution of a water-soluble polymer having an intrinsic viscosity of at least 5 dl/g (measured in 1M NaCl at 25).

205    Claim 2 describes a process which involves treating material which is the “underflow from a thickener”. A flocculant is combined with that material and transferred to a deposition area. The process comprises the co-disposal of coarse and fine solids as a homogenous mixture.

206    Claim 3 describes a process which involves treating material which is the “underflow from a thickener”. A flocculant is combined with that material at a “selected dosing point” and transferred to a deposition area. The process comprises the co-disposal of coarse and fine solids as a homogenous mixture.

207    Claim 4 is dependent on claim 3 and adds the requirement that the dosing point to be “near an outlet where the material is discharged in the deposition area”.

208    Claim 5 is dependent on claim 3 and adds the requirement that the water-soluble flocculant is added “to a conduit containing a flow of the material”. Page 7a of the 568 application discloses that “[t]he conduit is any convenient means for transferring the material to the deposition area and may for instance be a pipe or a trench”.

209    Claim 6 is dependent on claim 3 and adds the requirement that the flocculant has an intrinsic viscosity of at least 5 dl/G.

210    Claims 7 to 11 introduce limitations concerning the solids content of the material being treated.

211    Claims 12 to 18 introduce limitations concerning the properties of the flocculant.

212    Claims 19 to 21 introduce limitations on the source of the material being treated.

213    Claims 22 to 24 limits the material being treated by particle size and particle size distribution.

214    Claim 25 is a dependent claim which is expressed in similar terms to claim 1, given that the words “during transfer” in claim 1 of the 785 application means that the material being treated is thickener underflow.

215    Claim 26 is dependent on claim 25 and adds the requirement that coarse particles (wet or dry) are added to the underflow before or during the addition of the flocculant.

216    Claim 27 is dependent on claim 25 or claim 26 and adds the requirement that the material is transferred to a “holding vessel” before being pumped to the deposition area.

217    Claim 28 is dependent on any of the preceding claims and adds the requirement that the material is discharged from an outlet and allowed to flow over the surface of previously rigidified material then stand and rigidify to form a stack.

218    Claims 29 and 30 are dependent on any of the preceding claims and add the requirement that the mixing of the flocculant with the material being treated occurs “during or subsequent” to pumping (claim 29) or as the material “exits the outlet” (claim 30).

219    Claims 31 to 33 introduce limitations concerning recycling the water/liquor released from the material being treated recovering valuable materials contained in the water/liquor.

220    Claim 34 is an omnibus claim.

WITNESSES

(a)    SNF’s witnesses

221    Dr de Kretser is a technical engineering consultant in the areas of tailings and waste sludge management, solid-liquid separation, chemical conditioning, heap leach hydrodynamics, suspension rheology and the development and execution of laboratory and site based testing programs in these areas. He has a blend of specialised theoretical and practical experience gained across a range of industries including mining and waste water management in the period 1990 to date.

222    His doctoral thesis at the University of Melbourne completed in 1995 was in respect of “the compression dewatering and rheology of slurried coal mining tailings”. His area of research speciality was the “characterisation of the permeability and compressibility properties of materials” and developing techniques for understanding the behaviour of materials including how flocculation or particle size changes that behaviour. The focus of his research was on filtration methods, sedimentation methods, gravity permeation methods, all of which were methods which probed the permeability and compressibility of flocculated sediments over a range of solids concentrations. Dr de Kretser used that information in the context of optimising any one of those solid/liquid separation processes, ranging from settling through to consolidation in a tailings dam or filtration. I will return to discuss his thesis later as it has relevance to some of his evidence.

223    In the period 1996 to 2003, Dr de Kretser worked at the University of Melbourne, where he worked on numerous collaborative projects with industry, including various Australian Minerals Research Association (AMIRA) projects. His numerous collaborative projects involved site work and practical application in the field of research and development. He worked on many of the projects undertaken by AMIRA in the period 1996 to 2003, including:

(a)    AMIRA project 527 which was concerned with flocculating tailings in the context of the Bayer Process. This work involved laboratory, pilot scale and large scale field work at alumina sites in Western Australia and Queensland, including working on thickeners and examining the properties of the thickener underflow “going into a tailings deposition area”.

(b)    AMIRA project 266 which was concerned with improving thickener technologies.

224    Before the priority date, Dr de Kretser was involved in a number of projects which concerned the disposal of coarse and fine materials in mineral tailings, and in particular was aware:

(a)    through his work in the alumina industries of issues with coarse and fine materials through the thickening process;

(b)    of the coarse and fine tailings disposal and co-disposal work that was being undertaken in the oil sands industry in Canada; and

(c)    of the problems that arise when coarse and fines segregate during deposition and also during transport.

225    Following the priority date, Dr de Kretser continued to work at the University of Melbourne where he continued to collaborate on research based projects with industry. As part of his work for the University both before and after the priority date Dr de Kretser visited a number of mineral processing plants, including alumina refineries, copper, uranium, lead, zinc and coal processing plants. In 2011 Dr de Kretser left the University and was employed by Rio Tinto as a principal advisor in waste water and tailings for 4.5 years. As a result of that work, Dr de Kretser managed laboratory and field testing programs including in relation to dewatering technologies, dry and semi-dry tailings disposal, and inline flocculation.

226    Mr Russell Schroeter has been the Managing Director of SNF since 2002. Mr Schroeter holds a bachelor degree in Applied Science from the Gordon Institute of Technology. Mr Schroeter has worked in the field of mine tailings disposal since 1978 including for Nalco Chemical Company (Nalco). Between 1978 and the priority date, Mr Schroeter regularly visited mine sites to observe their tailings disposal processes and identify potential uses for flocculant in those processes. In particular, Mr Schroeter visited numerous mines around Australia which used thickeners in the treatment of tailings comprising a range of mineral types including coal, alumina, gold, nickel, mineral sands, copper, lead, zinc and uranium. Mr Schroeter has listed 29 mine sites around Australia which he personally visited before the priority date. Between 1994 to April 2002, Mr Schroeter worked for Nalco in New Zealand and then in Singapore. Although Mr Schroeters primary role concerned the sale of flocculants in the pulp and paper industry, he regularly attended internal meetings and remained aware of Nalco’s work in the mining industry during that time. He also continued to have some direct involvement in the mining industry. For example, Mr Schroeter became aware of the Pearson patent in around 1996, in the context of trying to sell flocculants to the Waihi gold mine in New Zealand.

227    Dr Nicholas Clarke is currently the Manager Metallurgy in the international division at AngloGold Ashanti Australia and has a PhD in mineral processing from Leeds University. Dr Clarke has over 40 years experience in the mineral processing industry, which included working for Iluka Resources Ltd (Iluka), which had been the majority shareholder of Consolidated Rutile Limited (CRL), between 2002 and 2006 as a principal process engineer within the R&D group at Iluka.

228    Mr Carl Buckland is the Operation Manager for Borals South-East Queensland Quarries and in the period 1999 to 2002, was the Assistant Quarry Manager at Borals Stapylton Quarry in Queensland.

229    Mr Peter Holtzman was employed during the period 1992 until May 2015 by Cable Sands (WA) Pty Ltd (Cable Sands) as a metallurgical technician. He is now retired.

230    Mr Michael Schmidt is self-employed. Mr Schmidt has a Diploma in Applied Science from North-West TAFE, Tasmania. From 1997 to 2009 Mr Schmidt was employed by Nalco as a Technical Sales Representative. Between 2010 and March 2016, he was employed by SNF as an account manager. In both of his roles he promoted and sold flocculants for use in treatment processes including the OreBind process, which I will describe and elaborate on later.

231    Mr Ronald Coleman has a Bachelor of Science (Tech) in Chemical Engineering and has been involved in the mining industry for over 30 years. His specific expertise is in the field of waste treatment and water recovery including advising on and developing processes for dealing with and recovering water from waste materials including at the Londonderry Mine in around 1984.

232    Mr Peter Woolley is an account manager employed by Integrated Water Management Pty Ltd. He has a Bachelor of Science degree from the University of New South Wales. Mr Woolley was employed by Nalco (and its predecessor) for 28 years. He was involved in sales and marketing for various applications including applications in the treatment of wastes in the mining industry. This involved regularly visiting mining sites, including the Londonderry Mine in New South Wales, and observing the processes used to treat and dispose of mining waste.

233    Mr Jim Cigulev has been a Technical Services Superintendent at Doral Mineral Sands for around 14 years, and has more than 32 years’ experience in the mining industry. Mr Cigulev has a Bachelor of Science degree with a major in chemistry from the University of Western Australia. In the period 1996 to 1999, Mr Cigulev was a metallurgist working for BHP at its Beenup Mine in Western Australia. In 1999 to March 2002, Mr Cigulev was a contract process engineer working for Iluka.

234    Mr Daniel Bembrick worked for Allied Colloids (which became Ciba) as a laboratory analyst for a number of years assisting with trials and laboratory work, which included analysing the effectiveness of flocculants and performing quality control tests on flocculants. Mr Bembrick has a Bachelor of Science majoring in Chemistry from Charles Sturt University. Mr Bembrick worked for Ciba from 1997 until April 2003, and then for SNF from November 2003 to January 2012 as an account manager. Mr Bembrick now works as a process consultant / sales manager for the BTX Group, which is a distributor of water treatment chemicals. His experience also included visiting mines on a regular basis for the purposes of attempting to sell flocculants to mine operators.

235    Ms Janine Herzig graduated from the University of Queensland in 1992 with a Bachelor of Engineering (Honours) in metal processing. She has approximately 25 years’ experience in the Field, including as a Plant Metallurgist at CRL’s mineral sand mine at Yarraman on North Stradbroke Island in Queensland for most of the time between August 1995 and February 2003. From February 2003 until April 2005, Ms Herzig was the principal metallurgist for Iluka.

236    All of these witnesses were cross-examined. SNF also tendered evidence from Dr John Ralston and Mr Brett Wroth but they were not cross-examined.

237    In general, many of the witnesses called by SNF gave their oral evidence truthfully, and to the best of their ability in circumstances where the events in question had often occurred decades earlier. But there were significant parts of their oral evidence that departed from their affidavit evidence, and reconstruction was manifest. Let me give some examples.

238    Each of Mr Holtzman, Mr Buckland, Mr Schroeter and Mr Cigulev sought to give truthful evidence when cross-examined, but in many instances their oral evidence significantly departed from their affidavit evidence.

239    The affidavit of Mr Buckland asserted that “[t]he process we implemented was to dewater the tailings in the impoundment area and there was definitely no further mechanical dewatering step. This material was dewatered on deposition and clear water was captured in the impoundment area. The excavator was subsequently used after the material had dewatered to move it to another location for ultimate disposal. … The material which was excavated out of the pit had very little water in it”. In contrast, his oral evidence was that “an essential component” of the process was the use of a digger, which would invariably “collect some water because … there was nowhere for the water to go out of the silt pond” (Mr Buckland qualified that although “[t]he pond was not full of water … there was always water left in the pond”), and that consideration was given to the use of a screw classifier instead of a digger and it is safe to assume that this is because the treated material was not sufficiently or adequately dewatered.

240    Mr Holtzman asserted in his affidavit evidence in the context of SNF’s allegations of anticipation that he had considered “collectively” four documents sent between Nalco and Cable Sands in 2002 that were said to disclose the process described in the opposed applications and that he recalled the precise details of conversation/s he had with a Nalco representative around the same time that the documents were created. But when cross-examined, there was doubt about him seeing one document. Further, he did not have any recollection of reading the documents together, although the three documents he read were “linked” in his mind, and that he did not have a recollection of the specific conversation(s) alluded to in his affidavit. Mr Holtzman was nevertheless an honest witness.

241    The affidavit of Mr Schroeter asserted immediately under the heading “secondary flocculation” and without any reference to mechanical dewatering, that people in the field knew, at the priority date, “that it was advantageous to add flocculant at more than one point in the tailings disposal process”. But when cross-examined, Mr Schroeter clarified that in his personal experience he had “only added it [flocculant] upstream of belt presses and high-speed centrifuges” and that, as at the priority date, people reporting to him within SNF only had experience with secondary dosing “upstream of some form of mechanical dewatering device”. But generally speaking I found Mr Schroeter to be very frank and commercial.

242    The affidavit of Mr Cigulev asserted that “The process used at Beenup was a co-disposal process”. Contrastingly, when cross-examined his evidence was “we were trying to remove the sand. So there was no conscious – you know, decision to add sand”. Earlier he had said “the process was inefficient and sand was in there due to that inefficiency”. I thought that Mr Cigulev was fudging some aspects of his evidence.

243    Mr Woolley and Mr Coleman sought to give truthful evidence when cross-examined regarding their recollections of events at the Londonderry mine. But for the reasons I have discussed elsewhere, there are some difficulties with their evidence.

244    Mr Woolley and Mr Coleman previously gave evidence in the 2008 proceedings regarding their work in 1980 and 1984 at the Londonderry mine. But their evidence before me conflicted in various respects with the evidence they had given earlier in the 2008 proceedings.

245    Both Mr Woolley and Mr Coleman confirmed that they gave their new evidence before me without obtaining any further materials to the materials relied upon to prepare their affidavits in the 2008 proceedings to assist them to recall events occurring more than 30 years earlier. Moreover, they had visited numerous mine sites in the decades after their work at Londonderry. Cross-examination revealed that their evidence was in some respects unreliable, and involved an attempt to reconstruct what they thought might have happened at Londonderry, by reference to photographs, videos, and documents they had subsequently reviewed over the ensuing decades, as well as conversations with one another. I agree with BASF that the best evidence of what occurred at Londonderry is not the reconstructions of Mr Woolley and Mr Coleman, but the three photographs adduced in evidence. These photographs, taken in 1981, were of the super flocculation process working at Londonderry installed by Mr Woolley. They were attached to an affidavit of Mr Woolley filed in the 2008 proceedings.

246    Further, Mr Schmidt sought to give truthful evidence regarding his experience with the OreBind process and the trial work he undertook at Ernest Henry. But some aspects were reconstruction. Other aspects I thought were beyond his expertise.

247    As with many of SNF’s witnesses, it became clear during cross-examination that his affidavit did not in many respects accurately reflect his evidence. Further, he gave his evidence after he had watched a promotional video made by Nalco in 2011 regarding the OreBind process, which he assumed to be the same as the process he had promoted in 1999 and which he relied upon to assist him to describe the OreBind process. But this placed the reliability of his evidence on an uncertain footing, particularly given his concession that he had not witnessed the implementation of the process depicted in the 2011 video as at the priority date. Moreover, the description given of the Ernest Henry testwork in his affidavit had some problematic aspects exposed during his cross-examination.

248    Further, I have rejected various aspects of the evidence of Ms Herzig. In cross-examination she adopted an argumentative and apparently partisan approach. Further, her evidence was on some occasions contradicted by the contemporaneous documents and the evidence given by other witnesses. In these circumstances, I am cautious about accepting some of the contested aspects of her evidence.

249    Further, I have adopted a cautious approach to aspects of the evidence given by Mr Bembrick, particularly his affidavit evidence concerning the work he undertook whilst employed by Ciba. It became apparent at trial that his affidavit evidence bore significant differences to his oral evidence and that his affidavit had involved reconstruction, undertaken many years after the relevant events had occurred, and in circumstances where Mr Bembrick had reviewed the opposed applications.

250    Further, Dr Clarke was a clever and honest witness. The oral evidence of Dr Clarke was that the secondary dosing process trialled at Yarraman (and subsequently Yoganup) was “a new process that had not been applied at full scale”, that he attended the Yarraman trials to see for himself how the process worked, that he was impressed with the results, and that he considered the trial work to involve new techniques confidential to Iluka and Ciba. As BASF submitted, Dr Clarke had extensive experience in the field such that weight should be placed on his evidence that he had never before his visit to Yarraman observed a process involving secondary dosing of flocculant close to the discharge point at any mine site.

251    Finally, I will discuss aspects of Dr de Kretser’s evidence in a moment.

(b)    BASF’s witnesses

252    Dr John Farrow holds a doctorate in physical chemistry and since 1984 has been employed by the CSIRO.

253    Since 1984, his main research interests have been in surface chemistry and solid-liquid separation in the mining industry, focusing on flocculation, thickener technology and filtration. The primary focus of Dr Farrow’s work was on thickener technology. Dr Farrow had less knowledge in the use of belt press filters or centrifuges.

254    There is a contrast between the backgrounds of Dr Farrow and Dr de Kretser. Dr Farrow had more extensive industry experience and had extensively worked in the field at the priority date. Further, Dr Farrow gave evidence on secondary dosing and common general knowledge which accorded more with the evidence of the fact witnesses than did the evidence of Dr de Kretser. In these circumstances, I have preferred the evidence of Dr Farrow on various aspects that I will discuss later.

255    Mr Stephen Scammell was a sales account manager employed by Ciba in the period from 1991 to 2007. Prior to 1991, Mr Scammell had worked for Catoleum (Nalco) as a sales representative. Mr Scammell was the Ciba sales representative responsible for the CRL account for CRL’s mineral sands mine on North Stradbroke Island in 2002 and 2003. He is one of the named inventors of the claimed invention.

256    Mr Scammell has not been employed by Ciba/BASF for over a decade and is now employed by Sibelco, which was formerly CRL. Mr Scammell gave his evidence carefully and reliably. Mr Scammell made concessions where appropriate and without suggesting that he had a precise recollection of all events occurring years ago. Where there is a conflict between the evidence of Mr Scammell and Ms Herzig, I have accepted his evidence.

257    Mr John Bellwood is Head of Mining Technology Asia Pacific for BASF and has had management roles with Ciba/BASF and its predecessors since 1996.

258    Ms Angela Beveridge is a Senior Technical Specialist with BASF and has performed this role with Ciba/BASF and its predecessors since 1995.

259    Mr Bellwood and Ms Beveridge both gave evidence regarding the Project Zenith / Son of Zenith project, including those aspects of the project they were involved with. Both witnesses were cross-examined regarding the contents of documents they had authored in 2002 to 2003. Their evidence was generally coherent and consistent with other evidence including the contemporaneous documents tendered. They were not cross-examined on large parts of their evidence. I did though consider that some of Ms Beveridge’s evidence was vague and that her memory on occasion was imperfect. I found Mr Bellwood to be an honest, technically skilled and impressive witness although at times overly careful.

260    Mr Martin Edgar is the Regional Sales Manager Asia for BASF. He was only cross-examined briefly.

261    Mr John Lee is a partner at Gilbert & Tobin, BASF’s solicitors. He was not cross-examined.

THE PERSON SKILLED IN THE ART

262    Let me at this point say something about the person skilled in the art.

(a)    Who is the person skilled in the art?

263    Now it is trite to observe that a patent application should be construed through the eyes of the hypothetical person (or team of persons) who is likely to have a practical interest in the subject matter of the claimed invention and may often work in the field of technology to which the claimed invention relates.

264    In the present case, the person skilled in the art is a team of people who had a practical knowledge of, and experience in, the treatment and disposal of mining wastes (including mine tailings) at the priority date including people working for flocculant suppliers, who typically have some tertiary or TAFE qualification in chemistry, consultants who have specialised knowledge and expertise in matters such as rheology and chemistry, and people working for mine operators, who typically have some tertiary or TAFE qualification in chemistry or metallurgy. The relevant field, to use Dr Farrow’s language, was “the art of dealing with tailings in mines, albeit adopting a description used in the 2008 proceedings. The relevant field was mine tailings disposal which of course included the dewatering of slurries, slimes and tailings generally and as part thereof encompassed tailings beaching and the deposit of tailings.

265    It is apparent from the evidence that flocculant suppliers had important knowledge and experience concerning the treatment and disposal of mining wastes (including mine tailings). Flocculant suppliers would regularly attend mine sites to assist and advise the mine operators on the use of their flocculants including the use in thickeners, belt press filters and centrifuges. Flocculant suppliers regularly attended mine sites to promote the use of their flocculants and undertake trials and tests for that purpose. Further, if a miner encountered an issue which involved the use of flocculants, their first port of call would have been to a flocculant supplier, such as Ciba, Nalco, Cytec or SNF. For example, a decision to add a second dose of flocculant was something that a mine operator might undertake after discussions between the flocculant supplier and the mine processing manager if confronted by a problem in the treatment of its tailings. A mine operator could look to the experience offered by flocculant suppliers to assist in dealing with that problem and the knowledge of in-house metallurgists and chemists employed by flocculant suppliers. Further, when more complex issues were confronted, mine operators could seek assistance from consultants with specialised knowledge and expertise. Those consultants could be sourced from the CSIRO or universities and could include consultants experienced or qualified in a broad range of different disciplines including metallurgists, chemists and chemical engineers.

266    In my view, all of SNF’s witnesses who gave oral evidence were persons skilled in the art at the priority date. All of those witnesses had relevant experience and knowledge of the use of flocculants, flow properties of mine waste/tailings and performance and interpretation of flocculation tests, although some were clearly more experienced than others.

267    Further, BASF’s witnesses were all persons skilled in the art at the priority date.

(b)    Information persons skilled in the art had regard to

268    The accumulated knowledge of the person skilled in the art was derived from a variety of sources which I will briefly summarise at this point.

269    Dr de Kretser said that people working in industry routinely refer to papers, patents, and others industry projects such as the AMIRA 266 project in order to keep abreast of developments in the field. Indeed, it was part of their job description to stay on top of those developments.

270    His experience at Rio Tinto was that there was an extensive internal library of relevant journals which people within that organisation would have referred to. Rio Tinto also had a specific knowledge management practitioner who assisted people in making searches if they needed to find information in relation to a certain task or topic.

271    Dr de Kretser accepted that there was a spectrum of levels of sophistication in terms of searching in the industry depending upon the size and sophistication of the particular operation or operator. He said that any mid-tier miner would necessarily have people in a technology development and deployment role, and that those people would necessarily keep on top of technological developments of interest and relevance to their operations.

272    Further, he said that if a persons role within the organisation involved technology development and protection of that technology, they would be focused on patent literature. If a company was developing technologies for the mining industry, it would consult patent literature in order to ascertain what its competitors were patenting, so as to understand the focus of their work and to avoid any infringement risk. Whilst working at Rio Tinto, Dr de Kretser regularly observed colleagues with patents on their desk.

273    Further, flocculant suppliers undertook patent searches, both for the purposes of assessing the patentability of their own processes and for freedom to operate.

274    Further, Dr de Kretsers evidence is supported by Dr Clarkes evidence regarding the paper presented by Mr Brian Dymond and Mr Don Luke entitled ‘Beyond Conventional Flocculation’ at the High Density & Paste seminar at the Paste and Thickened Tailings conference (the PTT conference) in Pilanesberg, South Africa, on 10 and 11 May 2001 (Dymond paper). Although Dr Clarke did not attend the conference at which the Dymond paper was presented, he became aware of it when he was undertaking research for Iluka in around 2002. I will return to the Dymond paper later.

275    Further, the Paste and Thickened Tailings Guide (PTT Guide) was another source of information to which persons skilled in the art had regard at the priority date; I will return to the PTT Guide later. However, the PTT Guide was one of many available sources of information and was not a complete set of the information available in the field at the priority date. Product technical literature and brochures prepared by flocculant manufacturers were also an important source of information.

276    Further, another source of information which was widely recognised in the water treatment industry was the Nalco Water Handbook. The first edition was published in 1979 and the second edition was published in 1988. Now flocculant suppliers were actively involved in both the water treatment and mineral tailings industries and personnel such as Mr Schroeter and Mr Scammell had experience in both industries. Mr Schroeter said that the Nalco Water Handbook was given to new sales staff when they commenced working at Nalco to use as a reference source. Mr Schroeter said that the Nalco Water Handbook was a widely recognised text within the water treatment industry.

277    Further, the sources of information referred to by a person skilled in the art also included the Minerals Engineering International (MEI) website, which was an aggregation source of information providing summaries of papers and recent advances in technology across all of mineral processing. The MEI website was free to access and an easy place to get a quick overview of work that was done in the field. The MEI website included information concerning mining industry news and developments and recent industry specific publications. Further, The MEI website also included a “new patents” section. Historical pages from that section of the MEI website include extracts of patents relating to the treatment of mineral tailings, including a Ciba patent for a mineral solids separation process in the name of Mr Don Luke, one of the co-presenters of the Dymond paper. Other patents extracted on the MEI website include “apparatus for treating fine ore”, “process for reducing the quantity of water contained in pulps of nickel-bearing oxide ores” and an application filed by Nalco for “Rheology modification of settled solids in mineral processing”.

278    Further, the sources of information referred to by persons skilled in the art included magazines published by industry bodies, promotional information and research updates published by industry bodies and research organisations, and commercial subscription based publications.

279    Let me now turn to questions of construction.

CONSTRUCTION OF THE CLAIMS

280    In MLA (No 1) I set out the relevant principles and it is convenient to repeat what I said at [213] to [220].

281    The principles governing the construction of patent specifications including claims are well established. A claim is construed from the perspective of a person skilled in the relevant art as to how such a person would have understood the patentee to be using the words of the claim in the context of the specification as a whole. Further, a claim is to be construed in the light of the common general knowledge including the art before the priority date.

282    A measure of common sense should be used. And ordinary words should be given their ordinary meaning unless a person skilled in the art would give them a technical meaning or the specification ascribes a special meaning.

283    In terms of how the body of the specification may be used in construing a claim, the claim should be construed in the context of the specification as a whole even if there is no apparent ambiguity in the claim. Nevertheless, it is not legitimate to narrow or expand the boundaries of the monopoly as fixed by the words of a claim by adding to these words glosses drawn from other parts of the specification. More particularly, if a claim is clear and unambiguous, to say that it is to be read in the context of the specification as a whole does not justify it being varied or made obscure by statements found in other parts of the specification.

284    Now the specification may stipulate the problem in the art before the priority date and the objects of the invention that are designed to address or ameliorate this. Accordingly, the specified objects may be useful in construing a claim in context. Nevertheless, the specified objects are not controlling in terms of construing a claim; glosses cannot be drawn from the objects.

285    A claim should be given a purposive construction (Product Management Group Pty Ltd v Blue Gentian LLC (2015) 240 FCR 85 at [39] per Kenny and Beach JJ). Words should be read in their proper context and a too technical or narrow construction should be avoided. Further, the integers of a claim should not be considered individually and in isolation. Further, a construction according to which the invention will work is to be preferred to one in which it may not. But to give a claim a purposive construction “does not involve extending or going beyond the definition of the technical matter for which the patentee seeks protection in the claims” (Sachtler GmbH and Co KG (formerly Sachtler AG) v RE Miller Pty Ltd (2005) 221 ALR 373; [2005] FCA 788 at [42] per Bennett J). To apply a purposive construction does not justify extending the patentee’s monopoly to the ideas disclosed in the specification. I also adopt what was said in Artcraft Urban Group Pty Ltd v Streetworx Pty Ltd (2016) 245 FCR 485 at [72] to [78] per Greenwood J (agreed to by Rares J at [142], [145] and [146]). Further, I would also refer to Lord Hoffmann’s observations in Kirin-Amgen Inc v Hoechst Marion Roussel Ltd (2004) 64 IPR 444; [2004] UKHL 46 at [27] to [34] concerning a purposive approach to construction.

286    As I have said, a claim is to be construed from the perspective of how a person skilled in the art would have understood the patentee to be using the words, informed by the notional skilled addressee’s relevant general knowledge and what has been disclosed in the specification. But to consider such a perspective does not entail that the Court necessarily requires expert evidence to assist on construction. If it is clear that the claims are to be read according to their ordinary meaning with no special meaning given to any word or phrase, if the science or technical issues are easily comprehensible and if, more generally, the Court does not require expert assistance in understanding the context of the claims, then expert evidence on construction may not only be unnecessary, but unhelpful and distracting. The nature and complexity of the patent in suit and the issues raised will determine the utility or necessity for expert evidence on construction. In the present case, I have to some extent been assisted on questions of construction by the expert evidence adduced by the parties, but the significance and weight of such evidence should not be over-stated. After all, the proper construction of a claim is ultimately a question of law for me, albeit that I must adopt the perspective that I have just described.

287    In terms of the skilled addressee, one is using a hypothetical construct. The following principles are applicable:

(a)    First, to identify the characteristics of the skilled addressee, the field to which the invention relates must be identified.

(b)    Second, the skilled addressee is taken to be a person of ordinary skill (as opposed to a leading expert) in that field and equipped with the relevant common general knowledge including the art before the priority date.

(c)    Third, the qualifications and experience of the skilled addressee will depend on the particular case, having regard to the nature of the invention and the relevant industry. Formal qualifications are not essential. Practical skill and experience in the field may suffice. A patent specification is addressed to those having a practical interest in the subject matter of the invention; such persons are those with practical knowledge and experience of the kind of work in which the invention is intended to be used.

(d)    Fourth, the hypothetical person skilled in the art may possess an amalgam of attributes drawn from a team of persons whose combined skills, even if disparate, would normally be employed in interpreting and carrying into effect instructions such as those contained in the specification.

(e)    Fifth, as the skilled addressee comes to a reading of the specification with the common general knowledge of persons skilled in the relevant art, they read it knowing that its purpose is to describe and demarcate an invention. But the person skilled in the art is not particularly imaginative or inventive.

(f)    Sixth, the skilled addressee does not come to reading the specification seeking failure.

288    As I have said, the legal construct may not be a single person but may be a team of persons whose combined skills would normally be employed in that art in interpreting and carrying into effect instructions such as those contained in the relevant instrument.

289    In the appeals before me the language of the claims which falls to be determined in the present case is as follows:

(a)    rigidify, “rigidification” and improving rigidification;

(b)    an effective rigidifying amount; and

(c)    co-disposal.

290    Otherwise, the meanings of the following terms and expressions in the opposed applications do not appear to be in doubt.

291    First, tailings” or “tails” refers to waste material produced during the process of extracting minerals from mined material. They are the end product of a mineral processing operation.

292    Second, the reference to bimodal distribution of particle sizes refers to tailings having a particle size distribution with two characteristic populous sizes.

293    Third, the reference to bimodal distribution of particle sizes comprising a fine fraction and a coarse fraction refers to the material having two characteristic populous sizes of particles, one being fine and one being coarse, in which the:

(a)    coarse particles have a size distribution peak (most common size) of greater than 75 microns,

(b)    fine particles have a size distribution peak (most common size) of less than 25 microns.

294    Fourth, a “coarse solid” is a particle greater than 75 microns and up to 10,000 microns. A fine solid is a particle smaller than 25 microns.

295    Fifth, the expression “during transfer” refers to the transfer of thickener underflow between the thickener and the deposition area.

296    Sixth, as I have already touched on, the expression “intrinsic viscosity” refers to a physical property of an aqueous flocculant solution which can be used to quantify the molecular weight of the dissolved flocculant species (size/length of the flocculant chain). The higher the molecular weight, the higher the intrinsic viscosity.

297    Seventh, the reference to wet or dry coarse particles being added to the underflow refers to coarse particles being added to a thickener underflow stream as:

(a)    a slurry comprised of coarse particles and water;

(b)    de-watered solids, such as the discharge from a vibrating dewatering screen; or

(c)    dry material;

and then mixed with the thickener underflow by some means appropriate to the specific application.

298    Eighth, the term “homogenous mixture refers to a mixture where coarse and fine particles are evenly dispersed with no appreciable segregation. Relatedly, the reference to co-disposal of coarse and fine solids as a homogenous mixture refers to a co-disposal process by which the coarse and fine particles are disposed together such that no appreciable segregation is evident in the finally deposited material, although I would note that the opposed applications do not exclude the possibility of some segregation on deposition.

299    Ninth, the term “heaped geometry refers to a beach with a relatively steep beach angle, with the material having a sufficient slope such that the rate of drainage from the deposited material prevents the build-up of free water over the top of the deposited material.

300    Let me now discuss in more detail the construction of the relevant terms and expressions which were the subject of more detailed focus before me.

(a)    Rigidification and improving rigidification

301    Now “rigidify” and “rigidification” are ordinary English words. “Rigidify” means “[t]o become rigid, set, or inflexible”; and “rigid” means, in relation to material, “stiff; not pliant or flexible; firm; hard” (Oxford English Dictionary (2nd ed, 1989)). The noun “rigidification” has a cognate meaning. No one doubted that in terms of the construction of the relevant claims these meanings should be applied.

302    Now I am prepared to accept that:

(a)    the underflow from a conventional thickener was material that had been rigidified;

(b)    the discharge from a paste thickener was material that had been rigidified;

(c)    the material which is discharged from a belt press filter is rigidified;

(d)    a specific measure of rigidification is the yield strength of the material which could be quantified and measured; and

(e)    stacking angle is an indicium of rigidification, with steeper stacking angles being consistent with extra structure having been introduced into that material.

303    I am further prepared to accept that persons skilled in the art:

(a)    were familiar with the concept of rigidifying tailings material that was subsequently deposited in a deposition area;

(b)    were familiar with using flocculant to treat tailings material to rigidify that material in thickeners, belt presses and centrifuges;

(c)    knew there were different degrees of rigidification;

(d)    were accustomed to examining the rigidification of material that had been treated with flocculant, and forming a view as to whether, after treatment, the rigidification of the material had improved; and

(e)    could conduct routine tests to determine if the material that had been treated with flocculant had achieved an improvement in rigidification and could improve the rigidification of the material being treated if desired.

304    In terms of “rigidification” I largely accept the evidence of Dr de Kretser.

305    The meaning of the term “rigidification” refers to the process of changing an initially fluid mixture of solid particles and liquid into one where the solid particles are held within a networked deposit with some level of mechanical strength by the following processes:

(a)    First, flocculation at some point prior to deposition, such that large and strong flocs are formed. Depending on the system, operating conditions and flocculant dosages, these flocs could range from centimetres in size up to an interconnected mass of order of the size of the outlet pipe.

(b)    Second, at the point of deposition, the permeability of the slurry at its transfer solids concentration is significantly increased as a result of the well-developed, almost granular floc structure. This means that the flocs settle/separate rapidly from the slurry stream as it slows on exit from the discharge point.

(c)    Third, the rapid settling/separation leads to release of a significant amount of free water and a corresponding solids concentration increase. As a result of both the solids concentration increase and the strength of the flocs themselves, the yield stress of the material increases to the point where the solids cease moving and form a well developed beach, with a relatively steep beach angle. Due to the over-flocculated structure, inter-floc water can almost freely drain down the beach governed by the increased permeability. But initially the over-flocculated material will have an open, porous, and relatively low-density, compressible structure.

(d)    Fourth, due to the rapid deposition of the flocs from slurry, material accumulates more rapidly over previously deposited solids. Thus compressive dewatering of the initially low-density flocs is promoted resulting in expression of water from the beach with a further increase in the solids concentration of deposited solids. The compressive dewatering would typically be at an enhanced rate due to the enhanced permeability.

306    The process of “rigidification” involves exploitation of increased permeability via settling/separation, drainage and consolidation rate to more rapidly reach a solids concentration exhibiting a desired level of mechanical strength. Additionally, the high flocculant dosage results in increased inter-particle forces meaning this strength is achieved at a lower solids concentration, requiring less water removal for its development.

307    I also agree that successful implementation of rigidification would result in the following behaviour after discharge of slurry into a deposition area:

(a)    Rapid attainment of a solids concentration at which the mechanical strength of the inter-particle network results in accumulation of a deposit with solid-like properties. Such a deposit has a strength sufficient to resist re-mobilisation by subsequently deposited material.

(b)    This solids concentration would be typically achieved through rapid settling and separation of clarified water immediately after deposition.

(c)    Formation of a relatively steep beach of deposited solids as a result of the processes in (a).

(d)    Retention of coarse particles present in the initial slurry within the flocculated fine particle matrix such that segregation within the deposited material is prevented.

308    Let me now turn to the more difficult question of the construction of the expression “improving rigidification”. The claims of the opposed applications are directed to a process of “improving rigidification”. But it is apparent from the opposed applications and the evidence that improving rigidification is a qualitative and relative concept.

309    The description of the process of “improving rigidification” appears in the body of the specifications of the opposed applications. What is taught is that:

(a)    the treated material is to stand and rigidify and therefore forming a stack of rigidified material;

(b)    the formation of stacks has the advantage that less area is required for disposal;

(c)    the rheological characteristics of the material are important since once the material is allowed to stand it is important that flow is minimised and that solidification of the material proceeds rapidly;

(d)    the rigidified material must be sufficiently strong to remain intact and withstand the weight of subsequent layers of rigidified material;

(e)    the process will preferably achieve a heaped disposal geometry and will co-immobilise the fine and course fractions of the solids and will allow released water to have a higher driving force to separate it from the material by virtue of hydraulic gravity drainage;

(f)    the heaped geometry will give a higher downward compaction pressure on underlying solids which seems to be responsible for enhancing the rate of dewatering;

(g)    as a result, the process will achieve a higher volume of waste per surface area;

(h)    it is not possible to achieve the objectives by adapting the flocculation step in the thickener and that it is essential to treat the material that has been formed as an underflow in the thickener; and

(i)    it is important that the liquor released as part of this process is clear and substantially free of contaminants, especially migrating particulate fines.

310    Further, to achieve improved rigidification, the opposed applications teach that the material must be treated with “[a] suitable and effective rigidifying amount of the water-soluble polymer solution”, at an effective point. The examples in the opposed applications then provide slump test results for a range of doses of various polymers in various forms on a variety of slurries as follows: mineral sands – tables 3 and 18; lateritic nickel, acid leach process – table 9; red mud – table 11; gold processing – table 13; lead/zinc – table 15; and coal – table 17.

311    The examples in the opposed applications are important in giving meaning to the phrase “improving rigidification” and how it is to be achieved.

312    By way of illustration, Example 12 in the opposed applications teaches that in a plant evaluation at a mineral sands process, where dosages of 100g/t of dry solids were used at a dosing point 20 metres (11 seconds) prior to discharge, the process achieved: (a) stacking angles of 8 – 10 degrees; (b) “clean water release”; and (c) “a retention of the fine material within the heap disposal”. Figure 6 then provides a visual illustration of material treated with the claimed process. This plant evaluation was undertaken following the laboratory results reported in table 19, which records different slump results achieved at different mixing times and different dose rates.

313    Similarly, table 20 in the opposed applications teaches, when read in the context of the opposed applications in their entirety but particularly page 15 of the opposed applications, that the process seeks to achieve “a significantly high-yield-stress material… so that when it discharges, it stands and does not move very far from the discharge point”. For example, table 20 teaches that a dose rate of 160 g/t and 10 seconds mixing time increased the yield stress of the material from 65 to 356 Pa. This teaches that yield stress increases significantly and that the treated material is to have the type of yield stress that would come from paste thickener and have all the beneficial properties of a paste-thickened material.

314    So in summary, the specifications of the opposed applications including the examples, tables and figures give meaning to the phrase “improving rigidification”.

315    Now Dr Farrow identified six qualitative characteristics of improved rigidification. The treated tailings would:

(a)    be less likely to spread laterally after deposition, enabling more efficient land use;

(b)    more rapidly form a solid structure in the form of a beach or stack;

(c)    have a greater yield stress when deposited;

(d)    have an increased uniformity or homogeneity of fine and coarse particles;

(e)    have a heaped geometry which would result in downward compression forces in the deposited material, forcing water out of the stack; and

(f)    have more rapid and improved clarity of water release in the disposal area.

316    In his evidence Dr Farrow accepted that the comparator using these indicia was thickener underflow which had not been treated with a second dose of flocculant (untreated tailings). That is, the “improvement” in rigidification required by the claims is to be assessed by comparing the rigidity of the deposited material, which has been treated with a second dose of flocculant, with the rigidity of material which has only been treated with flocculant in the thickener. I agree.

317    But as a matter of plain language, it follows that any improvement on the results achieved by depositing untreated tailings, even if small, would constitute improving rigidification.

318    Now Dr Farrow accepted in cross-examination that feed material which has been flocculated downstream of a conventional thickener and placed on a belt press filter is material that exhibits “improved rigidification”.

319    In his evidence, Dr Farrow explained his understanding of the process of “improving rigidification” as described and claimed in the opposed applications to the effect that:

(a)    it is a qualitative term not defined in the opposed applications in a quantitative way;

(b)    rigidification is caused in the pipe through the addition of flocculant to the tailings in the pipe;

(c)    when the material is discharged, all the material will stand at the discharge point forming a beach;

(d)    depending on the yield stress, the deposited material will only be pushed forward when extra deposited material provides enough force to exceed the yield strength of the deposited material;

(e)    there will be some flow of the deposited tailings depending on the level of yield stress in the material, but an important feature of the claimed invention is that the flow of material is minimised;

(f)    the extent to which the material flows on deposition will depend upon the amount of flocculant that is added and where it is added;

(g)    as the water leaves the deposited tailings, the solids concentration will increase;

(h)    there will be a rapid rise in the yield strength as the water is released and the material will be further rigidified or solidified; and

(i)    the steeper beach angles formed by the deposited material will increase the volume of tailings which can be stored in a given surface area.

320    Now Dr de Kretser’s understanding of the qualitative indicators of improved rigidification was largely consistent with that of Dr Farrow. In this respect he identified the following qualitative indicators of “improved rigidification”:

(a)    the tailings have been strongly flocculated such that large flocs of solids are present after deposition; he said that the large flocs possess an interconnected network between the particles such that the material has an integral strength which can withstand application of an external force (a yield stress);

(b)    the tailings have greatly enhanced permeability, reducing the time taken to dewater to the point where a beach develops;

(c)    the tailings achieve sufficient yield stress to promote beach development at a lower solids concentration than untreated tailings;

(d)    the tailings exhibit improved water release and recovery after deposition when compared with untreated tailings;

(e)    the tailings occupy a smaller surface area than untreated tailings; and

(f)    the tailings are more quickly rehabilitated than untreated tailings.

321    Dr Farrow agreed that improved rigidification results in the rapid formation of a solid structure in the form of a beach or stack. And Dr de Kretser accepted that the speed of creation of a stable deposit was relevant in determining whether there was an improvement in rigidification, but the actual time taken to achieve a stable deposit depended on the type of material that was being treated. For him, for certain types of material such as phosphate tailings, obtaining a stable deposit within weeks would be considered to be rapid. I am not sure that I completely agree with that characterisation.

322    Let me turn to the aspect of network structure and chemical bonding.

323    As I have indicated earlier in my reasons, flocculants cause solids in a slurry to aggregate into flocs that under certain conditions will start to stick together to form a structure that is permeable and allows for further dewatering. This causes an increase in the yield stress of the material.

324    Once the flocculant is added, the solid particles in the slurry attach to the flocculant forming aggregates, and the liquid between these growing solid aggregates carries them to the deposition area. The treated material is then deposited in the deposition area.

325    When particles or aggregates (flocs) are in close physical proximity there is an opportunity for flocculant bridging between the particles or aggregates. If there is exposed flocculant on the outside of an aggregate, there is an opportunity for flocs to then attach to neighbouring aggregates. This mechanism provides an opportunity for additional bonding to take place.

326    Now there was a dispute between the experts concerned the precise nature of the underlying chemical-physical nature of the deposited material that arises when improved rigidification is achieved. More particularly, the dispute between the Dr de Kretser and Dr Farrow was whether there were molecular bonds between individual flocs when improved rigidification is achieved.

327    Dr Farrow’s evidence was that in order to achieve “improved rigidification” a sufficiently high solids concentration was required to enable particles to be in sufficiently close proximity so that when a sufficient dose of flocculant was added those particles joined together and formed a single unified network structure. Dr Farrow’s evidence was that it would be feasible to achieve improved rigidification if the solids content was in a range of 15 to 80%.

328    His evidence in relation to the nature of the structure or bond created when improved rigidification is achieved was to the following effect. A flocculant induces aggregation of the solid mineral particles by the flocculant molecules co-adsorbing on two or more solid particles and binding them together through a molecular bond. Adsorption refers to the adhesion of the flocculants onto the surface of the solid particles in the tailings. This can be either physical or chemical and the specific nature of chemical bonding can vary. And in addition to the molecular bonds between the particles, there are also molecular bonds between individual flocs when improved rigidification is achieved.

329    Contrastingly, Dr de Kretser did not accept that when improved rigidification is achieved, there are molecular bonds formed between individual flocs. Dr de Kretser’s evidence was that the process of the claimed invention does not require that the treated material as deposited is fully interconnected in a chemically bonded sense. Dr de Kretser’s evidence was that after a sufficient dose of flocculant has been added to thickener underflow an interconnected network of large flocs will form. Those large flocs will consist of particles that have been molecularly bonded together to form the larger floc. But the large flocs may not themselves be bonded to other flocs.

330    Let me elaborate a little further on networking.

331    Neither of the opposed applications explicitly refers to “networking”. Nevertheless, the concept of networking is important in understanding how improved rigidification, as taught and claimed in the opposed applications, is achieved. Both Dr Farrow and Dr de Kretser agreed that networking was required to improve rigidification. But at trial, it became apparent that the extent of their disagreement was limited to the degree of networking required and the precise bonding mechanism for and explaining the networking.

332    The evidence of Dr Farrow was that the process claimed in the opposed applications requires the flocculant “to bridge all the particles together to form a network, not just form a multitude of individual aggregates”. Dr Farrow explained that this is a “dynamic process where small “sub networks” will initially form and then in the presence of the remaining flocculant these smaller “sub-networks” will progressively join to form an overall network”. The overall network is the result of “the flocculant forming molecular bonds between all of the particles i.e. the structure does not consist of a collection of discrete aggregates”.

333    Dr Farrow explained that this networked structure “has very different properties and characteristics, including in relation to dewatering, compared to a conventional settled deposit of a dry stacked deposit” and is the reason why “a rigidified deposit formed by the process described in the opposed applications has the characteristics” of improved rigidification, namely:

(a)    it is less likely to spread laterally after deposition enabling more efficient land use;

(b)    it will more rapidly form a solid structure in the form of a beach or stack;

(c)    it will have a greater yield stress when deposited;

(d)    it will have an increased uniformity or homogeneity of fine and coarse particles;

(e)    the heaped geometry results in downward compression forces in the deposited material forcing water out of the stack; and

(f)    more rapid and improved clarity of water released in the disposal area.

334    The evidence of Dr de Kretser on improved rigidification accorded in many material respects with the evidence of Dr Farrow. Dr de Kretser gave evidence that improved rigidification as claimed in the opposed applications would be achieved where the tailings had been “strongly flocculated such that large flocs of solids are present” and that the “large flocs possess an interconnected network between the particles such that the material has an integral strength which can withstand application of an external force (a yield stress)”.

335    Dr de Kretser gave evidence that “rigidification” requires that “extremely large and strong flocs are formed” and that, depending on the system, “these flocs could range from centimetres in size up to an interconnected mass of order of the size of the outlet pipe”. Dr de Kretser further explained that “improved rigidification” requires the building of “a more extensive, stronger flocculated network”.

336    As I say, there was really only one substantive point of disagreement, namely, the extent and nature of the networking required. Whereas Dr Farrow considered that in order for improved rigidification to be achieved the deposited material needed to form part of an overall network created by molecular bonds, Dr de Kretser asserted that improved rigidification could be achieved in a tailings slurry consisting of a series of discrete aggregates each constituting its own molecularly bonded network, which became “networked” as a result of an increasing solids concentration.

337    In his affidavit, Dr de Kretser set out a schematic illustrating “the range of solids concentrations and inter-particle structures”. The schematic, which I have set out earlier, is again set out for convenience:

338    Focusing on the teachings of the opposed applications, Dr de Kretser gave the following evidence when cross-examined, in respect of the top right stage of the schematic:

Mr Caine, in opening at transcript 85, line 16, suggested that those dotted lines were molecular bonds; you would accept that - - -?---They – they could - - -

- - - in the process described in the patent that’s what’s contemplated?---That is what is contemplated.

339    But he later clarified in the context of improved rigidification:

HIS HONOUR: Can I just ask you, the reference in 176 to inter-particle forces; are you referring there to the network of flocs that you had in your earlier diagram?---Yes. I mean - - -

Yes?---In the context of talking about a high flocculant dosage resulting in increased inter-particle forces, I am talking about the molecular bonds between them.

The polymer and the - - -?---Yes. And – so the rigidity of the flocs themselves.

Yes. Thank you. Rather than the diagram that you’ve got of a network of flocs it might - - -?---Yes. Yes.

Have physical forces that keep them together?---Because – well, I guess, ultimately, irrespective of the dotted lines, which I’ve drawn in that diagram which we could say are not necessarily molecular bonds, the underlying – as material deposits into a sediment, the overall strength of that sediment is largely governed by the molecular bonds within the flocs that formed that sediment.

340    Further, the evidence of Dr de Kretser was that for “extremely large flocculated masses”, the second stage of dewatering is “effectively instantaneous”, and that “time is important” when assessing improved rigidification and the opposed applications are concerned with rapid beaching and rapid separation of clarified water.

341    Nevertheless, Dr de Kretser maintained that there was no requirement for an overall networked structure to achieve improved rigidification because “you will see that my evidence in relation to the patented process suggests that it – that they’re exemplars of the patented process in which discrete aggregates are being deposited and in no way can be inferred as an extensive completely network[ed] material where all of the particles are connected together”.

342    In other words, the evidence of Dr de Kretser was that an overall networked structure was not required to improve rigidification because this was not present in some of the exemplars.

343    Further, when Dr de Kretser was cross-examined on his evidence concerning networking, he ultimately agreed that what was required was the formation of “large flocs” from smaller flocs:

What you’re suggesting, if I’ve understood it correctly – and let me replay it to see that I’m not misrepresenting what you’re saying. In the process of adding a significant dose of flocculant close to the discharge point in the tailings pipe, you have a process where particles are bonded with the molecular bonds to form a floc. Some of those flocs will form together to form larger flocs?---Yes.

What you’re saying is that, ultimately, there might be a number of the large agglomerated flocs, but they won’t necessarily themselves be bonded?---Between them, yes.

Yes?---Yes.

344    In my view, there is little substantive difference that I need concern myself with between the evidence of Dr Farrow and Dr de Kretser on networking. Both agree that flocculants work by creating molecular bonds between particles and both agree that to achieve improved rigidification, sufficient flocculant must be added so that “small flocs” or “discrete aggregates” or “sub networks” are in some sense “bonded” together to form “large flocs” or an “overall network”. So much was subsequently affirmed by Dr de Kretser during his cross-examination:

So you have a sufficient dose of flocculant so that large flocs of solids are present, an[d] the large flocs can either be a whole pile of particulate matter all bound together or groups of those smaller aggregates that have been bonded together to form a large floc?---Yes.

And the effect of that is that ... those large flocs possess an interconnected network between the particles. And when you say an interconnected network between the particles, that encompasses, also, the smaller aggregates that might be within that large floc?---Yes. It encompasses the physical interactions between flocs as well as any chemical interactions.

Yes. And there might be both?---Yes.

Yes. And such that the large flocs has an integral strength which can withstand an application of yield stress?---Yes.

And the consequence is that when you have these large flocs discharged you get what you’ve described in paragraph 177?---Yes.

345    Further, Dr de Kretser affirmed that the process claimed in the opposed applications “will generally develop a strong network and, potentially, a more extensive network” and that when he was talking about an “extensive network” this included “individual aggregates of particles themselves flocculated in a network being joined with other aggregates to form a larger floc”. Dr de Kretser also clarified that he did not wish to draw any distinction between a network of particulates bound by flocculants and a network between flocs:

MR SHAVIN: Now, can I break that down a little bit: there’s no dispute between you and Dr Farrow that in the context that we’ve discussed there are molecular bonds in the network?---Within a floc, yes.

Yes. And there’s no distinction – there’s no disagreement, is there, that when one is looking at what comes out of the thickener, we’re looking at the relative size, structure and extent of the network?---When you’re saying – comparing that to the patented process?

Yes?---The patented process, because of the conditions under which polymer is added and the generally higher dosage required to obtain a flocculated network will generally develop a stronger network and, potentially, a more expensive network depending largely on the conditions within the deposition area and the share conditions.

Yes. And by - - -?---But in all cases – they are molecularly bonded networks in both cases – in thickener underflow, and secondary flocculated thickener underflow.

Yes. And when you talk about an extended network, this includes but is not limited to individual aggregates of particles themselves flocculated in a network being joined with other aggregates to form a larger floc?---Yes.

Yes.

HIS HONOUR: Although, I just want to be clear about that. Your last sentence says “within a floc, a molecularly bonded network”, and counsel is now putting to you a different type of network, which is a network of flocs. Not a network of particulates bound by flocs. So there seem to be two concepts of network here. So in relation to not a network of particulates bound by flocculant, which produces a floc – but let’s talk about a network between flocs. What’s your evidence?---I prefer to, again, make no distinctions between a network formed of flocs of flocs, and simply just consider that it’s the extent of the network because, as I’ve said, in feed to a thickener, it is completely – it – the feedwell of a thickener is also designed to create flocs of flocs, so I cannot make a distinction between – on that basis between thickener underflow and secondary treated .....

There is one distinction, though, isn’t there? The thickener is normally physically proximate to the processing plant?---Yes.

The discharge point – the TSF I think it has been referred to by - - -?---Yes.

- - - in the literature, is often not physically proximate to the thickener?---That’s correct.

346    Moreover, Dr de Kretser accepted that the process claimed in the opposed applications would require flocs that are “stronger and potentially more extensive in size”, that a “consequence” of the process would be “large flocs”, and that flocs would “be allowed to grow to a higher size, larger size”. Dr de Kretser explained, in this respect, that “if a sufficiently high dosage of flocculant is added, a gelatinous blob…might just come straight out of the pipe and deposit with no movement”. That, of course, accords with the teaching in the opposed applications that “once the [deposited] material is allowed to stand it is important the flow is minimised”.

347    What is the effect of all this evidence? In my view, it establishes that whether the treated material is characterised as forming part of an “overall network” (as Dr Farrow prefers) or “large” and “strong” flocs which are themselves networks of smaller flocs (as Dr de Kretser prefers), the end result is the same.

348    Now SNF submits that the precise chemical and/or physical structure of the treated material deposited is not relevant to the construction of “improved rigidification” as such. Ultimately the way in which the structure of the deposited material is formed is largely irrelevant. What is important is that the deposited material has the requisite strength and permeability to exhibit improved rigidification on deposition. I tend to agree. In my view the person skilled in the art would not understand the terms “rigidification” or “improved rigidification” by reference to the underlying physical or chemical structure of the deposited material, but rather by reference to its qualitative characteristics which are visually discernible. Persons skilled in the art are unlikely to care about the details of the type of bond, just that each flocculant molecule adsorbs strongly two or more particles and binds them together. The assessment of the rigidified nature of deposited material is made through visual observation of the nature of the deposited material rather than any consideration of the underlying chemical or physical forces present in the material. In my view, persons skilled in the art were concerned with the functional outcome of the deposition strategy such as the formation of a beach which promotes drainage, stability and an increased rate of rehabilitation. Indeed, the person skilled in the art had no ready way of determining whether deposited material consisted of physical or molecular bonds within the flocs or between the flocs.

349    Let me elaborate further on the functionality of material standing until it is pushed forward by other material. Dr Farrow’s evidence was that a further feature of improved rigidification is that the material stands until it is pushed forward by other material. But Dr Farrow also accepted that there must be some flow of the deposited tailings. The deposited tailings must spread out in order to make efficient use of the impoundment space. But in any event the underlying mechanism for the formation of the deposited material, including the precise manner in which the material spreads, is not directly relevant to the assessment of the validity of the claimed invention. The assessment of the extent and nature of rigidification is to be made by reference to relevant qualitative indicators.

350    Let me also deal with the concept of improved rigidification as distinct from settling/sedimentation.

351    Dr Farrow sought to distinguish some prior art relied on by SNF on the basis that it disclosed “settling” or “sedimentation” as opposed to “improved rigidification” because:

(a)    the prior art used the word “settle”;

(b)    the process involved sub aqueous deposition; and/or

(c)    the process involved the use of containment walls or engineered structures.

352    But SNF says that Dr Farrow adopted a varying and inconsistent approach to his assessment of the prior art, which allowed for prior art to be conveniently and semantically distinguished. For example, Dr Farrow sought to distinguish the Backer & Busch 1981 paper by referring to disclosures of “immediate separation of water and solids” and “the flocculated solids readily settled and relatively clear water was liberated” as “exactly what would be expected from a flocculation/settling process”. But SNF says that the observations in the Baker & Busch 1981 paper are precisely those identified in the opposed applications as being the desired result of rigidification. Similarly, SNF says that Dr Farrow distinguished prior art as not disclosing rigidification merely because of references to the “settling” of the flocs or use of the word “sedimentation”. But SNF says that this is notwithstanding that the 785 application on page 16 lines 20 to 26 refers to the claimed invention involving the transfer of treated material to a “settling area” where the material is allowed to dewater to release liquor. Further, SNF says that Dr Farrow attempted to distinguish prior art on the basis that the material is deposited into a dam or other containment area and therefore cannot have achieved “improved rigidification”.

353    Now in my view there is a conceptual distinction between “improved rigidification” and the concepts of “settling” or “sedimentation” and I will come back later to discuss the s 7(3) prior art. But I do agree with SNF that “settling” or “sedimentation” concepts are not foreign to and may be part of the process(es) of the claimed invention.

354    First, the 785 application itself provides for the deposition of material into a “tailings dam or a lagoon”. The final paragraph of page 16 of the 785 application states:

In this form of the invention the aqueous polymer solution is applied to the material in a similar manner as described above. In this case, the polymer solution is applied in an effective dewatering amount and in the same way as a first aspect of the invention it is important that the fluidity of the material is retained during transfer. The material is transferred to a settling area, which can for instance be a tailings dam or a lagoon.

(Emphasis added.)

355    Second, in example 13 on page 38 of the 785 application it is noted that in this working of the claimed invention, the material was discharged into a series of pits that were filled sequentially.

356    Third, in evidence before me was a brochure produced by Ciba which refers to Ciba’s Rheomax ETD program. Rheomax ETD is the trade name for Ciba’s flocculants used in relation to the claimed invention and other applications. Pages 4 and 5 show a working of the claimed invention where the tailings are deposited into impoundment cells.

357    In my view, there is no reason why rigidification of material cannot also occur if the material is deposited into a dam, pond or lagoon. Under such circumstances, the deposited tailings could first settle with development of a sedimentary delta, but would eventually accumulate and form a beach above the water level. The material in such a beach would be subject to hydraulic gravity drainage.

358    Finally on the topics of “rigidification” and “improving rigidification”, let me deal with some other matters.

359    First, it is important to note that the opposed applications do not claim to have invented “rigidification” of tailings. The claims are concerned with “[a] process of improving rigidification of a material...”. Thus, the opposed applications are teaching and claiming “a process” which improves upon rigidification previously known.

360    Second, the meanings of rigidification and “improving rigidification” were debated at length in the 2008 proceedings before Kenny J who held that they were not terms of art (SNF (Australia) Pty Ltd v Ciba Speciality Chemicals Water Treatments Ltd (2011) 92 IPR 46 at [47]). Her Honour had regard to the competing evidence of the experts in the 2008 proceedings and ultimately adopted the evidence of BASF’s expert, Dr Farrow, as to meaning (at [69] to [75]). Her Honour then said at [267], which was upheld on appeal (SNF (Australia) Pty Ltd v Ciba Speciality Chemicals Water Treatments Ltd (2012) 204 FCR 325):

I accept Dr Farrow’s evidence that “rigidification” is a qualitative term, although one that Patent 944 explains clearly as “a networked structure”. Further, for the reasons already stated, I conclude that Dr Farrow’s evidence supports the Ciba respondents’ basic contention that, read within the context of Patent 944 (and the other patents in suit), compared with settling and sedimentation processes, rigidification is faster; produces more recovered water; and results in chemically-bonded tailings that occupy a smaller surface area, which are more quickly rehabilitated. I also accept that, as Dr Farrow said, rigidified tailings material would be less likely to spread laterally after deposition, enabling more efficient land use; and would more rapidly form a solid structure in the form of a beach or stack; and have a greater yield stress when deposited, with increased uniformity or homogeneity of fine and coarse particles. Further, by reason of its heaped geometry as a beach or stack, such rigidified material would result in downward compression forces driving water out of the stack and more rapid release of water, with better clarity.

361    Mr David Shavin QC for BASF submits that I should adopt the same construction in these appeals by simply following her Honour’s conclusions. Now it has been held that as a matter of judicial comity an earlier decision on the construction of a patent should usually be followed (see Neurizon Pty Ltd v Jupiters Ltd (2004) 62 IPR 569 at [32] and [35] per Kiefel J). But in the present case of course I am not dealing with the innovation patents dealt with by Kenny J and so have discussed and decided the matter for myself.

(b)    Effective rigidifying amount

362    The opposed applications refer to suitable doses ranging from 10 to 10,000 g/t of material solids, and preferred doses in the range of 30 to 3,000 g/t, with more preferred doses ranging from 60 to 200 or 400 g/t.

363    The opposed applications disclose that SDITB had been used to treat thickener underflow to improve the compaction of the fine waste material and clarity of the recovered water applying flocculants at “conventional doses”, but that this had produced little benefit.

364    It is well apparent that the opposed applications disclose that when flocculant had been added at conventional doses, the following problems described in the opposed applications had not been solved:

(a)    the coarse and fine particles were not being deposited together in a homogenous deposit;

(b)    consequently, the coarse material settled much faster than the fine material, causing banding or segregation; and

(c)    the run off water contained high proportions of fine particles that contaminated the recovered water.

365    There is little doubt that as at the priority date, a person skilled in the art would understand what conventional doses of flocculant would be required for specific purposes and would know that conventional dosages would vary depending on the type of tailings and how the tailings were to be treated.

(c)    Co-disposal

366    As I have already said, tailings are often produced in two size fractions: a coarse fraction and a fine fraction. The coarse tailings are made up of sand sized particles. Fine tailings or slimes are generally made up of silt or clay-size particles. For example, in mineral sands mining, two streams of waste material are typically produced, one being predominantly fine particles and one being predominantly coarse particles.

367    Now the coarse and fine particles in the tailings may have originally been part of the same tailings stream but may have been recombined after a prior upstream separation step. Alternatively there may have been tailings streams generated from different parts of the process employed at a mine. Alternatively, the coarse particles may have been sourced from elsewhere on site such as sources of waste rock or off site such as sand, and then added to the fine tailings stream.

368    It is not in doubt and both of the experts accept that co-disposal, as that term was used in the opposed applications, was the deliberate combining of a separate stream of coarse material to the thickener underflow.

369    In my view, the use of the term “co-disposal” in the opposed applications would be understood by a person skilled in the art at the priority date as referring to a process whereby there is a step of combining coarse and fine process streams to create a combined co-disposal stream.

370    I should also say for completeness at this point that a number of the claims of the opposed applications contain specific requirements as to when in the process of improving rigidification the co-disposal step is to occur. For example, claim 17 of the 785 application explicitly requires that the “wet or dry coarse particles are added to the underflow…before or during the addition of an effective rigidifying amount of the water soluble polymer”. Claim 1 of the 568 application explicitly requires that:

(a)    aqueous suspensions of fine and coarse particulates be combined for the purpose of co-disposal to form a material;

(b)    the aqueous suspensions be mixed into a homogenous slurry; and

(c)    during or after mixing of the aqueous suspensions, an effective rigidifying amount of an aqueous solution of a water-soluble polymer be combined.

371    The opposed applications also teach that the effective rigidifying amount of the water-soluble polymer solution will normally be added during or after the mixing of the different waste streams into a homogenous slurry.

372    I need not say anything further at the moment on this aspect.

COMMON GENERAL KNOWLEDGE

373    I have already dealt with matters of common general knowledge that are essentially not in dispute earlier in my reasons. Let me now elaborate and address some additional matters and also turn to more contentious areas.

(a)    Tailings beaching

374    First, let me deal with yield strength in the context of tailings beaching. In the context of tailings beaching the following was well known at the priority date. Mine operators could produce thickener underflow with a very high yield strength which would assist in tailings beaching. But often thickener underflow had to be pumped long distances to a deposition area. However, shear forces in the pipeline would cause the yield strength of the slurry to break down as it was pumped to the deposition area. The effect of shear degradation during transit was that the deposited material would have reduced yield strength (viscosity) and permeability and compressive strength. This would adversely affect the ability to obtain a stable deposit and clear water run-off. Now a person skilled in the art knew that flocculation in the thickener could be used to increase the yield stress of the material being deposited. But if the thickener underflow had a high yield strength, this would put strain on pumps and cost more in terms of power than a less viscous underflow. Additionally, if the yield strength of the tailings within the thickener was too high, the rakes in the thickener could be damaged. Therefore, there was a trade-off between obtaining a higher yield strength underflow to assist in tailings beaching, and having a less viscous underflow which was easier and less expensive to pump.

375    Second, let me deal with the beaching/stack angle in this context. In relation to tailings beaching, the following was also well known at the priority date. If the underflow was discharged in a deposition area, a beached structure would be formed depending on the nature of the material being discharged. And a sloped beach in the deposition area promoted the run off of water from the tailings into a decant pond for recycling. Further, it was desirable for the solids in the deposited tailings to rigidify and stop moving as soon as possible after leaving the outlet pipe so as to avoid contamination of the released water with fines, and to allow the water to run away to a collection point for recycling. Further, it was desirable to achieve an increase in the stack angle of the deposited tailings so that the land available for disposal could be used more effectively by storing a greater volume of tailings in the deposition area. Further, it was desirable for deposited tailings to have as high a solids concentration as practicable at the deposition point because this would promote steeper beach angles due to its higher viscosity. Further, the stacked material needed to have sufficient strength i.e. be sufficiently rigid to withstand further deposits of tailings material on top of it, and to permit the continued build-up of the stacks. But this did not mean that the deposited tailings had to stop moving immediately on deposition. If the material built up too quickly at the point of deposition it could impede the flow of further material from the outlet pipe. This would not result in the most efficient utilisation of the space in the deposition area. Further, if the stacking angle was too steep, not all of the deposition area would be utilised. Accordingly, the optimum stacking angle was a function of the deposition area to be filled and could be determined by simple geometry. Further, it was desirable to produce tailings in which the coarse and fine particles formed a homogenous deposit rather than to segregate on deposition. A homogenous deposit would have a greater and more uniform strength than a deposit with a non-uniform spread of particle sizes.

(b)    Use of belt press filters and centrifuges

376    Let me now turn to the question of the use of belt press filters and centrifuges. The evidence is that belt press filters and centrifuges have been widely used in Australia since the 1970s in conjunction with conventional thickeners to dewater thickener underflow. Belt press filters started to become widely used in the mineral processing industry in the early 1980s and by 1994 it was not new technology in the field. Indeed, many witnesses in this proceeding had experience with the use of belt press filters and centrifuges.

377    Mr Bellwood gave evidence that belt press filters and thickeners were the primary focus of persons skilled in the art who were working on solid / liquid separation in respect of tailings prior to the priority date.

378    Mr Scammell had significant experience with secondary dosing of flocculant in the use of belt press filters and centrifuges before the priority date and before he began his trials of secondary dosing at the Yarraman mine. Mr Scammell recalled working with belt press filters at Mount Thorley and Catherine Hill Bay before the priority date. Mount Thorley mine had four belt presses and at least two centrifuges. He also saw belt press filters in operation at another mine in Singleton/Cessnock, New South Wales where there were two belt presses in operation.

379    Mr Bembrick had experience with adding flocculant in solution to the feeds of belt press filters before he undertook the trials at Ardlethan. By October 2002, he was for example aware of secondary dosing being performed at the Tahmoor Colliery in New South Wales, which was operating two belt presses and at Port Kembla’s steelworks using a centrifuge. Additionally, he worked with Mr McColl on a belt press filter at Port Kembla’s steelworks.

380    Mr Schroeter was aware of belt press filters being used in coal mines in the Hunter Valley before the priority date, including observing one in operation at the Dartmoor Coal Washery in 2002. Further, Mr Schroeter was aware of the use of belt press filters to treat thickener underflow with a second dose of flocculant given that the sale of flocculant for use in belt press filters was one of SNF’s key target markets in the mineral processing industry before the priority date.

381    Mr Coleman saw belt press filters in operation in at least four mines before the priority date, including the Westcliff and Tahmoor collieries.

382    Mr Holtzman used secondary dosing in trials of belt press filters at Cable Sands’ Jangardup mine in the very late 1990s.

383    Further, the widespread use of belt press filters is consistent with the disclosure in one of BASFs predecessor’s papers published in 1992 authored by G Moody, “The Use of Polyacrylamides in Mineral Processing” (1992) 5(3-5) Minerals Engineering 479-492 (Moody paper). The Moody paper records (at 480 and 481):

Pressure Belt Filtration…started to become widely used within Mineral Processing approximately 10 years ago [ie, 1982]. A similar type of process had been used for some time for the treatment of sewage and in the paper industry. However, due to the highly particulate nature of mineral processing slurries belt pressing had to await the advent of high molecular weight synthetic flocculants in order to allow it to be used effectively in this area. It is necessary to produce a highly flocculated slurry which dewaters very rapidly to ensure little free water is present as the partially dewatered structure enters the compression zone. The flocs should then collapse to a certain extent to allow further dewatering to take place, but they must be strong enough to prevent penetration of solids into and through the belt.

384    The Moody paper also notes (at 484) that centrifuges have been used as equally widely as belt press filters since the early 1980s within the mineral processing industry. The paper records:

Centrifugation

As with pressure belt filtration it is only in fairly recent years (10 to 15 years) that centrifugation has become widely used within the Mineral Processing Industry. Again, this type of process had been used extensively for sewage treatment before that.

A diagrammatic representation of a counter - current centrifuge is illustrated in Figure 7. As the feed enters the centrifuge very high shear forces are encountered and therefore in order to take advantage of flocculation, the floes produced must be very shear stable. Centrifugation provides a useful illustration of how varying levels of cross-linking can be used to improve polymer performance.

385    Let me now re-iterate how belt press filters work. The thickener underflow is treated with a second dose of flocculant added into the pipeline from the thickener to the belt press filter in sufficient doses to achieve a highly flocculated structure in the underflow such that the material rapidly dewaters when deposited onto the belt in the belt press filter. Dr Farrow accepted that a person skilled in the art was accustomed to adding a second dose of flocculant to thickener underflow to pre-treat it before feeding it into a belt press filter. Indeed he was unable to recall any example where thickener underflow was transported or piped to a belt press filter without a second dose of flocculant being added.

386    The solids content of the slurry feed to the belt press was in the range of 20% to 50%. The flocculant that was then added was in solution. Indeed, Dr Farrow accepted that the flocculant was usually added in the form of an aqueous solution. And as a result of the second dose of flocculant being added, the underflow had a “cottage cheese” like structure and appearance prior to being pressed between the belts in the belt press filter.

387    Dr Farrow accepted that it was important to have a highly flocculated slurry with a degree of structural integrity as the slurry was fed into the compression zone of the belt press filter. This was necessary so that the solids could be squeezed between the belts, rather than being displaced by them. And as the Moody paper also notes, in the application of belt press filters:

It is necessary to produce a highly flocculated slurry which dewaters very rapidly to ensure little free water is present as the partially dewatered structure enters the compression zone. The flocs should then collapse to a certain extent to allow further dewatering to take place, but they must be strong enough to prevent penetration of solids into and through the belt.

388    Let me re-iterate how centrifuges work. Centrifuges operate by feeding the slurry into a rotating vessel. Centrifuges are similar to thickeners except that centripetal forces, rather than gravity, are used to separate the liquids and solids in the tailings. Use of centrifuges similarly involved the addition of a second dose of flocculant to the tailings in the feed line.

389    Now at this stage it is convenient to elaborate further on dose point and dose amount. Dr de Kretsers evidence is that a person skilled in the art, who was familiar with the use of flocculants to pre-condition tailings which are to be treated using a belt press filter or centrifuge, would be aware that the floc structure is obtained very quickly. This necessarily means that the flocculant needs to be added with a short mixing time, and therefore shortly prior to the deposition of the treated material onto the belt or into the centrifuge. Dr Farrow also accepted that a person skilled in the art was accustomed to adding a second dose of flocculant to thickener underflow to pre-treat it before feeding it into a belt press filter and that the flocculant needed to be added between the thickener and the belt press filter to achieve well flocculated material. To do so, such a person would make a judgment based on their skill and experience about the most effective point at which to add the flocculant so as to allow it to adequately mix with the thickened slurry to create a highly flocculated material.

390    The Nalco Water Handbook (first published in 1979 and republished in 1988) included a schematic diagram of a belt press filter showing flocculant being added to the feedline to the belt press.

391    Mr Bembrick’s evidence was that he was aware from his work for Ciba in relation to belt press filters “that if you secondary dose too far away from the belt press (in-line) the floc will just shear and not dewater”.

392    Mr Scammell gave evidence in relation to the use of belt press filters at Mt Thorley. He advised the operator of that mine in relation to the method of addition of the second dose of flocculant to underflow which had been thickened in a thickener increasing its solids concentration. The purpose of the second dose of flocculant was to improve the performance of the belt press filters. He also advised the operator of that mine in relation to the selection of the flocculant to be added to the thickener underflow before it was deposited onto the belt. The flocculant was added as an aqueous solution. A first dose of anionic flocculant was added into the outlet pipe several metres before the tailings were deposited onto the belt, and a second dose of cationic flocculant was added either into the feedwell, or just before the feedwell, of the belt press. In optimising the performance of the belt press, Mr Scammell adjusted the operational parameters of the flocculant dosing in order to produce a material with the required mechanical strength and rate of drainage on deposition onto the belt. As part of the optimisation process, he reviewed the mixing of the flocculant into the thickener underflow, the form and nature of the material that was produced, and how that material drained on deposition on the belt. He then modified the process so that the anionic flocculant used was the same flocculant (Magnafloc 156) that was used in the thickener at the mine. In both cases, Magnafloc 156 was added as an aqueous solution. He adjusted the mixing of the second dose of flocculant to ensure it was adequately mixed with the tailings so that the tailings had the necessary qualities when deposited onto the belt press. The point of addition of the flocculant could only be moved a small distance, but Mr Scammell was able to create more mixing in the pipeline after the second dose of flocculant was added by using an impeller. He was aware that the underflow would shear in the pipe and that this shear would reduce the yield stress of the material. At Mount Thorley, the impeller was ultimately removed because it was producing too much shear. But he was able to optimise the performance of the belt press despite only being able to move the dose point for the flocculant addition a small distance.

393    Further, the Mount Thorley mine also used centrifuges to treat thickener underflow. Mr Scammell said that he added flocculant to the thickener underflow before it entered the centrifuge so that more water would be released from the tailings in the centrifuge, and the solid material would be sufficiently dry to be discharged onto a conveyor belt.

394    In my view the above evidence establishes that secondary dosing in aqueous solution shortly prior to deposition onto the belt of a belt press filter or entry into a centrifuge in sufficient doses to obtain a highly flocculated material on deposition was widely practised in Australia before the priority date and formed part of common general knowledge.

395    Finally, let me address the question of improved rigidification in the belt press filter feed. Dr Farrow’s evidence concerning the material produced by secondary dosing of belt press filter feed was to the following effect. It is a highly flocculated slurry which dewaters very rapidly. It has a degree of structural integrity that prepares it for mechanical compression. The feed material “certainly [shows] improved rigidification”. If sufficient flocculant has been added, the feed material will be networked in the same way as material that exhibits “improved rigidification” within the meaning of the claims of the opposed applications. In his view there would have been improved rigidification in the belt press filter feed and it would have had that structure when it was deposited onto the belt.

396    In summary, the evidence makes clear that persons skilled in the art familiar with the operation of belt press filters and centrifuges well knew the following:

(a)    First, flocculant could be added in aqueous solution to thickener underflow in the outlet pipe shortly prior to a belt press filter or centrifuge to produce a highly flocculated slurry which on deposition would:

(i)    have a degree of structural integrity;

(ii)    be likely to exhibit “improved rigidification” within the meaning of the claims of the opposed applications.

(b)    Second, to achieve such flocculation was a matter of establishing a suitable dose and dose point taking into account the need for adequate mixing and the degrading influence of shear.

(c)    Third, the optimum point of addition of the flocculant to achieving such flocculation was one which enabled sufficient time for the mixing of the flocculant into the thickener underflow but insufficient time for the yield stress created by the flocculant to be degraded by shear thinning.

(d)    Fourth, the optimum point of addition to achieving such flocculation was usually a short time prior to the end of the outlet pipe, prior to the thickener underflow being deposited onto the belt or into the centrifuge.

(e)    Fifth to assess whether such flocculation had been achieved, the flocculated tailings could be visually observed, and if the result was not as required the operator could add more flocculant or change the dose point.

397    But it must be emphasised that this secondary dosing is all upstream of the mechanical dewatering device and close to the thickener. Moreover, the vast majority of the dewatering is done close to the plant rather than at the deposition point.

(c)    Other secondary flocculant dosing

398    In my view, the use of a secondary dose of flocculant prior to deposition of a slurry into a tailings disposal area was unusual outside the scenario of secondary dosing upstream of a mechanical dewatering device.

399    The evidence of Dr Farrow was that he had not directly encountered any operation using in-line flocculation for tailings disposal, co-disposal or not, at the priority date. Further, despite his role in managing the P266D Project for CSIRO, which was an industry wide project focused on enhancing tailings disposal, not a single one of the numerous industry participants had ever raised with him the possibility of adding flocculant to the underflow line of a thickener to improve dewatering of tailings in the disposal area.

400    Further, the evidence of Dr Clarke, who had a doctorate in mineral processing, and who had “learned a lot about existing practices and developments in the Field [mine tailings disposal] in the period from approximately 1974 to 7 May 2003…through visiting mine sites and viewing the practices conducted”, was that he had never before his visit to Yarraman in 2002 to observe the trials conducted by Ciba and CRL observed a process involving secondary dosing of flocculant close to the discharge point at any mine site.

401    Now SNF’s solicitors provided to Dr Clarke a draft affidavit that they had apparently drafted for him, which asserted:

I knew before visiting Yarraman and had known for years before that visit that one way of addressing such problems was to add another dose of flocculant to the combined tailings during transfer from the concentrator and thickener to the deposition area, so that the viscosity and bonded structure of the waste materials in the underflow was re-established before deposition. This was commonly referred to as secondary dosing or viscosity/rheology modification.

402    But upon reviewing the draft affidavit, Dr Clarke deleted that paragraph. He informed SNF’s solicitors that: “I am not sure about this. I am not even sure it is correct. What I was aware of was “mud farming” and also the installation of high compression thickeners, but neither of these is really relevant”. When cross-examined, Dr Clarke confirmed that the secondary dosing process trialled at Yarraman and subsequently Yoganup was “a new process that had not been applied at full scale”.

403    Further, the evidence of Ms Herzig, who had by 2002 been working as a qualified metallurgist for 10 years, was that she did not have any personal experience of secondary dosing of thickener underflow before it was trialled at Yarraman.

404    Further, the evidence of Mr Holtzman was that he commenced work in the mining industry in 1969 and started work at Cable Sands in 1992, but he did not encounter secondary dosing at all until the late 1990s and even then only in the context of dosing thickener underflow which was to be fed to a belt press filter. Further, his evidence was that, in 2002, “[w]e didn’t seriously consider it as part of our trials, mostly from an economic point of view”.

405    Further, the evidence of Mr Schroeter, the managing director of SNF, more supports BASF’s case. In his affidavit, he stated that people in the field knew at the priority date “that it was advantageous to add flocculant at more than one point in the tailings disposal process”. But in his oral evidence he said that in his personal experience he had “only added [flocculant] upstream of belt presses and high-speed centrifuges”, and that as at the priority date people reporting to him within SNF only had experience with secondary dosing upstream of some form of mechanical dewatering device.

406    Further, the evidence of Mr Scammell, a metallurgist previously employed by Ciba, who had worked in the mining industry since 1986, was that the first time he tested a secondary dosing process using flocculant in solution was at the Yarraman mine in 2002. He had only previously used a secondary dose of flocculant in aqueous solution in the context of a belt press where the objective was both “speed of drainage and the ability to maintain structure under compression”, whilst ensuring that the material did not “squeeze out the sides”.

407    Further, the evidence of Ms Beveridge, a senior technical specialist employed by BASF, who had been employed by Allied Colloids (a corporate predecessor to BASF) since 1989, was that she had “never previously trialled the administration of a rheology modifier dissolved as a solution in a tailings line” until she became involved in Project Zenith.

408    Further, the evidence of Mr Bellwood, who had been employed by Allied Colloids since 1982, and whose role from 1997 to 2016 involved managing global research projects for Ciba in mining technology, was that:

[B]efore the Priority Date, the only in-line secondary flocculation processes aimed at achieving stacking of the tailings that I was aware of, were those being developed by Ciba. I was not aware of other in-line secondary flocculation being used in dry stacking applications. Nor had I read about this in any literature before the Priority Date.

409    Mr Bellwood also explained that “prior to August 2002, certainly within Australia, we were only testing polymer in a powder”.

410    Now the affidavit evidence of Mr Bembrick, a former employee of both Ciba and SNF, was that by the time he began to undertake trials with Ciba (apparently around 1999) he was “very familiar with the process of secondary flocculation and how to achieve effective stacking of tailings”. But there was little foundation for this evidence. Under cross-examination he gave the following evidence:

So in the context where we are looking at a process where there is a thickener and the underflow of the thickener is discharged to a tailing dam, are you able to identify five mines with which you are familiar in October 2002 that involved the addition of a secondary dose of flocculant at or very close to the point to discharge into the tailings dam?---That I was involved with?

That you knew of?---I didn’t know of any.

411    Further, when cross-examined about his knowledge of secondary dosing and slump angles in 2003, Mr Bembrick gave evidence that: “I would say that back in 2003, my experience was quite low at that time. I would have said that I would – I wouldn’t know too much back then.”

412    The above evidence is also consistent with the following further material.

413    First, reference should be made to the state of the art as recorded in the 2002 version of the PTT Guide. Dr Farrow considered the guide to reflect “the state of knowledge and focus of efforts in the area at the time” and Dr de Kretser acknowledged that he and other persons in the field “regularly” referred to it as at the priority date. The guide was published in conjunction with the PTT conference which Dr de Kretser identified as the leading conference in the field at the priority date. That guide made no reference to secondary dosing and did not do so until well after the priority date. Indeed, Dr de Kretser said that “the focus of that conference series at that time” was on variations of thickener processes. Let me elaborate.

414    The 2002 PTT Guide in evidence before me made no reference in terms to in-line or secondary flocculation.

415    Section 4.3.2.1 on beach geometry stated the following:

When tailings are deposited as a slurry, they flow like a liquid from the point of discharge towards the lowest point of the impoundment. This flow is typically channelised and the meandering channels move back and forth on the beach as the deposit develops. If the deposition water content is high, as with conventional tailings disposal, segregation of the coarser and finer particles will occur, with the coarser particles settling out near the point of deposition and the finest particles being transported all the way to the tailings pond, finally settling out under water. In this gravity driven process, the amount of segregation that occurs and the distribution of particle sizes that occur along the resulting ‘beach’ are a function of both the deposition water content, the particle size distribution, and the mineralogy of the tailings. The beach profile that results under these conditions usually has pronounced upwards concavity, as described by Blight et al (1985).

If a tailings stream is sufficiently thickened, segregation of particles will be extremely limited (essentially will not occur under ideal circumstances) and even very high-density slurries (and possibly some pastes) will flow as a viscous liquid. The flow is again dominated by channelisation but there is the potential for much more uniform profiles and, under ideal conditions, even a relatively uniform slope. The overall beach slopes will be a function of many parameters including, of course, the solids content and the consistency and viscosity of the thickened product. Based upon industry experience to date, even with high-density slurries the slopes tend to be relatively flat (1 to 2%) in most cases though there can be exceptions. It is expected that tailings thickened to a paste like consistency would form a slightly steeper beach. Due to processes such as drainage, consolidation, thickener efficiency fluctuations, tailings mineralogy changes (thickener feed variation), weather extremes etc., P&TT deposits also tend to demonstrate some degree of concavity in most cases. This concavity is not as pronounced as for conventional slurries and, as noted, it is conceptually possible to have a uniform slope under ideal conditions (though it is doubtful such conditions could be sustained over an entire mine life).

Preliminary assessments of likely beach slopes are often obtained from laboratory flume testing. There is however, no accepted method as yet for predicting this beach angle from laboratory tests and it is probably advisable to also carry out pilot-scale field tests. Similarly, there is no guarantee that even large-scale field trials can build in all of the variability of the actual operation (let alone the very real scale influences of deposit geometry).

It is expected that as research and development in this field gain pace, improved predictive techniques will emerge. However, it is unlikely that any method will develop that will allow precise beach geometry predictions that includes all potential process upset (breakdowns or periods of sub-optimal performance) conditions and geological variability in the thickener feed.

416    Section 5.8.2 only referred to improving the efficiency of thickeners in the following terms:

Over the last 10-15 years there has been an effort by thickener manufacturers and industry to improve the efficiency of thickeners and develop something better than the “conventional thickener”. The two pioneers in this field were EIMCO Process Equipment Company (Baker Process) and Enviroclear. Their work resulted in thickeners that could produce more tonnage per unit area and had a faster settling rate than previous thickeners. This was accomplished by the smart use of flocculants using special feedwell systems and controls. The “smart use of flocculants involved understanding under what condition flocculants work best. This was accomplished by understanding the flux and concentration relationship. Figure 5.20 shows this relationship graphically and demonstrates that for any given feed material there is a feed concentration at which the flocculant behaves at an optimum. This is determined by laboratory, bench or pilot tests. Once this relationship was understood, special deep feedwells were designed with and without bottom plates and special dilution features to reach that optimum flux concentration zone. Once this was achieved then the high rate thickener was born and thickener size reduced or throughput increased. A large percentage of currently produced thickeners are of this type. Two of the most widely [used] dilution systems are the EIMCO Process E-DUC® feed dilution system … and the Supaflo AutodilTM system …

417    Second, reference should be made to the “Tailings Disposal” memorandum written by Mr Cigulev in 1999 when he was employed by Iluka. In that memorandum, Mr Cigulev identified project aims of reducing mining area, increasing beach angle, obtaining consolidation as quickly as possible, providing a substrata that is amendable to rehabilitation, and achieving all this cost effectively. Under the heading “State of Current Research”, Mr Cigulev reported: “The only option that we currently have that satisfies the aims above is to use belt presses for the dewatering of the clay to a truckable paste”. Whilst Mr Cigulev did address secondary dosing, he reported that it “can only be used in low slime orebodies” and all of the options examined involved some form of mechanical dewatering (such as “a vibrating dewatering screen”) or slimes dams limited to recovering “decant water”, although there was reference to testing this method on a pilot scale.

418    Third, the state of the art recorded in 2004 in the Australian Coal Association Research Program (ACARP) research proposal reported in relation to a proposed two year research program into secondary dosing that: “the proposed research program is unique and its work methods and strategies have been developed with the aim of collecting base data which will act as a foundation for future assessments”. The 2004 ACARP research proposal recorded that the applicant organisations were Tamplin Resources Pty Ltd and Xstrata Coal Pty Ltd. But the second project leader was noted as being Mr Bembrick of SNF. SNF was a supplier of flocculant to the Bulga mine owned and operated by Xstrata. The funds requested from ACARP were $91,900. The total project cost was estimated to be $695,900.

419    The Executive Summary set out:

This study proposes to conduct a series of washery trials using a variety of linear and structured flocculants in order to assess their relative impact on tailings bed level density, water retention and the propensity for tailings to form surface crusts suitable for subsequent rehabilitation. The study will have a duration of 2 years and will be conducted wholly at the Bulga Mine which is owned and operated by Xstrata Coal Australia. The project has a total expenditure of $696,000, of which ACARP are requested to contribute $91,900.

Investigating the degree in which various types of flocculants affect tailings density and water retention is crucial in optimizing tailings disposal strategies which aim to maximise tailings pond capacity while minimizing overall water losses within the impoundment area. If successful, the studies findings will be applicable to the majority of Coal Mines in Australia as they will demonstrate how flocculant selection can impact on tailings water retention, the formation and depth of turbid zones within tailings dams and the delay between the cessation of tailings disposal and the subsequent rehabilitation of the dam.

All works associated with the proposal will be managed under the Bulga Coal BSafe Management System. As such, all work will be conducted by inducted and appointed personnel as per normal operating practices at Bulga Coal. A detailed risk assessment and work procedure will be developed in conjunction with Bulga personnel following the projects approval.

420    The project objectives were stated to be:

The objective of the proposed study is to assess the impact of flocculant chemistry on the tailings bed level density, water retention and the propensity for tailings to form surface crusts suitable for subsequent rehabilitation. Investigating the degree in which various types of flocculants affect tailings density and water retention is crucial to optimizing tailings disposal strategies which aim to maximise tailings pond capacity while minimizing overall water losses within the impoundment area. Tailings density has direct impacts on the volume of water retention within the dam as well as the propensity for tailings to form surface crusts suitable for subsequent rehabilitation.

The Hunter Valley has a number of old and existing tailings dams that contain tailings that have failed to form surface crusts and therefore cannot be rehabilitated. The trials proposed in this study will provide baseline data which will allow the development of disposal strategies that should increase tailings bed level density and ultimately facilitate the rehabilitation of high clay content tailings dams.

421    The state of the art was described in the following terms:

Recent advances in chemical engineering have seen the development and introduction of branched and cross linked flocculant polymers. These polymers are stronger and more robust than the old-style linear polymers and result in an improved clarification of washery water. Apart from the improved dewatering characteristics, the molecular weight of these polymers can be varied, to target a range of surface characteristics and particle size distributions. These developments have facilitated the ability to manufacture specific flocculants to suit the varied requirements of different coal· operations.

Flocculant engineering has traditionally focused on optimizing the speed and effectiveness of solid liquid separation. Historically little consideration has been given to the impacts of flocculant selection on the settlement characteristics, capacity impacts and subsequent rehabilitation potential of the resultant tailings deposited within impoundment structures. More recently, these concerns have become more important due to the increased environmental and regulatory constraints that are now applied to new and existing tailings dams. Moreover, the majority of coal operations now work in a regime where there is only limited capacity to construct new tailings dams and as such, research projects that could potentially optimize the capacity of existing structures will be beneficial to the industry.

Literature reviews have failed to identify any similar projects to the proposal that are directly related to the Australian Coal Industry. In the context, the proposed research program is unique and its work methods and strategies have been developed with the aim of collecting base data which will act as a foundation for future assessments.

Advances in flocculant injection techniques and mechanical dewatering methods (such as belt presses) have recently proved successful in thickening tailings. These techniques have the advantage of minimising water retention and improving the propensity of the tailings to form surface crusts, however, they add considerably to the capital and operating expenses of CHPP’s. Consequently, the old technology of using traditional thickener systems will continue to be applied for the next 10 to 20 years as existing plants are operated to their full life cycle capacity. In this context, the outcomes of the proposed study will be relevant to the majority of existing Coal Mines for the foreseeable future.

422    The project was anticipated to take 2 years.

423    Clearly, secondary flocculation not in the context of a mechanical de-watering device was to be investigated. This was a research project. And as the proposal stated, “the proposed research project is unique”.

424    Contrastingly to the foregoing evidence, Dr de Kretser deposed that: “In the case of treating underflow from a thickener, flocculant was commonly used to restore a freely dewatering structure in the tailings which had been broken down [due] to shear during transport of the tailings to the TSF” (emphasis added). Dr de Kretser then gave detailed evidence regarding elements of in-line flocculation which he asserted constituted common general knowledge at the priority date.

425    But Dr de Kretser did not have any personal knowledge of the use of in-line flocculation in Australia at the priority date. And he accepted under cross-examination that:

(a)    he had never trialled in-line flocculation before the priority date;

(b)    he had never published any articles on the topic of in-line flocculation;

(c)    the only instance of in-line flocculation in Australia of which he personally was aware before the priority date was at the Beenup mine, and even then he had not visited the Beenup mine or observed the use of in-line flocculation at that mine (I would note that the Beenup mine shut down because it could not deal with its tailings); and

(d)    the first time that he undertook any field work or research concerning secondary flocculation was in 2011, years after the priority date.

426    Dr de Kretser accepted that he could not view his solitary example of secondary dosing at the Beenup mine as constituting common use. Moreover, he gave the following evidence that hardly assisted SNF’s case relying upon his opinion on this aspect of the case:

So apart from the failed example of Beenup, you can’t actually identify a single instance in Australia before the priority date of which you had knowledge before the priority date of the use of a second dose of flocculation in-line in the tailings stream, can you?---That’s – that’s correct if you’re taking out Beenup…

When you say in paragraph 60:

…in the case of treating underflow from a thickener, flocculant was commonly used to restore a freely dewatering structure in the tailings which had been broken down to shear during the transport of the tailings to the TSF –

That statement is simply wrong, isn’t it, doctor, at the priority date in Australia?---If you’re talking specifically in Australia?

Yes?---Yes.

It wasn’t commonly used in Australia before the priority date, was it?---As I said, it was still something that was part of what was considered as a – as in the broader suite of tailings deposition or management strategies.

427    Further, when Dr de Kretser was questioned on why CRL and Iluka needed to undertake research and development projects into secondary flocculation in 2002/03 if it was common general knowledge, Dr de Kretser said: “they are specific applications where further work was required to understand if it was possible” (emphasis added). But as BASF submits, having perhaps appreciated the potential implications of such a statement, Dr de Kretser sought to step back from it, before then saying that “[i]t was a body of experience that was evolving”.

428    Further, Dr de Kretser deposed that “based on [his] experience before the [p]riority [d]ate”, the “typical flocculant dosages ranges” used in “in-line flocculation” “[r]ange[d] from 150 to 1000g/t” and involved “addition into pipeline close to or at [the] discharge point”. But that evidence also lacked an adequate foundation. It would seem that the basis for this assertion by Dr de Kretser was that he had read the Backer & Busch papers before the priority date. But I agree with BASF that the fact that a dose rate falling within this range might have been disclosed in the Backer & Busch papers says nothing about whether such a dose rate was common general knowledge or typically used in Australia. Let me elaborate.

429    First, Dr de Kretser accepted that he did not know whether anybody in Australia had read the Backer & Busch papers at the priority date.

430    Second, Dr de Kretser under cross-examination said:

You don’t know anyone who had used 150 grams a tonne before or after the priority date in the flocculation – or the secondary flocculation of tailings in a tailings pipe, do you?---I can’t say whether I do or don’t.

431    Third, the evidence given by Dr de Kretser concerning secondary dosing is not supported by his own PhD thesis. In his thesis introduction, he identified as a problem with the state of the art in 1999 that “conventional de-watering technology is either inefficient at de-watering the tailings or can only perform de-watering after substantial capital expenditure”. When he proceeded to survey the different available “methods and technology involved in tailings de-watering” he discussed the use of “conventional clarifier[s] or thickener[s]”, tailings dams in which material could “settle and eventually form a dry consolidated waste”, the “mechanical de-watering for processing their fine tailings streams”, and various prior art publications. But no methods described in his thesis included secondary dosing practices used in the field without a mechanical dewatering step. This is made apparent by section 1.3 and other parts of his thesis.

432    I have referred to Dr de Krester’s thesis not for the significance of what it says but what it does not say.

433    First, it makes no express reference to secondary flocculation in the absence of some form of mechanical dewatering device.

434    Second, it makes reference to the Backer and Busch publications, referenced as Backer and Busch, 1981 and “Stewart et al, 1986”, which SNF has relied upon as s 7(3) prior art, but draws little of the conclusions that SNF now says that a person skilled in the art would have drawn.

435    Third, his thesis generally makes no reference to co-disposal of the type dealt with in the 785 application and the 568 application. I will return to this later.

436    In summary, I agree with BASF that other than the work undertaken by Ciba and the trial work undertaken by Nalco in relation to the OreBind process, the only relevant examples of secondary dosing identified in the evidence related to the failed process at Beenup, the process implemented at Londonderry in the early 1980s, and a single confidential trial undertaken by Iluka at Yoganup. But such evidence falls short of establishing that at the priority date secondary dosing of a slurry prior to deposition into a tailings disposal area was known in the industry outside the scenario of secondary dosing upstream of a mechanical dewatering device.

437    But I am prepared to accept that it was common general knowledge that secondary dosing could have possible advantages, whether in the context of a mechanical dewatering device or outside that context.

438    So, for example, I note that secondary dosing was described by Xu, Y and Cymerman, G (1999) “Flocculation of fine oil sand tails”, Polymers in Mineral Processing, in Proceedings 3rd UBC-McGill International Symposium on Fundamentals of Mineral Processing, J.S. Laskowski (ed), Metallurgical Society of CIM, August 22-26 1999, Quebec, Canada, pp. 591-604 at 599 in the following terms:

Two-Step Dosing

Flocculation can be improved by splitting the polymer dose into two or more streams injected into the suspension at different points. This is because the capture of fine particles is directly related to mixing. However, during extensive mixing some flocs already formed will be broken, resulting in smaller floc sizes. The broken floc fragments will not be able to re-flocculate in single-step dosing, the mixing to capture most particles can be achieved in the first-stage addition and large floc sizes may be achieved at the second stage, where the broken floc fragments can be re-flocculated. Therefore, with two-stage addition, higher settling rate and clearer supernatant are expected.

(d)    Adjustment of variables

439    In my view it was common general knowledge that variables could be adjusted when undertaking a tailings beaching process and using belt presses and centrifuges in order to manipulate the characteristics of the tailings on deposition in the particular mining application including variables such as:

(a)    the flocculant to be used;

(b)    the form in which the flocculant was added;

(c)    the dilution of the flocculant;

(d)    the dose of the flocculant; and

(e)    the dose point/s.

(e)    Secondary Dosing in Tailings Beaching (SDITB)

440    SNF accepts that based on the evidence in this proceeding it has not been established that SDITB formed part of common general knowledge in Australia.

(f)    Co-disposal

441    Co-disposal was practised before the priority date in an attempt to overcome the difficulties in obtaining a stable deposit of tailings which consisted of predominantly fine particles. In an attempt to overcome those difficulties, coarser grained material such as sand or crushed rock (in either wet or dry form) could be added to the tailings during transfer to the deposition area or at the point of deposition. This could be done to encapsulate the fine material within the coarse matrix to provide enhanced drainage and stability of the deposit so as to produce tailings with better mechanical strength.

442    There were various possible ways of implementing co-disposal, including:

(a)    blending the coarse and fine material at the plant;

(b)    transporting the coarse material separately to a deposition point and combining it with the fine material in the thickener underflow just prior to deposition; or

(c)    intermingling the coarse and fine material at deposition.

443    Depending upon the set up of the mining operation, the coarse and fine particles may have originally been part of the same slurry stream which had been separated earlier in the process, the coarse and fine particles may have been generated from different parts of the mining process, or the coarse particles may have been sourced from elsewhere on site (such as sources of waste rock) or off site (such as sand), and then added to the fine tailings stream.

444    Co-disposal if it could be achieved had many advantages when compared with depositing fine tailings alone, which advantages were known to persons skilled in the art at the priority date. These advantages included the following:

(a)    producing tailings with increased mechanical strength on deposition and improved geochemical stability;

(b)    the rapid consolidation of the mixture to a high density;

(c)    increased permeability of deposited tailings thereby improving water release; the increase in permeability could be orders of magnitude larger;

(d)    improved clarity of the water that was recouped for recycling;

(e)    the potential to substantially enhance water recovery from the total washery waste by up to 40%;

(f)    reducing the operating costs and the time to complete rehabilitation; and

(g)    reducing the volume occupied by deposited tailings due to storage of the fines within the void spaces of the coarser particle matrix, substantially reducing the volume taken up by the two wastes if they had been separately disposed of.

445    Persons skilled in the art were aware that the benefits of co-disposal, if it could be achieved, included the fact that co-disposal would reduce the operating costs and the time to complete rehabilitation and that it was for these reasons that a lot of companies “were very interested in trying to achieve co-disposal to use Dr Farrow’s words, but with my emphasis. For the moment I will put to one side the significant challenge with co-disposal of preventing segregation and other difficulties.

446    It would seem that co-disposal was a method of tailings treatment at the priority date. Dr de Kretser’s evidence is that by the priority date there were various coal mines operating co-disposal processes in Australia, although there were problematic aspects with his evidence that I will return to. Dr Farrow also acknowledged that people were performing co-disposal before the priority date.

447    Further, other witnesses gave evidence that co-disposal was implemented prior to the priority date.

448    Mr Schroeter was aware of co-disposal being used extensively in the treatment of mineral sands and coal tailings from the early 1980s onwards and also in phosphate mining.

449    Mr Scammell gave evidence that Yarraman had been using a co-disposal process since at least 2000. The combination of a coarse stream with thickener underflow whilst being pumped to the deposition area, as was being practised at the Yarraman mine prior to Ciba’s trial work, was co-disposal.

450    Ms Herzig said that the Yarraman mine used a co-disposal process before the priority date (a process she referred to as “blended co-disposal”). She was aware in 2002 that blended co-disposal was increasingly being used in mining operations that had high fines content, as at this time many ore bodies were being mined in areas of high fines content, such as Yarraman. Moreover, she attended a lecture in Perth at which Professor John Ralston presented on co-disposal processes in red mud and mineral sands mines in Western Australia.

451    Mr Holtzman said that he had worked on the testing and development of a co-disposal process for Cable Sands in the late 1990s and early 2000s as part of his role as a metallurgical technician. Further, from the mid-1990s to the early 2000s, Cable Sands had been undertaking trials of co-disposal processes at its mineral sands mines, including operating a co-disposal process at the Jangardup mine for 12 months prior to 2002.

452    Mr Bembrick said that co-disposal was used in the treatment of coal tailings since at least the 1990s. And, he was aware of and trialled co-disposal processes at a number of mines prior to May 2003. Further, the Ardlethan tin mine was using a co-disposal process in 2002, when he whilst working at Ciba trialled the implementation of secondary dosing in conjunction with the existing co-disposal process at that mine.

453    Dr Clarke and Mr Cigulev also gave evidence to the following effect:

(a)    Iluka had been working on trials and tests to improve its co disposal processes since the late 1990s, including trialling secondary dosing with Nalco at Yoganup in 1999. Iluka had trialled co-disposal in conjunction with secondary flocculation at Yoganup in 2000.

(b)    Iluka was using co-disposal at its Eneabba mine in June 2002.

(c)    The trials of secondary dosing of co-disposed tailings being undertaken at Yarraman were of interest to Iluka for its application to other mines. Co-disposal was a process being considered by Iluka in 2002 at its mines in Western Australia and in the United States.

(d)    Iluka undertook laboratory testing to evaluate the effectiveness of SDITB of co-disposed tailings for one of its mines in Virginia in April 2003.

(e)    Dr Clarke was aware that a field trial had been undertaken by Nalco for Iluka sometime in August 2002.

(f)    The scope of works prepared by Dr Clarke in July 2003 for the trials at Yoganup records that Iluka had by that time investigated tailings disposal by non-segregating co-disposal for many years, with co-disposal being practised in its mines in most states of Australia except its south-west Australian operations.

(g)    Iluka had prior to the priority date undertaken work in relation to SDITB in co-disposal which it referred to as non-segregating co-disposal (NOSCOD) at its Concorde mineral sands mine.

454    Further, Mr Schmidt said that he was trained at Nalco that the OreBind process was suitable for treating co-disposed tailings and he promoted the OreBind process to potential clients on this basis.

455    Further, Ms Beveridge gave evidence that Ardlethan was practising co-disposal since at least as early as 2002 and that the problem the mine was experiencing was segregation of the coarse and fines. The fine material was leaving the coarse and “carrying over” on deposition. Further, the Sandalwood mine was practising co-disposal since at least as early as 2001. Further, by at least July 2002, Ciba considered that the Gallagher process would be useful to assist miners that were undertaking or proposing to undertake a process of co-disposal. Further, the Ciba trial work at the Osborne mine was at a copper/gold mine which was already undertaking a co-disposal process before Ciba trialled powder addition in November 2002 and solution addition in February 2003.

456    Further, the evidence also establishes that the following laboratory tests and field trials which Ciba undertook were designed to simulate use in a co-disposal process because that was the process the mine operator was either implementing at the time or was proposing to implement:

(a)    laboratory tests at Jangardup and Sandalwood in October 2001 and at Sandalwood in October 2002, January 2003 and February 2003;

(b)    field trials at Sandalwood in October 2002 and March 2003;

(c)    laboratory tests at Yarraman in August and September 2002;

(d)    field trials at Yarraman in October and December 2002 and March 2003;

(e)    laboratory trials at Osborne in November 2002 and February 2003;

(f)    field trials at Ardlethan in February and June 2003;

(g)    laboratory trials at Iluka’s Douglas, Eneabba and Yoganup mines in June 2003; and

(h)    field trials at Yoganup in July 2003.

457    Some of these though were after the priority date.

458    Further, co-disposal was the subject of papers which were published in journals, textbooks and presentations at conferences and also the subject of patent applications, throughout the 1990s. Dr de Kretser referred to various examples of such documents published in the period 1990 to 2003. For example, a publication by DJ Williams in 1992 (I will return to this in a moment) described co-disposal as a “promising new technique”, which records that 11 years before the priority date persons skilled in the art were contemplating co-disposal. Further, Heather Hutcheson of Cable Sands presented a paper on co-disposal at the 2001 Paste and Thickened Tailings Conference, the leading industry conference attended by persons skilled in the art at the time, in South Africa. I will return to this.

459    Further, the Condolios patent and the Ledden patent both disclose the use of SDITB adding flocculant in aqueous solution in co-disposal processes.

460    The Ledden patent is a patent filed in the name of American Cyanamid Inc which later became Cytec. At the priority date, Cytec was one of the four major flocculant manufacturers. The Ledden patent describes a process of SDITB adding flocculant in aqueous solution to phosphate tailings in a co-disposal process. In the Ledden patent the following was disclosed. Sand was added to the fines for the purposes of solidifying the material to improve its rigidity. Flocculant was added to the sand and then mixed with the slimes just prior to deposition in the land-fill area. And on deposition there was a rapid release of water (Column 5, line 34 and Column 6, line 8) and the sand and slimes compacted homogenously (Column 7 line 69 and Column 8 lines 2-3). It would seem that when using the process in the Ledden patent, rapid water release was achieved. Further, the coarse and fine particles in the deposited material were “sufficiently homogenous to settle in a compact form, which has useful load bearing characteristics” (column 4, lines 51-54).

461    The above evidence establishes that co-disposal was practised in Australia before the priority date and formed part of common general knowledge. But it would seem that achieving co-disposal without segregation of the coarse and fine particles on deposition (co-immobilisation), which segregation would result in ineffective stacking, was a challenge for the mining industry at the priority date. Let me at this point say something more about the segregation of coarse and fine materials. At the outset it seems clear that persons skilled in the art were aware that, notwithstanding co-disposal’s known benefits, one of the significant difficulties associated with co-disposal was minimising the segregation of the coarse and fine materials on deposition. Let me elaborate.

462    Dr de Kretser said that it was known that the successful operation of co-disposal processes required minimisation of segregation of coarse and fine tailings (thickener underflow) after deposition to maintain a degree of homogeneity. And the more homogenous the deposited material, the better the outcome. Further, he gave evidence that the extent of segregation could be managed via various means, including making use of flocculation, increasing the solids concentration of the thickener underflow and reducing the difference between the average particle sizes of the coarse and fine fractions.

463    Mr Schroeter said that it was known that segregation on deposition was a practical difficulty associated with co-disposal. But he said that segregation could be reduced by increasing the dosage of flocculant and segregation could be controlled by removing water from the tailings, thereby increasing the solids content of the tailings as well as the yield stress.

464    Dr Farrow gave evidence that whilst the concept and benefits of co-disposing of fine and coarse particles in order to achieve homogenous tailings was known as at the priority date, it was not routinely implemented. And where it was attempted, it was not done with good results. I must say that Dr Farrow’s evidence reflected the industry evidence.

465    Further, in relation to CRL’s Yarraman mine, Ms Herzig deposed that:

CRL’s need at the time was to be able to directly co-dispose the thickened slimes from the thickener underflow with coarse tailings (sand)… However, CRL found that the coarse fraction and the fine fraction within the slimes / tailings mixture would segregate on deposition, which was not desirable. In an attempt to fix this problem, CRL experimented with altering operational variables in the concentrator and the thickener without success.

466    When cross-examined, Ms Herzig explained that segregation of the coarse and fine tailings at Yarraman made it more difficult for CRL to rehabilitate the land, made it more difficult to reuse water from the deposition area, required extensive and expensive bulldozer work to be undertaken to build levees to prevent slimes returning to the dredge pond, and required more work to be done with machinery to manage the tailings.

467    Further and indeed, there was considerable documentary evidence recording the trials and experiments undertaken by Iluka, the parent company of CRL, after the priority date, to try and achieve effective co-disposal. These documents record that it was difficult to successfully implement a co-disposal process. For example the scope of works written by Dr Clarke dated 22 July 2003 which I have referred to elsewhere recorded that the co-disposal technique “in use at most sites” had the result that “sand and fines are disposed of together, but then tend to segregate in the impoundment. The segregation leads to increased costs and planning difficulties”. Moreover, Dr Clarke confirmed when cross-examined that this reflected his experience, at least “without the [addition] of flocculant”. The scope of works identified a number of important criteria to evaluate whether “viscosity modified blended co-disposal”, a technique intended to prevent segregation and enhance drainage could be effectively achieved by Iluka at Yoganup. Further, a technical report written by Ms Herzig dated 24 October 2003 noted that:One major concern with blended co-disposal as currently practiced, is that the fines segregate from the sand as the material flows over the beach. That segregation leads to separate sand and clay rich zones. That in turn leads to a number of problems.” The report then described the results of a co-disposal trial which sought to limit segregation. The trial was described as being “of great importance”, and concluded that “[t]he data unfortunately suggests that a considerable amount of segregation occurred in the Yoganup trial”.

468    I agree with BASF that the difficulties encountered by Iluka and CRL in successfully implementing a co-disposal process accorded with the difficulties encountered in the mining industry more generally. If co-disposal was part of common general knowledge, so too was its notorious problems and difficulties concerning segregation. So much is illustrated by the five key documents concerning co-disposal identified by Dr de Kretser in his written evidence. Let me deal with each of these documents in turn.

469    The first is a paper, DJ Williams, “Co-disposal of Coal Mine Tailings and Coarse Reject: A promising new technique” (1992) 22 Australian Geomechanics Society 50. Although it identified co-disposal as being in use at the Jeebropilly mine in the Ipswich coalfields it noted that “[T]he technique of co-disposal remains to be proven for other washery waste materials at other mine sites.” It also suggested to overcome issues at the Jeebropilly mine that “[t]he segregation of fines could be substantially reduced by pumping the mixture at a higher solids concentration and at lower velocity”. The paper concluded:

The co-disposal option, involving the combined pumping of coarse reject and tailings slurry, ideally to an elevated landform, offers the best potential technically, economically, and environmentally. It remains to carry out the research necessary to optimise this technique and ensure its successful application at a range of coal mines having different washery waste materials and conditions. Application of the technique to mining operations other than coal may also be possible.

470    It was silent on secondary or in-line flocculation.

471    The second is a presentation given in 1999 by A Vietti and McAlinden (De Beers Consolidated Mines Ltd), “Towards Co-Disposal: Progress on Southern African Diamond Mines” made at the Paste Technology for Thickened Tailings Conference, which discussed co-disposal, identified a 12 year “Towards Co-disposal research process”, noted that a challenge was “Operational Control (failure in many cases)”, and concluded that “[o]nce the technology has been fully explored, complete co-disposal is planned”. I would also note at this point that no reference was made to also adding a secondary dose of flocculant.

472    The third is a paper, PH Morris and DJ Williams, “The porosity of co-disposed coalmine wastes” (2000) 14 International Journal of Surface Mining, Reclamation and Environment 63, which included a “diagrammatic representation of a co-disposal beach” and recorded that co-disposal results in “[s]econdary beaches which comprise mostly fines that remain too soft to permit free access long after deposition has ended”. After discussing the porosity of co-disposed tailings, the paper concluded that: “The comparatively high initial [porosity] of the co-disposed wastes are attributable to the segregation of fines that occurs on the co-disposal breaches. Thus the potential of co-disposal to increase the volume of wastes that can be stored in a given impoundment can be fully realised only if this can be reduced by, for example, modifying the particle size distributions of the input wastes to reduce or eliminate gap grading… or by adopting up-slope deposition”.

473    The fourth is a presentation dated 2001 by Heather Hutchison, “Co-disposal Trials for a Mineral Sands Operation”, Conference: High Density & Paste, which described two co-disposal methods. The first method involved combining coarse and fine streams for “subaqueous disposal of fines and sands” in an underwater viewing tank. The second method involved “discrete co-disposal – separate streams” with the subsequent application of a “sand cap” onto the disposed streams. But in my view neither bore any resemblance to co-disposal as taught in the opposed applications. The second method was then tested in a trial dam, which was constructed, filled with water, filled with the separate streams, and then drained. The presentation concluded that it was necessary to “continue with test work and understanding”. The presentation also noted that: “The higher the underflow solids the better because less moisture would be added at the mixing/pumping stage and hence more [likelihood] of the mix not segregating”. Clearly, this presentation described an experiment, not a routine application of common general knowledge.

474    The fifth is a presentation dated 2003 by WL Daniels at the Sustainable Development Indicators in the Minerals Industries Conference, titled “Strategies for the return of heavy mineral sands mines to productive agricultural uses”. This presentation was concerned with “active mining and reclamation”, not co-disposal, reported that “slower than anticipated settling rates for the dewatering tailings/slimes mixtures” were achieved, and only contained a fleeting reference to the trial work being undertaken by Iluka in 2003.

475    I agree with BASF that these documents do not support the assertions of Dr de Kretser that co-disposal was at the priority date “a well-established approach routinely considered and implemented” in the field and that there was “widespread and successful implementation of co-disposal processes” in the field. To the contrary, the evidence establishes that to the extent that co-disposal was attempted at the priority date, it was attempted with difficultly and poor results.

476    Further, it is not irrelevant to point out that the method of co-disposal taught by these papers and presentations bears little resemblance to the process taught in the opposed applications. As Dr Farrow observed, “[a]ll papers suggested [that] the approach to co-disposal at their time of publication was to physically mix a flocculated fine suspension with unflocculated coarse material.” But that process may be contrasted with the process claimed in claim 1 of the 568 application.

477    But there was also other evidence led before me concerning the difficulties of co-disposal.

478    The difficulties associated with the successful implementation of co-disposal were also well illustrated by the evidence of Mr Holtzman who deposed that:

[I]n the late 1990s and early 2000s, the combining of these coarse and fines streams into a stable mass suitable for reclamation and re-use proved to be difficult with no ‘one size fits all’ process having yet been developed. To get the sands and the fines together and for them not to segregate was the biggest problem.

479    In his oral evidence, Mr Holtzman reiterated that “[i]t had been tried, and what generally happened was only a small portion of the – of the fines stayed with the tails”. It was only after lengthy and complicated testwork that Cable Sands overcame “that problem with segregation”. The testwork in which Mr Holtzman was involved commenced in the “mid to late-nineties” and yet, as pointed out by BASF:

(a)    in 2001, Cable Sands still needed to undertake further testwork to understand how co-disposal would best work;

(b)    in 2002, Cable Sands was still progressing the development of a co-disposal process; and

(c)    in October to December 2002 and March to April 2003 Cable Sands undertook what Mr Holtzman agreed to be “complicated and detailed and lengthy trial work”, at the end of which Mr Holtzman still considered that “[i]t wasn’t proved that it would – would work in a full scale trial”.

480    Further, Mr Holtzman described in some detail the testwork he undertook in relation to different co-disposal processes at Jangardup and Sandalwood. In relation to Jangardup, he agreed that Cables Sands undertook “research work” and that “It’s obviously not routine optimisation of something you’ve already got working in the field, is it?---No. No.” And in relation to Sandalwood, his evidence was that:

And as I understand it you agree with the proposition that there was complicated and detailed and lengthy trial work at dams 1 and dams 2 but your evidence is that that was work you were undertaking – or Cable Sands?---Yes. That’s - - -

And you worked in technical services department at the time, didn’t you?---Yes.

And that department was focused, I think you’ve said, on the development of new equipment or processes?---Yes.

And that’s what you’re trialling?---Yes…

The trials were the culmination of research you had been undertaking into dry mining co-disposal for a number of years?---Yes.

481    Further, the evidence of Mr Schroeter was that “[t]here were issues with segregation” and that people in the field were encountering difficulties at the discharge point where slurries containing fine and course materials were segregating. Indeed Mr Schroeter conceded that he had not previously suggested that co-disposal constituted common general knowledge in the proceedings before Kenny J, even though it was an integer of some of the claims of the innovation patents in suit before her Honour.

482    Further, the evidence of Mr Cigulev also made plain that co-disposal was not “routinely considered and implemented” or “widespread” in the field. The evidence of Mr Cigulev was unequivocal: “[i]t was not common”.

483    Further, Mr Ron Coleman could only identify a single instance of a co-disposal process which he had observed. Further, he did not suggest that he had ever implemented or been involved in working on a co-disposal process himself. And he accepted that his observation of one co-disposal process in 33 years was “hardly common”.

484    Further, the evidence of Mr Woolley was that “the typical method employed in the mining industry” for disposing of waste materials did not involve co-disposal. His knowledge of the typical method extended at least until the end of 2001, at which point he ceased working for Nalco.

485    In my view there was no widespread and successful implementation of co-disposal processes in the field. To the contrary, the evidence adduced before me revealed that persons skilled in the art working in the field before the priority date usually separated, treated and separately disposed of, the coarse and fine waste streams. For example, the evidence of Mr Cigulev regarding the tailings disposal process he oversaw at the Beenup mine was that rather than consciously trying to combine sand (coarse) to the fines stream, “in fact, we were trying to remove the sand. So there was no conscious – you know, decision to add sand”. The evidence of Mr Woolley regarding the tailings disposal process he oversaw at the Londonderry mine was that the material being treated “was fine and watery. There were clays”, after coarse material (including sand) had been removed by use of a hydrocylcone. Further, the evidence was that none of the trials of the OreBind process at Ernest Henry, Wemen or Boral Stapylton involved the use of a co-disposal process.

486    Moreover, when Dr de Kretser proposed a tailings disposal solution in his PhD thesis in 1999, he recommended a two-step process, commencing with a traditional settling process of the coarse material, and concluding with the separate flocculation of the fine material, so that “flocculation adsorption will then be specific to the fines rather than on the larger particles that can settle rapidly without flocculant”. This stands in stark contrast with the type of co-disposal process which he asserted before me to have been well known, well used and common general knowledge at the priority date.

487    It is worth setting out the following aspects of section 9.9.1 of his thesis:

As a result of the presence of the fine coal taking longer to settle, flocculant addition would still be required in practice to maintain a clear liquid overflow. However, in the literature review of Chapter 8 it was noted that the process of flocculation is one of adsorption and will thus be area dependent. With the increases in particle size due to controlling the break-up of the high surface area clay platelet aggregates there would be a huge reduction of the overall surface area open to flocculant adsorption in the suspension. The flocculant consumption required for effective flocculation of the tailings would be expected to be dramatically reduced. In effect, flocculant addition could be reduced to only adding enough to flocculate the fine clay and coal particles alone.

The observation that the majority of the solids in the controlled dispersed simulated tailings settle out rapidly without the addition of a flocculant lead to an interesting consideration. Flocculant adsorption is essentially indiscriminate in a highly heterogenous system such as coal tailings, so why waste additional flocculant on material that will settle out rapidly by itself? It would be more efficient in terms of operating costs, to have a two-stage tailings sedimentation system where the first stage involves sedimentation of the rapidly settling material in the tailings with no flocculant addition, whilst the second stage involves clarifying of the turbid supernatant from the first stage with a polymeric flocculant. More in-depth design considerations will be addressed in the Chapter 10 along with more in-depth recommendations regarding application of the research presented within this thesis.

488    He later concluded in section 10.5.2 that:

A two stage sedimentation/flocculation procedure was proposed in the discussion at the end of Chapter 9 involving; i) separation of the processes of thickening of the majority of the suspension solids and ii) clarification of the supernatant using polymeric flocculant addition. The advantages of such a system are that flocculant adsorption will then be specific to the fines rather than on the larger particles that can settle rapidly without flocculant.

It was observed that the settling rates of even the fines that were left in suspension after the majority of the solids had settled out were still rapid enough that, provided a large holding volume was available in the thickener, sedimentation without flocculant addition may be feasible. In any case, the first stage of the process would not require a high volume thickener and would most probably be best suited to a deep cone thickener type of arrangement. The second stage would require a larger volume thickener similar to conventional operations.

The underflow from the first de-watering stage would have a rather coarse size distribution and as such may be prone to sedimentation problems within the pipeline to the disposal or utilisation area. To rectify this the fine flocculated waste from the second de-watering stage could be blended in to impart a structure less likely to settle in short times within the pipeline.

489    Clearly none of this was secondary flocculation in the context of co-disposal of the type dealt with in the opposed applications.

490    Finally, to the extent that Dr de Kretser purported to identify during his re-examination a number of mines which operated co-disposal processes at the priority date, the weight of that evidence is doubtful. BASF points out that that evidence was given for the first time during re-examination. I also clarified with Dr de Kretser that he had not referred to any of these alleged instances of co-disposal in his affidavit(s), although Dr de Kretser gave evidence that some of these names were in the papers authored by Mr David Williams, which represent two of the five documents that I have referred to earlier. The mines Dr de Kretser identified in re-examination were:

Goonyella– Goonyella North, Jeebropilly – some of these are actually – in – in some of the papers by David Williams. Stratford. Tarong. Moranbah North. Kestrel. West Cliff. Coppabella. And at the priority date, Hail Creek, Clermont and Moorvale were in the process of implementing a co-disposal process. And the final one was Dartbrook, which was also at the priority date in the process of undertaking various trials of co-disposal processes.

491    The Jeebropilly mine is the case study referred to in the first document (DJ Williams, “Co-disposal of Coal Mine Tailings and Coarse Reject: A promising new technique”). Jeebropilly colliery is also referred to in the third document as a ‘full-scale co-disposal operation’ (PH Morris and DJ Williams, “The porosity of co-disposed coalmine wastes”). The latter document also refers to ‘Goonyella-Riverside’ as the site of a ‘field trial’. But the oral evidence constituted nothing more than an assertion by Dr de Kretser that co-disposal was practised at those mine sites. Dr de Kretser did not provide any further explanation of what type(s) of co-disposal processes were used, and with what effect or for how long at any of those mines. Further, there was no evidence that any other person skilled in the art was aware of any of those mines practising co-disposal.

492    Finally, I cannot leave this topic without noting that the 1999 article of Messrs Williams and Seddon titled “Thickened tailings discharge: A review of Australian experience” (published in Tailings and Mine Waste 99, 1999, Balkema, Rotterdam) made no reference to either secondary flocculation or co-disposal.

493    In summary, in my view although co-disposal was part of common general knowledge, so too were the known problems concerning segregation and the known difficulties in seeking to overcome them.

494    Let me conclude the section on common general knowledge by also observing the following.

495    In his written evidence, Dr de Kretser provided a lengthy exposition of what he asserted to constitute common general knowledge at the priority date. But at trial, it became apparent that the way in which that evidence had been prepared had problematic aspects.

496    Dr de Kretser revealed during his cross-examination that he had previously studied the 785 application before giving any evidence in this proceeding. He had reviewed the 785 application in 2013 because “I needed to understand what was being done in that space” in order to assess whether his employer, Rio Tinto, was “potentially infringing”. This was not a casual review, but was undertaken “to understand the parameters of the patent and whether something that we were contemplating within Rio Tinto was within the scope or outside the scope”. So his evidence as to what constituted common general knowledge was given with full knowledge of the contents of the 785 application. This is one reason why the evidence of Dr de Kretser on common general knowledge should be given reduced weight; I will come to the relevant authorities in a moment.

497    Further, when Dr de Kretser came to give his evidence on common general knowledge in this proceeding, he again studied the 785 application and also a vast body of additional material before identifying what constituted common general knowledge at the priority date. Dr de Kretser confirmed that he was provided with and read the 785 application, the judgment of Kenny J, videos of trial work at Yarraman, the ACARP report, the Bulga video, a Rheomax brochure, and that he had read all of that material before he started to write his report. When further questioned about this, Dr de Kretser confirmed that:

My recollection is that I reviewed the patent and as a – I can’t remember the exact order in which I wrote the documents, but predominantly the first part that I sat down and wrote was the common general knowledge section.

But you accept that, as far as you can recall, you wrote that after you had received and read the patent?---I believe that that’s probably the case. I can’t recall exactly.

498    Dr de Kretser also revealed that he had not finished the common general knowledge section of his report at the time of being provided with the Condolios patent, the Backer & Busch 1981 paper, the Backer & Busch 1986 paper, the OreBind video, the OreBind product sheet, and an affidavit of Mr Michael Schmidt. Now he said that “I would have started writing the – the CGK section before I read a lot of this”. But the evidence of Dr de Kretser was that he had only written “the majority of my CGK section” by 28 February 2017, by which time he had been provided with and reviewed all of the documents identified, as well as various other documents, including further prior art and documents concerning the OreBind process.

499    In my view the process followed by Dr de Kretser was not fully conducive to resulting in completely objective evidence as to what constituted common general knowledge.

500    Further, the way in which Dr de Kretser was instructed to prepare his evidence had some other problematic aspects to it. In total, Dr de Kretser was provided with eight separate letters of instruction, between 22 June 2016 and 14 March 2017. Throughout this period of time, Dr de Kretser apparently dictated, in “lots and lots of meetings”, the contents of his affidavit to SNF’s solicitors. However, he was not provided with a hard copy of his affidavit to review, or even the parts he had dictated on any day, “which made it a difficult process”. Indeed, it was not until “a couple of weeks beforehand, before we finalised the affidavit”, that he was given a full printed copy to review. Further, throughout this lengthy period of meetings and dictation, Dr de Kretser was repeatedly provided with, and then instructed to ignore, a mass of material which ultimately formed no part of the evidence in this proceeding. It is unnecessary to elaborate further.

501    BASF submits that what Dr de Kretser has done with the benefit of having read the opposed applications, the prior art, various discovered documents, and affidavits filed in this proceeding, is to reconstruct a working thesis of what was known in 2003, which bore little resemblance to his own knowledge or that of a person skilled in the art at the priority date. Such a criticism is not completely devoid of force on some aspects of his evidence. But at the end of the day it is for me to make my own assessment of these matters based on the totality of the evidence. And in that context I found Dr de Kretser to be highly intelligent, very technically skilled and articulate, albeit that in terms of what was known at the priority date I have preferred the evidence of Dr Farrow on the more contentious matters which evidence is better supported by the industry evidence and the contemporaneous documents and publications at and prior to the priority date.

OBVIOUSNESS - LEGAL PRINCIPLES

502    The applicable provisions of ss 7(2) and (3) are the following:

(2)    For the purposes of this Act, an invention is to be taken to involve an inventive step when compared with the prior art base unless the invention would have been obvious to a person skilled in the relevant art in the light of the common general knowledge as it existed in the patent area before the priority date of the relevant claim, whether that knowledge is considered separately or together with the information mentioned in subsection (3).

(3)    The information for the purposes of subsection (2) is:

     (a)    any single piece of prior art information; or

     (b)    a combination of any 2 or more pieces of prior art information;

   being information that the skilled person mentioned in subsection (2) could, before the priority date of the relevant claim, be reasonably expected to have ascertained, understood, regarded as relevant and, in the case of information mentioned in paragraph (b), combined as mentioned in that paragraph.

503    In relation to the relevant legal principles, I adopt what I said in Meat & Livestock Australia Limited v Cargill, Inc (2018) 354 ALR 95 at [678] to [694] to the following effect.

504    The question is whether the claimed invention lacks an inventive step over the prior art base. An invention is taken to involve an inventive step when compared to the prior art base unless it would have been obvious to a person skilled in the relevant art in light of common general knowledge as described in Minnesota Mining and Manufacturing Co v Beiersdorf (Australia) Ltd (1980) 144 CLR 253 at 292 per Aickin J (Minnesota Mining) as it existed in the patent area (the then s 7(2) of the Act) before the priority date, whether that knowledge is considered separately or together with information of the kind described in the then s 7(3).

505    The term “obvious” means “very plain” (Aktiebolaget Hässle v Alphapharm Pty Ltd (2002) 212 CLR 411 at [34] per Gleeson CJ, Gaudron, Gummow and Hayne JJ (Aktiebolaget Hässle) and Lockwood Security Products Pty Ltd v Doric Products Pty Ltd (No 2) (2007) 235 CLR 173 at [51]). The inventive element needed to sustain a patent can be small. A “scintilla of inventiveness” will be sufficient and “no smallness or simplicity will prevent a patent being good” (Meyers Taylor Pty Ltd v Vicarr Industries Ltd (1977) 137 CLR 228 at 249 per Aickin J). Relevant further content has been given to determining obviousness in The Wellcome Foundation Ltd v VR Laboratories (Aust) Pty Ltd (1981) 148 CLR 262 at 286 per Aickin J in stating:

whether the hypothetical addressee faced with the same problem would have taken as a matter of routine whatever steps might have led from the prior art to the invention, whether they be the steps of the inventor or not.

506    Further, in relation to experiments, his Honour said (at 280 and 281):

In the present case it was admitted by the respondent that the test of obviousness was an objective one, but it was argued that evidence of a subjective character was admissible. That is no doubt true in some cases because expert witnesses are often properly asked whether they found a particular invention “surprising” to them. That however does not answer the question whether evidence of the steps which the patentee took is relevant and therefore admissible. Evidence of what was in the patentee’s mind may be admissible as evidence of the state of the art, but could seldom be otherwise admissible. Evidence of what he did by way of experiment may be another matter. It might show that the experiments devised for the purpose were part of an inventive step. Alternatively it might show that the experiments were of a routine character which the uninventive worker in the field would try as a matter of course. The latter could be relevant though not decisive in every case. It may be that the perception of the true nature of the problem was the inventive step which, once taken, revealed that straightforward experiments will provide the solution. It will always be necessary to distinguish between experiments leading to an invention and subsequent experiments for checking and testing the product or process the subject of the invention. The latter would not be material to obviousness but might be material to the question of utility.

507    The question to be posed was whether putative experiments leading from the prior art to the invention as claimed were part of the inventive step or were of a routine character to be tried as a matter of course. That question has an affinity with the Cripps question posed in Olin Mathieson Chemical Corporation v Biorex Laboratories Ltd [1970] RPC 157 and paraphrased by French CJ in AstraZeneca AB v Apotex Pty Ltd (2015) 257 CLR 356 at [15] in the following terms:

Would the notional research group at the relevant date, in all the circumstances, which include a knowledge of all the relevant prior art and of the facts of the nature and success of [the existing compound], directly be led as a matter of course to try [the claimed inventive step] in the expectation that it might well produce a useful alternative to or better drug than [the existing compound]?

508    Further, the plurality said in Aktiebolaget Hässle (at [58]) that:

The tracing of a course of action which was complex and detailed, as well as laborious, with a good deal of trial and error, with dead ends and the retracing of steps is not the taking of routine steps…

509    Now the Act does not direct an inquiry respecting each integer of the claimed invention. Rather, the correct inquiry is whether the invention as claimed, that is, the combination of integers was obvious, not each of its integers.

510    Further, it is erroneous to characterise as obvious the variation of all parameters or the trying of all choices until one proves successful, where the prior art did not point to it. Similarly, it is erroneous to characterise as obvious the exploration of a new technology or a promising field of experimentation, where the prior art gave no more than general guidance.

511    Further, in Aktiebolaget Hässle the plurality cited (at [74]) Judge Rich in In re O’Farrell (1988) 853 F 2d 894 at 903 who said:

[F]or many inventions that seem quite obvious, there is no absolute predictability of success until the invention is reduced to practice. There is always at least a possibility of unexpected results, that would then provide an objective basis for showing that the invention, although apparently obvious, was in law nonobvious.

512    Now impermissible hindsight should be avoided in determining whether a claimed invention lacks an inventive step. Indeed, the misuse of hindsight is most common in relation to combination claims (Minnesota Mining at 293 per Aickin J).

513    For a combination invention, the question is whether the combination, not each integer, is obvious. It is simply impermissible to take any one integer or take each integer seriatim and ask whether each integer involved an inventive step. The invention is the combination. The combination is what is claimed as the monopoly. And the relevant question is whether the combination involves an inventive step. As the plurality in Aktiebolaget Hässle (at [41]) said:

… The claim is for a combination, the interaction between the integers of which is the essential requirement for the presence of an inventive step. It is the selection of the integers out of “perhaps many possibilities” which must be shown by Alphapharm to be obvious, bearing in mind that the selection of the integers in which the invention lies can be expected to be a process necessarily involving rejection of other possible integers. This expression of the issue follows what was said by Aickin J in Minnesota Mining.

(Citation omitted.)

514    Further, Lord Davey stated in In the matter of Klabers Patent (1906) 23 RPC 461 at 469, in terms approved by Dixon J in Palmer v Dunlop Perdriau Rubber Co Ltd (1937) 59 CLR 30 at 73:

A proper combination for a patent is the union of two or more integers, every one of which elements may be perfectly old, for the production of one object which is either new, or at any rate is for effecting an old object in a more convenient, cheaper or more useful way. But the point in a combination patent must always be that the elements of which the combination is composed are combined together so as to produce one result.

515    More generally, in Aktiebolaget Hässle (at [21] and [78]) the importance of avoiding hindsight was reinforced in the following terms:

[T]he warnings in the authorities against the misuse of hindsight are not to be repeated as but prefatory averments and statements of trite law. The danger of such misuse will be particularly acute where what is claimed is a new and inventive combination for the interaction of integers, some or all of which are known. It is worth repeating what was said by Lord Diplock in Technograph Printed Circuits Ltd v Mills & Rockley (Electronics) Ltd:

Once an invention has been made it is generally possible to postulate a combination of steps by which the inventor might have arrived at the invention that he claims in his specification if he started from something that was already known. But it is only because the invention has been made and has proved successful that it is possible to postulate from what starting point and by what particular combination of steps the inventor could have arrived at his invention. It may be that taken in isolation none of the steps which it is now possible to postulate, if taken in isolation, appears to call for any inventive ingenuity. It is improbable that this reconstruction a posteriori represents the mental process by which the inventor in fact arrived at his invention, but, even if it were, inventive ingenuity lay in perceiving that the final result which it was the object of the inventor to achieve was attainable from the particular starting point and in his selection of the particular combination of steps which would lead to that result.”

After its review of the evidence, the Full Court concluded that Astra's “development” of the formulation “was essentially an exercise in trying out various known possibilities until the correct solution emerged” (emphasis added). That view of the matter wrongly takes as the starting point the assumed result. It succumbs immediately to the seduction of hindsight. Also, the notion of trying out possibilities invites the repetition of criticisms made earlier in these reasons.

(Emphasis added, citations omitted.)

516    In Lockwood (No 2) (at [46]) the High Court repeated the caveat against the misuse of hindsight. Clearly, any inordinate use of hindsight is likely to conceal or gloss over all of the problems, blind alleys and choices of options encountered by the patent applicant along the path to the invention. To present its evolution as some seamless and inevitable linear progression involving no more than the application of routine steps or methods is to engage in some intellectual property lawyer’s version of a Whiggish ahistorical approach.

517    Further, it is self-evident that problems with hindsight are even further elevated where the claimed invention concerns the application of known principles.

518    I will return to the question of hindsight later in evaluating SNF’s allegations of obviousness in this proceeding. But let me make one other observation at this point on this topic.

519    In Minnesota Mining & Manufacturing Co v Tyco Electronics Pty Ltd (2002) 56 IPR 248, Heerey, Emmett and Dowsett JJ said that little weight should be given to expert evidence that a claimed invention was obvious if the expert was provided with a copy of the patent or other relevant information before giving their evidence (at [45] and [46]):

The manner in which the evidence of some of the experts in the present case was bought into existence suggests that relatively little weight should be given to certain of that evidence. For example, witnesses were provided with a copy of the patent. They were either provided with a large number of other documents or found them in response to the task that was set them. That is hardly calculated to result in objective evidence as to what the hypothetical uninventive but skilled worker would have done. To give the patent to a prospective witness is tantamount to leading the witness. Further, unless the other documents were part of the common general knowledge in Australia before the priority date, they are not relevant to any question of obviousness.

Evidence by “experts” on the question of obviousness … is not always likely to be helpful: see Firebelt Pty Ltd v Brambles Australia Ltd (2002) 188 ALR 280; 54 IPR 449 at [46]. Indeed, where evidence is obtained in circumstances such as just described, the evidence is not likely to be helpful at all.

520    These principles have been consistently applied by this Court. Lesser weight is to be given to the evidence of experts asserting an invention to be obvious after they have reviewed the patent or been provided with other information concerning the state of the art. Undoubtedly, at the least there is a need for caution where an expert asserting obviousness knew both the problem and the solution.

521    Further, expert evidence as to what asserted s 7(3) information would have disclosed at the relevant time has lesser weight if an expert has undertaken an exercise of looking for the integers of the claims in the prior art and trying to find them, if necessary by combining different parts of the article without explanation of why he would have done so at its publication date in the absence of knowledge of the claims (Austal Ships Pty Ltd at [156]).

522    Let me say something further about common general knowledge.

523    Section 7(2) first requires consideration of what would have been obvious to a person skilled in the relevant art in the light of the common general knowledge as it existed before the priority date, putting to one side for the moment s 7(3) information. Common general knowledge is knowledge “generally known and accepted without question by the bulk of those who are engaged in the particular art” (British Acoustic Films Ld v Nettlefold Productions (1936) 53 RPC 221 at 250 per Luxmoore J). Information cannot be treated as part of the common general knowledge unless there is evidence of its general acceptance and assimilation by persons skilled in the art. And information does not constitute common general knowledge merely because it might be found for example in a journal, even if widely read by such persons. Further, as Luxmoore J said at 250:

In my judgment it is not sufficient to prove common general knowledge that a particular disclosure is made in an article, or series of articles, in a scientific journal, no matter how wide the circulation of that journal may be, in the absence of any evidence that the disclosure is accepted generally by those who are engaged in the art to which the disclosure relates. A piece of particular knowledge as disclosed in a scientific paper does not become common general knowledge merely because it is widely read, and still less because it is widely circulated. Such a piece of knowledge only becomes general knowledge when it is generally known and accepted without question by the bulk of those who are engaged in the particular art; in other words, when it becomes part of their common stock of knowledge relating to the art. Whatever else common general knowledge may be, it has never in my judgment included public knowledge of particular documents reports or scientific papers and the like. The knowledge of a number of individuals that a particular suggestion or particular suggestions has or have been made for the use of biasing in a particular apparatus, or a number of particular apparatus, cannot be held to be common general knowledge. It is certainly difficult to appreciate how the use of something which has in fact never been used in a particular art can ever be held to be common general knowledge in the art.

524    Further, as stated in Minnesota Mining by Aickin J (at 292), the notion of common general knowledge:

involves the use of that which is known or used by those in the relevant trade. It forms the background knowledge and experience which is available to all in the trade in considering the making of new products, or the making of improvements in old, and it must be treated as being used by an individual as a general body of knowledge.

525    So it must be knowledge that is known and available to all in the trade or at least the bulk of those who are engaged in the relevant art. Accordingly, information ascertainable by a routine literature search is not of itself taken to be common general knowledge. And patent specifications do not form part of common general knowledge without evidence that they have been absorbed into common general knowledge.

526    And as further elucidated by Jagot J in Gilead Sciences Pty Ltd v Idenix Pharmaceuticals LLC (2016) 117 IPR 252; [2016] FCA 169 at [216] (affirmed in Idenix Pharmaceuticals LLC v Gilead Sciences Pty Ltd (2017) 134 IPR 1 per Nicholas, Beach and Burley JJ), it is erroneous to treat a document as being part of common general knowledge simply because skilled persons could readily locate and assimilate its contents. Her Honour went on to explain (at [217]):

It may be accepted that instant recall of an article is not required. This does not mean, however, that documents found by searching for a subject-matter, rather than by some form of recall or reminder of what is already known to exist, are common general knowledge. This is so irrespective of the fact that experts in the field read widely. Further, it is not the case that mere publication and republication proves that a document and its contents have entered the common general knowledge. Nor is it the fact that a document and its contents necessarily form part of the common general knowledge merely because one expert knows or has managed to locate it and assimilate its contents. Such a document may or may not form part of the common general knowledge. The relevant inferences are to be drawn on the basis of the whole of the evidence.

527    There is another point to be made concerning common general knowledge. There is no general principle permitting admissions in the specification of a patent to be used to establish in and of themselves that information is common general knowledge. Whether information has become so widely assimilated that it forms part of common general knowledge must be determined on the evidence, although admissions can be considered as part of that evidence.

528    Let me say something at this point on s 7(3) albeit briefly.

529    In addition to using common general knowledge on a stand-alone basis, common general knowledge can be aggregated with s 7(3) information. That part of the prior art base which is common general knowledge and the information referred to in s 7(3) are considered for the purpose of looking forward from the prior art base to see what the skilled person is likely to have done when faced with a particular problem. Now in a case where the problem is known and is part of the common general knowledge, the problem may be similar to that which the patentee claims to have solved with the claimed invention. But where the problem addressed by the patentee does not form part of common general knowledge, the relevant starting point is the prior art base, but not including the problem as identified in the patent specification. As was noted in AstraZeneca AB v Apotex Pty Ltd (2014) 226 FCR 324 by the plurality (at [203]):

If the problem addressed by a patent specification is itself common general knowledge, or if knowledge of the problem is s 7(3) information, then such knowledge or information will be attributed to the hypothetical person skilled in the art for the purpose of assessing obviousness. But if the problem cannot be attributed to the hypothetical person skilled in the art in either of these ways then it is not permissible to attribute a knowledge of the problem on the basis of the inventor’s “starting point” such as might be gleaned from a reading of the complete specification as a whole.

530    The purpose of the inquiry is to determine whether the invention is obvious to solve the perceived problem, looking forward from the prior art base. But of course this may not have been the patentee’s starting point.

531    Finally, on the question of inventive step, let me say something about secondary indicia. There are a number of matters from which it might be inferred that an invention involved an inventive step including the following.

532    The first matter is a long felt need. The existence of a known need for the invention, which had not been satisfied by any product on the market, is a recognised indication of inventiveness (Minnesota Mining at 297 to 298 per Aickin J). Such evidence “has a role to play in a case concerning an inventive step” and it is said that I “should be slow to ignore secondary evidence or to rely on [my] own assumed technical expertise to reach conclusions contrary to such evidence” (Lockwood (No 2) at [115] and [116]).

533    The second matter is commercial success. Evidence that the invention when introduced on the market had substantial commercial success also suggests that the invention was not obvious.

534    The third matter is the failure of others. Evidence that others sought but failed to find the invention may also be of relevance (ICI Chemicals & Polymers Ltd v Lubrizol Corp Inc (1999) 45 IPR 577 at [94] per Emmett J). This may be particularly so where many others had been working in the field, as revealed by a large volume of prior art, but failed to find the invention (Welcome Real-Time SA v Catuity Inc (2001) 113 FCR 110 at [158] per Heerey J). Indeed, it has been said that “[w]hen skilled, non-inventive persons…looking for improvements, fail to arrive at the invention, it is impossible to suggest that it would have been obvious to the skilled and not necessarily inventive person” (Lockwood (No 2) at [119]).

535    The fourth matter is lapse in time. Where there is a long period of time between the claimed invention and the closest prior art, this may also suggest an inventive step. It has been said that if some years went by following publication of the relevant prior art before that which is the substantial equivalent of the invention was claimed and patented, this may point towards inventiveness (Graham Hart (1971) Pty Ltd v SW Hart & Co Pty Ltd (1978) 141 CLR 305 at 331 per Aickin J).

536    I will return to the evidence concerning these secondary indicia later. Let me now turn to the first main question.

OBVIOUSNESS IN LIGHT OF COMMON GENERAL KNOWLEDGE ALONE

(a)    Summary of SNF’s case on this aspect

537    SNF has identified the relevant problem that was to be overcome in the following terms.

538    It says that tailings commonly consisted of a mixture of coarse and fine particles. The tailings may have inherently comprised coarse and fine particles, or a separate stream of coarse material may have been deliberately introduced to the thickener underflow downstream of the thickener. It then says that adding coarse material into thickener underflow (co-disposal) was a technique commonly employed to dispose of coarse waste material and to assist in obtaining a stable deposit of tailings. I would note at this point that the evidence does not support such a proposition in the sense of commonly.

539    It then says that a problem to be overcome in the context of tailings beaching was to achieve a stable deposit with clear water release where the coarse and fine particles did not segregate on deposition (the problem). The problem was related to the desirability of reducing the footprint of deposited tailings, maximising recovery of relatively clean water, and faster rehabilitation of the deposited tailings. But I would note at this point that the focus at the time was looking at water recovery in an earlier part of the process.

540    It is said that the problem was particularly an issue for miners practising co-disposal. If the coarse and fine particles segregated on deposition, the deposited material would be less likely to form a stable deposit, resulting in a larger footprint, the released water would contain fine particles, and the deposited solids would take longer to rehabilitate. But again, water release was dealt with earlier in the process in terms of using mechanical dewatering devices such as belt press filters.

541    Instead of SNF’s articulation of the problem, I would prefer to re-express the matter with the following emphasis.

542    Let it be accepted that in general terms it has always been a challenge for mineral processing operations to treat and deal with tailings in order to deposit them in such a way that:

(a)    clean, particle free water is rapidly released from the solids within the tailings so it can be reused in the mining operation;

(b)    the tailings take up a minimum area in the deposition or impoundment facility where they are deposited; and

(c)    the coarse and fine particles are deposited in a homogeneous manner rather than being segregated, which optimises the tailings disposal and enhances the rehabilitation and re-use of the land in the mining operation.

543    Now as at the priority date, the minerals industry had broadly recognised the need to treat mine tailings in a more environmentally and commercially acceptable manner. This included the need to maximise water recovery and minimise the deposition area. It was also recognised in the field to be desirable to try to ensure that tailings were deposited in a homogeneous manner without segregation of fine and coarse material. But in the period leading up to and even beyond the priority date, the industry’s main focus in trying to improve tailings disposal was on enhancing thickener technology.

544    Moreover, in terms of water recovery, the focus at the time was using mechanical dewatering devices to remove water earlier i.e. through belt press filters and centrifuges. And the use of secondary flocculation at the time was to aid that process to maximise water recovery at that point, in preference to maximising water recovery at the point of deposition.

545    In my view, in terms of the problem, SNF has incorrectly linked:

(a)    the desire for improved rigidification and co-disposal on the one hand; and

(b)    the question of water release at the discharge point on the other hand.

546    As at the priority date, they were not together the recognised problem to be solved together. Rather at the time the focus was on removal of water at an earlier point through improving thickener technology and improved use of mechanical dewatering devices closer to the plant so as to maximise water recovery at that point.

547    Contrastingly the claimed invention did not contemplate water removal at that earlier point but rather at the point of deposition, leaving the water in enhanced fluidity and pumpability through to the point of deposition with water recovery at that later point. There would be flocculation near the discharge point so that it could stand and rigidify, releasing a large volume of water to be pumped back. As Mr Shavin QC submitted, this involved “(n)o mechanical dewatering, cost savings, and, counterintuitively, no water release up close to the processing plant”.

548    Let me continue for the moment with a summary of SNF’s submissions.

549    SNF submits that it would have been obvious to a person skilled in the art seeking to address the problem in the context of a co-disposal process to implement the claimed invention for the following reasons.

550    It was known that the problem was caused by the decrease in the yield stress of the tailings caused by shear forces imparted on the tailings during pumping to the deposition area.

551    It was known that the problem could be addressed by removing water from the tailings and increasing the yield stress of the tailings on deposition to inhibit the segregation of the coarse and fine particles. It was known that flocculants removed water from tailings and increased the yield stress.

552    Persons skilled in the art used water soluble flocculants in aqueous form to increase the yield stress of the tailings in the thickener to recover water from the tailings out of the thickener and create material in the thickener underflow suitable for tailings beaching.

553    However, there was a limit to how much yield stress could be introduced at the thickener stage due to the effects of shear thinning on the underflow and other issues.

554    SNF says that an obvious solution to the problem was therefore to add further flocculant to the thickener underflow prior to deposition.

555    The addition of further flocculant to thickener underflow was commonly practised in the use of belt presses and centrifuges.

556    It says that a person skilled in the art would inevitably add the flocculant in the form of an aqueous solution (being the form in which flocculant was almost always used in all applications).

557    The person skilled in the art knew that the peak level of flocculation, and therefore viscosity, will occur within 10 to 20 seconds after addition to the slurry if flocculant is added in solution.

558    The person skilled in the art therefore knew that it was necessary to add the flocculant near the end of the pipeline because flocculant added in aqueous solution would increase yield strength very quickly and then to break down very quickly due to the presence of shear forces.

559    The person skilled in the art knew how to routinely adjust parameters such as the appropriate dose, dilution rate and dose point in order achieve a highly flocculated structure (improved rigidification) which was stable on deposition with non-segregation of coarse and fine materials and clear water release.

560    SNF submits that the person skilled in the art faced with the problem and armed only with common general knowledge would have been directly led as a matter of course to try the claimed invention with the requisite expectation that it might well produce a useful alternative to or better result than existing processes.

561    SNF submits that before implementing the process referred to above, a miner would need to assess the additional capital costs, the ongoing cost of flocculant and associated labour costs at the particular mine as opposed to pursuing other options including simply increasing yield stress in the thickener underflow at the thickener to account for the shear forces during pumping.

562    At this point let me elaborate on SNF’s submission that it was obvious to address the problem by SDITB.

563    SNF says that persons skilled in the art knew that pumping thickener underflow from the thickener to the deposition area exposed the flocculated structures of the slurried tailings to shear forces which reduced the yield stress of the tailings on deposition.

564    It is said that Dr Farrow acknowledged that thickener underflow was pumped distances to a deposition point from 300 metres to at least 10 kilometres, or up to even 25 kilometres.

565    Dr de Kretser’s evidence was that typical pipeline pumping velocities were between two to five metres per second, but could be slower depending on the type of pump being used. Higher velocities were often used to maintain homogeneity using the turbulence in the flow to keep the particles together. However, those high velocities would impart higher shear forces on the tailings and reduce the yield strength of the material, which would increase segregation on deposition.

566    It is said that Dr Farrow accepted that it was well known that thickener underflow would shear thin as it was being pumped due to shear forces and that flocculated material once exposed to shear forces will start to degrade without further flocculant being present. More particularly, it was well known that flocculated slurries were shear sensitive when exposed to shear forces when pumped through a pipe from a thickener to a deposition area or to a belt press filter, the flocculation state would be degraded if the slurry was pumped over a long distance, and the level of shear thinning will depend on the degree of turbulence in the transfer pipe and the transit time over which the flocs are subject to turbulence. Mr Bellwood also accepted that when subjected to turbulent forces in a pipeline the tailings would suffer shear degradation.

567    The effects of shear forces on yield stress were well known to others skilled in the art within Ciba. For example, on 19 November 2001 Mr Keith Parker sent an email to Mr Tim Cameron and others at Ciba and noted in relation to the Gallagher process that [the] whole idea [of flocculation] is just to rebuild yield stress that is lost through pumping when the flocculated structure is broken down” and that “[with] dry addition we have continuous dissolution over 30 – 60 minutes thus always have floc available to rebuild yield stress”.

568    It is said that addressing the degrading effects brought about by shear in the pipeline could not be overcome by the simple expedient of adding more flocculant in the thickener. If the thickener underflow had a high yield strength:

(a)    it impaired pumpability and put strain on pumps;

(b)    it cost more in terms of power than a less viscous underflow;

(c)    if the yield strength of the tailings within the thickener was too high the rakes in the thickener could be damaged; and

(d)    as flocculants are expensive using higher doses of flocculant in the thickener would increase costs.

569    It is said that a person skilled in the art knew that there needed to be a balance struck between the need to have the material sufficiently fluid to be pumpable and the desire to create a material with sufficiently high yield stress on exit from the pipeline. Operators had extensive experience varying a number of parameters to optimise the operation of the thickener and the flow characteristics of the tailings.

570    That knowledge would teach the person skilled in the art of the need to flocculate the thickener underflow to get a higher yield stress material to address the problem before shear forces reduced the yield stress again. They would therefore be looking at a way to introduce more flocculant into the underflow shortly prior to deposition of the treated tailings in the deposition area.

571    Further, SNF submits that in addressing the problem a person skilled in the art would draw on their knowledge and experience in the use of flocculants including their knowledge and experience in relation to the use of flocculants in thickeners and for pre-treating thickener underflow fed into belt press filters and centrifuges. SNF submits that in using their accrued knowledge, experience and skill, one obvious technique they would consider is adding a second dose of flocculant to the thickener underflow in the outlet pipe near the outlet end.

572    SNF says that the techniques involved in the secondary dosing of belt press filter feed and the techniques involved in SDITB are in practical terms the same. Importantly, in each case, a highly flocculated (rigidified) material is created. In each case, the thickener underflow is treated with a second dose of flocculant to improve the rigidification of the material compared with the rigidity of the material when it emerges from the thickener. If the second dose of flocculant is sufficient, the material will rapidly dewater and exhibit “improved rigidification” as claimed. The only difference is that in the use of belt presses the pre-treated material is deposited onto a belt and it then passes through the belt press filter, following which the material is conveyed to a deposition area, whereas with SDITB the material is deposited on a sloped area so that released water flows to a lower point and can be pumped back to the plant for re-use.

573    It is said that a person skilled in the art’s experience with belt press filters was that flocculant in aqueous solution could be effectively mixed into the high concentration thickener underflow in the pipeline shortly prior to the point at which the treated underflow entered the belt press. This produced highly flocculated tailings that rapidly dewatered and exhibited improved rigidification when deposited onto the belt of the belt press filter.

574    It says that Dr Farrow’s evidence in relation to the use of belt press filters before the priority date was that:

(a)    a person skilled in the art was accustomed to adding a second dose of flocculant to thickener underflow to pre-treat it before feeding it into a belt press filter;

(b)    the flocculant was almost invariably added in the form of an aqueous solution;

(c)    a person skilled in the art had to determine where the flocculant needed to be added between the thickener and the belt press filter to achieve well flocculated material. To do so they would make a judgment based on their skill and experience about the most effective point at which to add the flocculant so as to allow it to adequately mix with the thickened slurry to create a highly flocculated material;

(d)    that second dose of flocculant was introduced into a high solids concentration thickener underflow;

(e)    it was essential to have a highly flocculated slurry (with a degree of structural integrity) as the slurry was fed into the compression zone of the belt press filter;

(f)    the material discharged from a belt press filter is rigidified;

(g)    improved rigidification would enhance beaching;

(h)    the feed material “certainly [showed] improved rigidification” in the feed to the belt press filter; and

(i)    if sufficient flocculant had been added, the feed material would be networked in the same way as material that exhibited “improved rigidification” within the meaning of the claims.

575    SNF says that it was put to Dr de Kretser that the person skilled in the art would not, at the priority date, have had experience in the practice of secondary dosing of flocculant to thickener underflow. But Dr de Kretser’s response was:

The person in the field with experience of – or what I would consider routine training in flocculation of mineral tailings would have been exposed to the concept of addition of aqueous polymer into concentrated tailings streams as part of practices such as feeding to belt press filters, centrifuges, to develop a highly flocculated structure with requisite or with clear properties that are consistent with improved rigidification.

576    SNF says that Dr de Kretser’s evidence was that the person skilled in the art would not confine their understanding of the use of flocculants for secondary dosing arising from their experience with belt press filters and centrifuges to only those particular pieces of dewatering technology.

577    Similarly, Mr Holtzman, when asked whether the first time he heard about SDITB was when he was informed of this by Nalco in October 2000, responded that he knew of SDITB as a result of his work with the use of belt press filters at Jangardup in the late 1990s.

578    Moreover, so SNF says, it is apparent from Ciba’s contemporaneous documents that Ciba’s employees appreciated the parallel between SDITB and secondary dosing of belt press filter feed. Mr Scammell, for example, in his email to his colleagues regarding his work at Yarraman dated 4 August 2003, Mr Scammell said that:

I believe the addition of a [viscosity modifier] to the thickener underflow alone does not add any value to the operation. The VM should be added to the combined thickener underflow and sand tails stream. This is where we have had success in Queensland…

What we have done in Queensland that is different to the original slump test work is use the diluted VM to basically flocculate the slurry and hence release water from the slurry. It can be said we are producing a belt press filter feed, but then discharging the treated slurry onto the ground.

579    Under cross-examination, Mr Scammell’s evidence was that the reference to “belt press filter feed” was the best equivalent he could think of to describe the qualities of the tailings he produced at Yarraman to his colleagues within Ciba.

580    Similarly, Mr Cameron in an email to his colleagues at Ciba dated 10 June 2003 dealing with the optimisation of co-disposal said:

As with optimising a belt press or centrifuge, success is all about correct conditioning of the slurry – and continual optimum conditioning relies on ·a consistent feed of uniform slurry. Variations in slurry characteristics requires repositioning of the dosing point or change in polymer concentration or dose.

581    Further, SNF says that Dr Farrow accepted that if a person skilled in the art prior to the priority date had thought of implementing SDITB they would “invariably” have decided to add the flocculant in solution. Indeed, his own experience using flocculants was that he always added the flocculant as an aqueous solution.

582    SNF says that based on accrued knowledge, experience and skill, a person skilled in the art would have expected to succeed in mixing the flocculant into the thickener underflow so as to arrive at rigidified material which on deposit was stable, released clear water and the coarse and fine particles in the material did not segregate. By undertaking tests of a routine trial and error nature, a person skilled in the art could readily assess the level of yield strength which had developed in the flocculated material and improve the rigidity of the material by adjusting the dose point and the amount of flocculant added until the desired outcome was achieved.

583    Further, SNF says that Dr Farrow accepted that for many years before the priority date, persons skilled in the art relied on their skill and experience to mix aqueous solutions of polymer with slurried material to achieve the outcome that they were content with.

584    It is said that Dr Farrow accepted that persons skilled in the art were accustomed to undertaking tests of a routine trial and error nature to assess the level of yield strength developed in flocculated material and to discern whether or not there was an improvement in the rigidification of the material.

585    It is said that Dr Farrow also accepted that persons skilled in the art knew:

(a)    when flocculant was added to tailings it needed shear forces to mix the flocculant through the tailings, but at the same time, the shear forces operated to break down flocs formed by the flocculant;

(b)    they would have to strike a balance between how long it took for the flocculant to mix with the tailings and not allowing too much time in the pipeline for the shear forces to break down the yield stress unacceptably;

(c)    that mixing conditions were important and that they would be able to modify the mixing conditions in one way or another;

(d)    that it was desirable that the mixing of the flocculant or application of shear should be relatively short and that mixing did not continue for significant periods after the flocculant was dissolved;

(e)    that adding flocculant in aqueous solution to tailings achieved a peak structure very rapidly and that it degraded as a result of shear thereafter; and

(f)    the flocculant needed to be added close to the discharge point so that there could be sufficient time for mixing (to form a fully networked structure), but not too much time for there to be unacceptable degradation through shear forces.

586    It is said that Dr Farrow also accepted that the person skilled in the art, relying on their knowledge, experience and skill could readily determine how much flocculant to add when introducing it into the pipeline, ascertain the right dose point and determine where a suitable balance was struck so that flocculation would occur and shear degradation would not compromise the degree of flocculation required to get the outcome desired on deposition.

587    If a person skilled in the art did not achieve an effective outcome due to what Dr Farrow described as “inefficient mixing”, they could make routine adjustments to overcome that issue, such as moving the point of addition further back in the outlet pipe.

588    If the material was deposited in a deposition area and the mine operator was not satisfied with the rigidity of that material, the person skilled in the art would know that rigidification would be improved by adding more flocculant.

589    It is said that Dr Farrow also accepted that if a person skilled in the art did not get the rigidification he or she desired, they could make modifications to the way the flocculant was put in and increase the dose, to obtain improved rigidification within the meaning of the opposed applications. He further accepted that a person skilled in the art could routinely determine the right dose point and effective method of addition as part of their commonly undertaken adjustments of operating parameters.

590    If the fine particles were segregating from the coarse particles, a person skilled in the art could inhibit that by increasing the dose of flocculant “if you did it in the correct way”. Dr Farrow accepted that it was within the skill set of the person skilled in the art to do it in the correct way.

591    If the flocs did shear to some extent, the person skilled in the art could simply add more flocculant and still get an effective result on deposition.

592    It is said that Dr Farrow’s evidence was that achieving improved rigidification was simply a matter of adding enough flocculant at the right dose point.

593    Dr Farrow accepted that if enough flocculant was added, a network structure would be produced in the tailings, which is the hallmark of improved rigidification.

594    Now Dr Farrow advanced a number of reasons why he considered it would not have occurred to he or others in the field to add flocculant in aqueous solution to the tailings line shortly prior to deposition in a tailings disposal area. His reasons were:

(a)    the focus at the time was on improving thickener outcomes. The work he and his colleagues were undertaking in the field at the priority date was taking them in a “completely different direction”;

(b)    the difficulty of effectively mixing flocculant into the thickener underflow; and

(c)    adding flocculant in aqueous solution to the tailings would mean adding further water to the system. Improving thickener underflow involved “[eliminating] as much water as possible from the underflow and [recovering] it in the thickener”. It would have been “counterintuitive” to retain a significant volume of water in the underflow and try to recover it in the tailings disposal area. It would have been considered that this would result in a low yield stress material that would wash away material from the discharge point.

595    But as to the first matter, SNF says that Dr Farrow’s evidence that the focus of research was solely directed to thickeners should not be accepted for the following reasons.

596    It is clear from Dr Farrow’s evidence that the primary focus of his work was on thickener technology. However, improving thickener performance was only one of the approaches to tailings disposal that was known and used in the industry at the time.

597    Dr Farrow had little experience or knowledge concerning the use of belt presses and centrifuges which had been used widely in the field in Australia since the 1980’s. Dr Farrow candidly conceded that he had “very limited experience” in belt press filters. He had only “seen a few in operation”. Indeed, it appears that it was only at Coal & Allied where he was “within touching distance” of a belt press filter. He did not consider himself to be an expert in belt press filters.

598    SNF says that Dr Farrow was also not aware of the numerous instances of third parties who were trialling SDITB processes before the priority date.

599    SNF says that Dr Farrow’s myopic focus on thickeners is well illustrated by the fact that he was apparently unaware of the presentation given by Mr Dymond, which I have referred to as the Dymond paper, which disclosed SDITB. It appears that the presentation of the Dymond paper immediately preceded Dr Farrow’s presentation at the same conference.

600    Dr de Kretser’s evidence is that this conference was the leading regular conference in the field and that he regularly received papers that were presented at these conferences. Dr Farrow evidence is that it was likely that persons skilled in the art would have, as did he, attended this conference. Mr Schroeter was aware of the Dymond paper (presented in May 2001) before the priority date. Dr Clarke also read the Dymond paper prior to the priority date when he was undertaking research for Iluka in around 2002.

601    SNF says that Dr Farrow was also apparently unaware of the publication by Ciba of the paper presented by Mr SJ Adkins and Mr DT Smith (Adkins paper) at the Minerals, Metals and Materials Society annual meeting held Washington in February 2002 and then published in that society’s journal (Rheology Aids for Effective Mud Disposal (Light Metals 2002)).

602    As to the second matter, ineffective mixing, one of Dr Farrow’s key contentions was that a person skilled in the art would not have considered adding the flocculant in aqueous solution to the tailings as they would know that, as the flocculant would need to be added close to the discharge point, there would be insufficient time for mixing to occur prior to discharge. Indeed, his evidence was that if high dosages of flocculant were added into a high solids concentration suspension such as a thickener underflow, it would have been “impossible” to effectively mix the flocculant amongst the particles in the slurry. But SNF says that Dr Farrow’s evidence should not be accepted for the following reasons.

603    First, Dr Farrow’s contention is inconsistent with the evidence that flocculant in aqueous solution had been routinely and effectively added in high doses to the thickener underflow fed into belt press filters and centrifuges. Dr Farrow accepted in cross-examination that the thickener underflow fed into belt press filters was a high solids concentration slurry.

604    SNF says that any person skilled in the art who had worked with belt press filters or centrifuges was aware that flocculant could be effectively added in aqueous solution to a high solids concentration environment in the outlet pipe close to the discharge point in order to achieve an extensive flocculated network.

605    Second, Dr de Kretser’s evidence is that if there was non-uniform mixing, this was addressed by increasing the dose flocculant. SNF says that Dr Farrow accepted that if a person skilled in the art had a problem achieving uniform mixing of flocculant added in aqueous solution, the consequence was that they needed more flocculant to achieve the result they were seeking. Mr Schroeter explained that inefficient mixing requires the addition of more flocculant to achieve optimal mixing. This imposes an economic penalty rather than a technical one.

606    It is said that this is consistent with Dr de Kretser’s evidence that although mixing flocculant in aqueous solution into a high concentration of solids flowing in a pipe can be difficult, it was practised successfully in the field at the priority date. Dr de Kretser’s evidence is that persons skilled in the art were aware of any necessary engineering practices required to achieve the requisite mixing.

607    Third, difficulties in mixing a viscous solution into a viscous slurry were addressed in the use of belt press filters and centrifuges by adjusting the dilution of the flocculant.

608    SNF says that after Ms Beveridge had undertaken her laboratory testwork adding flocculant in solution to Ardlethan tailings, and before any field trial work had been undertaken by Ciba at Yarraman or Sandalwood, Ms Beveridge recorded that where the flocculant is added in solution for maximum efficiency [the flocculant] must be added “enough distance from the discharge point to allow homogenous mixing into the sample, but not too far away so that in-line shear effects break up bonds connecting slimes particles”.

609    It is said that clearly Ms Beveridge knew in October 2002, before any field trial work was undertaken by Ciba of SDITB adding the flocculant in solution, that if flocculant was to be added for SDITB in aqueous solution, effective mixing could be achieved provided a suitable dose point was identified that took account of shear effects and pipeline residence time.

610    As to the third matter of adding further water to the tailings, SNF says that Dr Farrow’s contention ignores the fact that:

(a)    the amount of water associated with the added polymer would be very small compared to the amount of water flowing with the tailings stream; and

(b)    the net water recovery would be significantly higher than without the aqueous polymer addition.

611    Furthermore, it says that Dr Farrow’s assertion is inconsistent with the evidence that persons skilled in the art using belt press filters routinely added a second dose of flocculant in aqueous solution to the slurry after it had been treated in the thickener. The reason they did so was that the net effect of the improvement in water recovery afforded by flocculation far outweighed the minor impact of the added water.

612    Accordingly, SNF submits that the person skilled in the art, faced with the problem and armed only with common general knowledge would have been directly led as a matter of course to try the process of claim 1 of the 785 application with the requisite expectation that it might well produce a useful alternative to or better result than existing processes.

613    SNF also says that the evidence indicates that the various features added by the dependant claims do not confer inventiveness. In this regard it has submitted the following:

(a)    Claims 2 to 8 of the 785 application simply describe the chemistry of the range of commonly available flocculants as at the priority date. This is not in dispute between the parties. BASF has accepted that the invention does not reside in the chemistry of the flocculants.

(b)    Claims 9 and 10 of the 785 application require that the tailings being treated comprised dispersed particulate solids which are mineral or which are residues from a mineral processing operation. This refers to the disposal of tailings always found in a mineral processing operation.

(c)    Claim 11 of the 785 application requires that the process to provide a “heaped geometry”. Obtaining a “heaped geometry” was a key indicator of the successful achievement of rigidification.

(d)    Claim 12 of the 785 application requires the material to be derived from the tailings from a mineral sands process. This simply specifies a particular type of mineral tailings, being mineral sands tailings, which were regularly encountered by persons skilled in the art at the priority date. They possess no different functional requirements in terms of treatment than any other mixture of coarse and fine tailings that could be treated by the process.

(e)    Claim 13 of the 785 application requires the dispersed particulate solids to have particle sizes less than 100 microns, in which preferably at least 80% of the particles have sizes less than 25 microns. This simply specifies a typical distribution of particle sizes found in mineral tailings encountered and treated by persons skilled in the art at the priority date.

(f)    Claim 14 of the 785 application requires dispersed particulate solids to have a bimodal distribution of particle sizes comprising a fine fraction and a coarse fraction, in which the fine fraction peak is substantially less than 25 microns and the coarse fraction peak is substantially greater than 75 microns. This simply specifies a typical distribution of particle sizes characteristic of co-disposal processes performed by persons skilled in the art at the priority date.

(g)    Claim 15 of the 785 application requires the material to have a solids content in the range 15% to 80% by weight, preferably in the range 40% or 50% to 70% by weight. This simply specifies a typical solids content range found in mineral tailings encountered and treated by persons skilled in the art at the priority date.

(h)    Claim 16 of the 785 application requires that the tailings being treated are underflow from a thickener. As there is no apparent dispute between Dr de Kretser and Dr Farrow that the words “during transfer” in claim 1 refer to a process in which the flocculant is being added at any point between the point at which the underflow is discharged from the thickener to the point at which the treated material is discharged, claim 16 is not relevantly different to claim 1.

(i)    Claim 17 of the 785 application requires that wet or dry coarse particles are added to the underflow from the thickener either before or during the addition of an effective rigidifying amount of the water soluble flocculant. The final outcome of a process as described in this claim is identical to the addition of flocculant to any tailings inherently having a broad particle size distribution. This claim appears to be narrower in scope than claim 1, as it requires the addition of flocculant either during or after the addition of coarse particles. In this regard, the process the subject of claim 17 of the 785 application is equivalent to the process the subject of claim 1 of the 568 application; I will discuss this in a moment. SNF submits that it would be obvious to a person skilled in the art to add the flocculant before, during or after the addition of coarse particles depending on their particular mine set-up as:

(i)    a person skilled in the art knew flocculant added in aqueous solution would need to be added close to the deposition point;

(ii)    to add the coarse material after the addition of flocculant may require transporting the coarse material a long distance to the deposition point in a separate pipeline which would add infrastructure cost and pumping cost; and

(iii)    if however the additional coarse material such as sand was readily available near the deposition area it would save costs to add it during or after the addition of the flocculant to save the costs of transporting the coarse material to the deposition area.

(j)    Claim 18 of the 785 application requires that the material is transferred to a holding vessel prior to being pumped to the deposition area. Use of holding tanks for either buffering (to smooth out process flows) or mixing (of material with reagents) purposes was common practice in the field at the priority date.

(k)    Claim 19 of the 785 application requires (amongst other things) that the process provides a “heaped geometry” and co-immobilises the fine and coarse fractions of the solids in the material. Achieving a heaped geometry and co-immobilisation of all solids within the tailings were key indicators of the successful achievement of rigidification. A heaped geometry would necessarily result in a higher driving force of the water released from the material, due to gravity drainage.

(l)    Claim 20 of the 785 application requires that the tailings are pumped to an outlet, where they are allowed to flow over the surface of previously rigidified material wherein the material is allowed to stand and rigidify to form a stack. Deposition of tailings over previously deposited rigidified tailings and the standing and rigidifying to form a stack was commonly undertaken as part of tailings beaching and was a key indicator of the successful achievement of rigidification and a natural consequence of achievement of rigidification.

(m)    Claims 21 to 23 of the 785 application require the flocculant to be mixed with the tailings prior, during or subsequent to a pumping stage, or as it exits the outlet pipe. The experts agree that it was within the ordinary skill of a person skilled in the art to determine the precise point of addition of flocculant in light of the particular circumstances at the mine.

(n)    Claim 24 of the 785 application requires the material to dewater during rigidification releasing liquor. Dewatering is a key indicator of the successful achievement of rigidification and a natural consequence of achievement of rigidification.

(o)    Claim 25 of the 785 application requires that the liquor is recycled to a mineral processing operation. It was common practice to recycle reclaimed water/liquor from tailings back to the mineral processing operation. This reduced the amount of fresh water required by the process.

(p)    Claim 26 of the 785 application requires that the clarity of the liquor is improved by the addition of an aqueous solution of water-soluble polymer. Improved clarity is a key indicator of the successful achievement of rigidification and a natural consequence of achievement of rigidification.

(q)    Claim 27 of the 785 application requires that the released liquor is subjected to further processing to reclaim or reuse the valuable materials. Re-use of recovered water from tailings to make use of contained valuable materials was commonplace at the priority date.

(r)    Claim 28 of 785 application has the same integers as claim 1 of the 785 application.

(s)    Claim 29 is an omnibus claim and so is limited to the examples in the 785 application which include laboratory tests and a description of two field trials involving mineral sands mines.

614    SNF also submits that the claims of the 568 application were obvious in light of common general knowledge alone.

615    Let me first deal with the independent claims. Claim 1 of the 568 application is expressed in similar terms to claim 1 of the 785 application, except that the “process” claimed includes the following additional language:

(a)    combining aqueous suspensions of fine and coarse particulates for the purpose of co-disposal to form the material;

(b)    mixing of the aqueous suspensions into a homogeneous slurry; and

(c)    adding the flocculant in aqueous solution during or after the mixing the aqueous suspensions.

616    Claim 1 of the 568 application is therefore limited to a process where the flocculant in aqueous solution is added “during or after” the mixing of the coarse and fine particulates for the purposes of co-disposal. This may be contrasted with claim 1 of the 785 application, which encompasses the flocculant in aqueous solution being added before, during or after the combination of the coarse and fine particulates for the purposes of co-disposal. The scope of claim 1 of the 568 application is therefore narrower than the scope of claim 1 of the 785 application.

617    SNF submits that the differences between claim 1 of the 568 application and claim 1 of the 785 application cannot confer inventiveness. There is no dispute between the experts that the selection of a dose point was a matter of routine within the ordinary skill of the person skilled in the art. Accordingly, the precise point of addition during transfer relative to the combination of coarse particles with the thickener underflow it is said cannot confer inventiveness.

618    Claim 2 of the 568 application is in substance the same scope as claim 1 of the 785 application. Accordingly for the reasons outlined above, it is said that claim 2 of the 568 application is obvious.

619    Claim 3 of the 568 application differs from claim 1 of the 785 application only in that it does not limit the flocculant to having a particular intrinsic viscosity. Claim 3 of the 568 application therefore encompasses flocculants of any intrinsic viscosity and therefore molecular weight. Consistent with the evidence of both experts, BASF has accepted that the invention does not reside in the chemistry of the flocculants. Accordingly, it is said that claim 3 of the 568 application is obvious for the reasons set out above in relation to claim 1 of the 785 application.

620    Finally, SNF says that the dependent claims of the 568 application introduce various additional integers which are the same as, or similar to, the additional integers added to the dependent claims of the 785 application. Accordingly, it says for similar reasons above that none of the additional features of the dependent claims of the 568 application confer inventiveness.

(b)    Analysis – Non-secondary indicia

621    Let me begin at this point by analysing the question of inventiveness without regard to what I will describe as secondary indicia of inventiveness.

622    In summary, I am inclined to the following views.

623    First, as BASF correctly submitted, it seems to me that as at the priority date secondary flocculation was not the focus of interest in the art. Rather, the focus was improving thickener underflow technology.

624    Second, the focus was to remove water at or around the thickener phase by improving such technology or enhancing the removal of water by the use of mechanical dewatering equipment, rather than to leave a significant volume of water in the underflow to be recovered in the tailings disposal area, and indeed with added water, albeit a modest additional amount, by the use of an aqueous form of flocculant at that later stage.

625    Third, it is not apparent why a person skilled in the art would appreciate that adding more flocculant would provide a solution to the problem.

626    Fourth, although the use of co-disposal was part of common general knowledge, there were known to be significant difficulties as I have elaborated on earlier in my reasons. Moreover, it is not apparent why a person skilled in the art would combine co-disposal with secondary flocculation with an expectation of success.

627    Let me now deal with each of these matters in turn.

628    First, in my view, secondary flocculation was not the focus of interest in the art.

629    As BASF correctly submits, neither Dr Farrow nor Dr de Kretser, the two independent experts in this proceeding, had ever trialled or observed secondary dosing before the priority date. Further, neither Ms Herzig (who had by 2002 been working as a qualified metallurgist for 10 years) nor Dr Clarke (who had very extensive experience in the industry from 1974) had ever trialled or observed secondary dosing, until they became involved in the trials undertaken by Ciba at Yarraman. Further, neither Mr Schroeter nor any employee of SNF had any experience of secondary dosing of tailings except in conjunction with the use of mechanical dewatering equipment. Further, Mr Bembrick, who had worked in the industry since 1997, was not aware, in late 2002, of a single mine anywhere in Australia which added a secondary dose of flocculant at or close to discharge point into a tailings area. Further, none of the Ciba employees who have given evidence in this proceeding, Mr Scammell, Ms Beveridge or Mr Bellwood, had any experience, prior to Yarraman, of the secondary dosing of polymer in aqueous solution to achieve stacking.

630    In these circumstances, I am not satisfied that it would be a matter of routine for an unimaginative person skilled in the art to attempt secondary dosing outside the mechanical dewatering context to improve rigidification with an expectation of success at the priority date. Rather, the unimaginative person skilled in the art, at the priority date, would not have been exposed to or attempted secondary dosing save perhaps in the context of mechanical dewatering devices.

631    I also agree with BASF that this is well illustrated by the secret use evidence relied upon by SNF, which I will discuss later. Further, other than the work undertaken by Ciba, the only evidence of secondary dosing of flocculant in aqueous solution adduced by SNF related to:

(a)    the OreBind process, which concerned mechanical dewatering;

(b)    the secondary flocculation process at Beenup, which failed;

(c)    the process implemented at Londonderry in the early 1980s, which involved treating tailings constituting 3%wt solids as part of a conventional settling process;

(d)    the use by SNF of secondary dosing as part of a mechanical dewatering process; and

(e)    a single small scale and confidential research trial undertaken by Iluka at Yoganup in 2000.

632    The work undertaken at each of those mines suffered from numerous limitations, which I will discuss later.

633    Second, the focus of the work in the field at the priority date was on improving thickener underflow technology, not secondary flocculation. The focus on thickener technology is illustrated by the following matters, as BASF has pointed out. The leading publication in the field, the PTT Guide published in 2002, contained considerable focus on thickener technology, but did not contain a single reference to secondary dosing. Further, the method Mr Holtzman developed as part of his trial work at Cable Sands in 2001 focused on improving thickener technology. Further, the attempts made by CRL to address the problem of segregation at the Yarraman mine in 2002, before Ciba became involved, focused on altering operational variables in the concentrator and the thickener without success. Further, the attempts made by Boral to address the tailings disposal problem at the Stapylton quarry involved experimenting with the flocculants being used in the thickener.

634    It seems well apparent that to the extent that people in the field were encountering problems with tailings disposal, they were seeking to solve those problems at the thickener stage.

635    Third, there is something counter-intuitive about SNF’s case.

636    As BASF points out, the evidence of Dr Farrow, which I accept, was that “[i]t would have been quite counter-intuitive to me and other people skilled in the field at the time to retain a significant volume of water in the underflow and try to recover it in the tailings disposal area”. This evidence was supported by a number of documents which recorded the reactions of people working in the field to the invention ultimately claimed in the opposed applications. Let me give some examples.

637    When Dr Clarke first learned of the testwork being undertaken by Ciba, including that being undertaken by Mr Scammell at the Yarraman mine, he sent a lengthy email to Mr Scammell in which he identified “the points of greatest interest to Iluka”. In that email, he wrote that one option would be to flocculate the tailings at “an appropriate point in the pipe”, but that “[t]he negative aspect [of this] would be the amount of water to be pumped to the disposal point and back”. When cross-examined, Dr Clarke confirmed that he had perceived that there was a cost advantage in recovering the water close to the plant and that a problem with secondary flocculation was the need to pump the water a substantial distance.

638    Similarly, when Mr Scammell reported within Ciba on the results of his testwork at Yarraman, Mr Mike Gallagher responded as follows: “The results you achieved look excellent but from the photo’s it looks more like flocculation to produce a plug rather than increasing the yield stress of the slurry. You also appear to be getting a lot of water release which is helping the slump height in the test but will this released water be a hindrance at the deposition site?”

639    As BASF points out, these contemporaneous responses of Dr Clarke and Mr Gallagher provide strong corroboration of the evidence of Dr Farrow that it was counter-intuitive, at the priority date, to add “further water to the system after the thickener” which was “completely contrary to the focus of our work which was to remove as much water as possible during the thickening phase”. They also accord with the evidence given by other witnesses at trial.

640    Mr Bellwood explained that the way in which yield stress is increased by the secondary dosing of polymer in powder is quite different to the way in which yield stress is increased by the use of the solution, because with the use of powder, “part of the effect is also the absorption of the water by the particle, the polymer” such that “yield stress developed because some of the water in the system is necessarily tied up by the hydrating polymer”. Mr Bellwood further explained that: “If you flocculate material, then yes, generally speaking you will get a – a yield stress increase; but you also usually get a water release, so you get a two-phase system in the pipeline”.

641    Mr Bellwood also explained that he would not have expected the second dosing of a low solids slurry with aqueous solution to be effective, because although this might “bind all the solids together; but at the same time, you will release a lot of water. And therefore that material then will not have a great deal of dynamic viscosity, because you will effectively have big clumps of particles surrounded by a lot of water”. And as he explained, the end result of secondary dosing with solution, with lots of water release, in fact “looks quite different” to the use of the Gallagher process. In this way, a difference between a secondary dosing process using powder and a secondary dosing process using solution, is that when a low solids containing slurry is dosed with solution, “yield stress would only increase once that water has been removed. So in a pipeline, where the water can’t go anywhere, the yield stress of the material in the pipeline would actually decrease on flocculation”.

642    Further, Mr Scammell explained that: “Part of the advantage of adding the dry powder is that it absorbs some of that liquid or liquor that’s in the red mud, changing the viscosity while it slowly dissolves…whereas the solution is already dissolved and you’re injecting … it straight in”. This has the disadvantage that you are adding more water to a stream which will impact yield stress.

643    Further, as I have already referred to, people in the field were at the priority date focused on mechanical dewatering methods to remove water. This is illustrated by the trial work undertaken by Nalco in respect of the OreBind process. Further, the evidence of Mr Schroeter was that as at the priority date people reporting to him within SNF only had experience engaging in secondary dosing “upstream of some form of mechanical dewatering device”.

644    So, as Dr Farrow explained, “[t]he thinking was, though, about the performance of that equipment, per se, so if it’s the thickener or if it’s a belt-press filter, that’s what was the focus of where you added the flocculant – was to enhance the performance of that unit operation”. I agree with BASF’s submission that it would have been counterintuitive for a person skilled in the art, experimenting with and failing to achieve satisfactory results with a thickener or mechanical dewatering equipment to remove water from the tailings, to simply discard that equipment in its entirety, in the hope of getting a better result without the very equipment upon which reliance was being placed to achieve adequate dewatering.

645    Further, it would not have been obvious to a person working in the field at the priority date that, in order to improve rigidification, they should simply add more and more flocculant to the tailings stream. People working in the field were concerned about making the treated tailings difficult to pump and the possibility of blocking the tailings pipe. For example:

(a)    Mr Holtzman gave oral evidence that a difficulty he encountered with secondary dosing was that it made “the underflow difficult to pump” which was undesirable for a range of reasons.

(b)    Mr Schroeter explained that “if the dose is too high, then it can cause the slurry to cease moving during transfer to the deposition area”.

(c)    Mr Bellwood explained that in August/September 2002 he believed that the use of polymer in powder form would be superior to the use of polymer in solution form, because it would avoid “rapid viscosity spikes which may lead to pipeline blockages and/or excessive pumping pressures”.

(d)    Dr Farrow explained that, at the priority date, “[n]obody was looking to get a high yield stress in a thickener. That was – the whole issue about having to add a flocculant actually had the downside that it increased the yield stress of the material in the thickener”.

646    In my view it would not have been obvious to a person skilled in the art to simply add more and more flocculant to improve rigidification. To the contrary, the person skilled in the art would have been cautious to say the least in adding flocculant, for fear of blocking the tailings line.

647    In my view the person skilled in the art, at the priority date, would not know or suspect that improved rigidification as taught in the opposed applications could be achieved by secondary dosing. It would have been highly unlikely for a notional person skilled in the art, at the priority date, to ignore the risk of blocking the pipeline, and the pumping inefficiencies associated with increasing viscosity, and instead add more flocculant.

648    Fourth, the evidence at trial established that as at the priority date, co-disposal was known to be difficult, was not routinely implemented, and generally produced poor results. In those circumstances, a person skilled in the art would not at the priority date have been directly led as a matter of course to implement a co-disposal step as part of a secondary dosing process with an expectation of success. Let me elaborate.

649    When CRL attempted to implement a co-disposal process at Yarraman, CRL found that the coarse fraction and the fine fraction within the slimes / tailings mixture segregated on deposition. This was not desirable. CRL only trialled secondary dosing in conjunction with co-disposal once this was suggested as a technique by Mr Scammell.

650    Further, none of the trials of the OreBind process undertaken by Nalco at Ernest Henry, Bunbury, Wemen or Boral Stapylton involved a co-disposal process.

651    Further, the experience of Dr Clarke was that when co-disposal was attempted, the coarse and fines segregated in the impoundment.

652    Further, Mr Holtzman gave evidence that when co-disposal was attempted “what generally happened was only a small portion of the – of the fines stayed with the tails”. Further, Mr Schroeter gave evidence that when co-disposal was attempted “[t]here were issues with segregation”.

653    In my view, this evidence suggests that a person skilled in the art would be wary about attempting co-disposal. Indeed, if they had decided to trial secondary dosing, they would likely have focused on separately treating the fines, with the coarse fraction to be disposed of elsewhere.

654    Now there was some evidence that some people in the field contemplated combining co-disposal with secondary dosing.

655    Mr Scammell said that when he initially undertook testwork at the Yarraman mine in May 2002 it was not promising. That work initially focused on the separate dosing of the underflow. In his first written report to CRL, Mr Scammell wrote that:

Addition of thickener underflow to course tailings at a ratio of 1:10 resulted in a significant deterioration of slump angle. Even when the thickener underflow was dosed at 300gpt, the slump angle deteriorated significantly once this slurry was combined with the course tailings.

656    Further, Dr Clarke said that most of his work before the priority date was focused on treating the coarse and fine streams separately. And when he initially contacted Mr Scammell to discuss how a secondary dosing process might be used in conjunction with co-disposal, he wrote that:

Two ways that this could be done are:

    Thicken the fine and “dry stack” or similar directly in the pit OR

    Dewater the sands and then mix with fines thickened as at present to make a non-segregating mixture which would be disposed of directly in the pit.

657    But I agree with BASF that none of these suggestions bears significant resemblance to the process claimed in claim 1 of the 568 application or any of its dependent claims.

658    Further, Mr Holtzman said that the testwork undertaken by Cable Sands regarding the use of secondary dosing in conjunction with co-disposal, was complicated, detailed and lengthy trial work, following which Mr Holtzman still considered that “[i]t wasn’t proved that it would – would work in a full scale trial”.

659    Generally, BASF asks rhetorically: Why would a person skilled in the art, seeking to achieve a tailings disposal solution at the priority date, combine a rarely used and little known process (secondary dosing outside the mechanical dewatering context) with a difficult and uncommon process (co-disposal)? I agree with BASF that to do so would involve risk taking and experimentation. This is hardly the stuff of routine steps.

660    Now before proceeding further, it is convenient to note at this point that there are problems with the evidence of Dr de Krester on inventive step, some of which I have already touched on.

661    First, it is apparent that the way in which the evidence of Dr de Kretser on common general knowledge was prepared was flawed and affected by hindsight as I have indicated.

662    Second, some of the evidence of Dr de Kretser in relation to secondary dosing and co-disposal lacked any strong basis and was contradicted by much of the other evidence adduced.

663    Third, in circumstances where Dr de Kretser had not trialled, personally observed, or even published on secondary dosing before the priority date and first observed a secondary dosing process eight years after the priority date, he did not have strong experience to express an opinion on whether the process claimed in the opposed applications was obvious.

664    Fourth, the evidence of Dr de Kretser on obviousness had other difficulties. Dr de Kretser appeared to assess obviousness from the perspective of a person who was “analytical”, had “critical problem solving skills”, was resourceful, had “a willingness to trial novel, especially simpler, less technologically intensive solutions” and had “a willingness to develop / trial solutions in the field, even before fundamental aspects may be understood”. As Dr de Kretser conceded in cross-examination, the person skilled in the art he had in mind, when assessing obviousness, was “the opposite of unimaginative” and “proactive and resourceful and innovative” in the context of solving problems. But his perspective is problematic. The person skilled in the art is unimaginative without inventive capacity.

665    Fifth, and relatedly, the notional person skilled in the art posited by Dr de Kretser did not reflect the real world practical experience of those people working in the field. The evidence of Dr de Kretser was that a person skilled in the art would have routinely undertaken patent searches. But when it was put to him that no witness had identified or read the Condolios, Gallagher or Ledden patents before the priority date, Dr de Kretser conceded “it does surprise me” and that his experience at Rio Tinto between 2011 and 2015 might have been different to the practice in the field before the priority date.

666    In summary, in my view SNF has not made out its case of obviousness based upon common general knowledge alone. Moreover, that view is fortified by a consideration of the secondary indicia which I will now discuss.

(c)    Secondary indicia – General

667    There are a number of secondary indicia that I propose to elaborate on.

668    First, evidence of SDITB being implemented or published before the priority date is relevant and probative on the issue of obviousness. Prior acts including the use of SDITB provide probative evidence of what a team of skilled persons actually did prior to the priority date when attempting to address the problem.

669    Now SNF submits that the evidence of the prior acts is “at least as probative, if not more so [than] … the hypothetical opinions of the experts as to obviousness”, but I tend to think that its proposition is over-stated.

670    Second, SNF submits that the conduct of the inventor is also relevant on the issue of obviousness. I agree. But as I observed in BlueScope Steel Limited v Dongkuk Steel Mill Co (2017) 135 IPR 1; [2017] FCA 1537 at [38], evidence of the inventors conduct has secondary significance, but can be used to fortify the primary evidence and analysis in relation to inventive step. I will discuss in some detail Ciba’s own trial work.

671    Third, I will then say something briefly concerning long felt need, commercial success and the ACARP project.

672    Let me address each relevant secondary indicator in turn.

(d)    Evidence of third parties undertaking SDITB prior to the priority date

673    Now although BASF contends that I can infer that the claimed invention was not obvious from the fact that, despite a considerable number of persons skilled in the art undertaking testwork in the field, no person skilled in the art arrived at the precise invention claimed in the opposed applications, SNF submits that the evidence of the prior acts is to the contrary. It says that the evidence demonstrates that persons skilled in the art were undertaking SDITB in co-disposal processes, adding the flocculant in aqueous solutions, to address the problem and to achieve improved rigidification.

674    Now I should note at this point that as at the priority date, aside from SNF, there were three major suppliers of flocculant to the mining industry, being Ciba, Nalco and Cytec. Each of these flocculant suppliers had developed SDITB processes as follows:

(a)    Ciba had developed the Gallagher process (powder) and the Moody process (water insoluble flocculant);

(b)    Cytec had developed the process the subject of the Pearson patent (emulsion); and

(c)    Nalco had developed the OreBind process (aqueous).

675    Before proceeding further to discuss the evidence of third parties undertaking SDITB, let me say something about the Moody process; I will discuss the Gallagher process later.

676    The Moody process is the subject of European patent O 338 108 filed 12 March 1990 (but claiming priority from 13 March 1989) with G Moody the inventor that I have earlier referred to. The invention related to a process for the treatment of waste materials from the Bayer process of recovering alumina from bauxite, in particular to treat red mud to render it more easily disposable.

677    As described (p 2 and 3):

In a new process according to the present invention a material comprising an aqueous liquid with a dispersed particulate solids is pumped as a fluid then allowed to stand and rigidify and the rigidification is improved whilst retaining the pumpability of the material by, prior to pumping the material, blending into the material particles of a water-absorbent, water-insoluble polymer.

The blend of the dispersion with polymer is a pseudo plastic material, that is it is capable of being pumped but, even after being pumped, when allowed to stand, ie whilst under very low shear, forms a relatively rigid, stackable material.

The process of the invention is suitable for treating materials wherein the dispersed particulate solids have very small particle sizes, for instance, substantially all having sizes less than 100 microns, and even for materials wherein substantially all of the particles have sizes of less than 50 microns. It is of particular value where at least 90% of the particles have sizes less than 20 microns. The particles are usually of an inorganic material, usually a mineral. Although it may be useful for other materials requiring to be pumped and then stacked especially for materials which are filter cakes, for instance other mineral tailings or slimes, including phosphate, diamond or gold slimes, the major use of the present process is in the treatment of red mud from the final thickener of a Bayer process. The red mud may thus be the solids settled from the final thickener stage by the addition of flocculant alone, or, usually, the material is the filter cake from the pressure filtration of the slurry produced in the final wash stage. The red mud or other material which is pumped may have a solids content in the range 30% to 80% by weight, usually in the range 40% or 50% to 70% by weight. for instance 55% to 65% by weight. The sizes of particles in a typical red mud sample are substantially all less than 25 microns, for instance about 95% by weight of mud is particles less than 20 microns and about 75% is less than 10 microns, and about 95% more than 3 microns.

It is surprising that the process according to the invention forms a product which rigidifies far better, so as to minimise the area taken up by a stack of the material of given volume. This is achieved whilst maintaining the pumpability of the material, it being possible to provide a material which, under shear, has a viscosity not significantly different from the material without addition of polymer. In addition, in the absence of the water absorbent polymer, the material can be irreversibly shear thinning so that after pumping it does not rigidify at all. We have now found that the addition of polymer according to the present invention overcomes this problem and provides a material which becomes rigid when pumping is stopped.

678    In terms of polymer addition, it was said (at 5):

The polymer is mixed into the material in a manner which ensures good dispersion. If the polymer is added in the form of aggregates, the mixing should be such as to break the aggregates into the individual particles. Mixing may be carried out in line during the pumping process or may be carried out in a separate stage prior to the pumping stage. If the polymer is added during pumping, it is suitably added in line immediately before the pump, for instance which may be a centrifugal pump or a mono pump or other type of pump conventionally used for pumping the red mud. If the polymer is mixed in a separate stage it should be mixed in a mixer which encloses the slurry long enough for satisfactory mixing to occur, for instance in a plough share mixer which encloses the mixture during mixing and imparts sufficient shear to give adequate mixing. One suitable type of plough share mixer is a Loedige mixer. Another suitable type of mixer is exemplified by a ‘Turbulator’ manufactured by Ferro-Tech.

679    Claim 1 stated:

A process in which material comprising an aqueous liquid with a dispersed particulate solids is pumped as a fluid then allowed to stand and rigidify and the rigidification is improved whilst retaining the pumpability of the material by, prior to pumping the material, blending into the material particles of a waterabsorbent, water-insoluble polymer.

680    There are several points to note about the Moody patent. First, it is referring to the use of a water insoluble flocculant. Second, it says nothing about co-disposal. Indeed it is about the treatment of red mud and treating materials where the dispersed particulate solids have very small particle sizes. Third, the injection of polymer is before pumping. But nevertheless secondary flocculation is envisaged.

681    Let me now say something concerning third parties undertaking SDITB.

682    Mr Schroeter had experience of SDITB processes distinct from the OreBind process, which I will discuss in more detail in a moment, whilst working for Nalco.

683    Mr Schroeter undertook laboratory work involving SDITB at the Ranger Uranium Mine in the Northern Territory in around 1982. That work did not progress beyond laboratory work for economic, rather than technical, reasons. In or around 1982, Mr Schroeter implemented processes at two sites where tailings which were not treated in a thickener had flocculant added in the form of an emulsion to the tailings in transit to the deposition area. In recommending that the mines add the flocculant in the form of an emulsion rather than the conventional practice of flipping the emulsion into an aqueous solution before adding it to the tailings, Mr Schroeter recognised that doing so would use more flocculant. However, neither mine had the available funds to purchase the equipment necessary to flip the emulsion. Mr Schroeter was also aware of the process of adding flocculation in aqueous solution to a sewage slurry stream shortly before the deposition point and then allowing the material to drain and form a solid mass. The Nalco Water Handbook which he received during his employment at Catoleum, included a section on this practice. Mr Schroeter also read the Pearson patent in or around 1996.

684    Further, Mr Bembrick, who had worked for both Ciba and SNF, gave evidence that in around 2000 he implemented a process which involved adding flocculant to tailings close to the end of the outlet pipe at Benedict, a site near the Parramatta River. Mr Bembrick’s evidence is that in 2000 when Benedict was confronted with segregation in its tailings, and wished to obtain a stable deposit with clear water release, Mr Bembrick and his colleague Mr Tudor implemented a process of adding flocculant in aqueous solution into the tailings pipeline in which the flocculant was added close to the end of the pipeline. Mr Bellwood did not dispute Mr Bembrick’s evidence that flocculant was added into the tailings pipeline. Mr Bellwood, who did not attend the site and who was not Mr Bembrick’s supervisor in relation to this work, surmised however that the flocculant was added as a powder. But the references in Mr Bembrick’s visit reports to the use of a “wet Shutz” is a reference to a 1,000 litre container which is used to convert flocculant from a powder into an aqueous solution. I accept Mr Bembrick’s evidence on that limited aspect. But there are other more general problems with his evidence that should be noted at this point. The affidavit evidence given by Mr Bembrick regarding the work he claimed to undertake for Benedict was unreliable in many respects.

685    First, the detailed account of that testwork in his affidavit stands in contrast to the limited recollection he had of that testwork when giving oral evidence in the 2008 proceedings. Mr Bembrick gave evidence that “I can’t be certain here, but I don’t know where we dosed it into the line, but I assume from the experience now that it was closer to post-discharge”. But in the present proceedings Mr Bembrick deposed that “[f]locculant was diluted and added in an aqueous solution of 0.25% solution to the discharge of the dredger”.

686    Second, his oral evidence regarding the ACARP project, which he undertook in 2004 to 2006, was that:

… in my experience – short-term experience prior that, I was – I was really only involved in belt presses and centrifuges and dewatering equipment. So I didn’t have an opportunity to really – it goes in the machine; it comes out dry. This was an opportunity to actually see the flocculant structure and actually visually see how it behaves in front of me.

687    But that evidence is difficult to reconcile with the evidence contained in his affidavit that in 1999/2000 he implemented a similar process to that which he trialled as part of the ACARP project in 2004 to 2006.

688    Third, to the extent that he gave evidence regarding the secondary dosing trial work he undertook for Ciba at the Ardlethan mine in 2002 to 2003, he confirmed that what prompted that work was the work undertaken by Mr McColl and Mr Scammell at Yarraman, not his earlier work at Benedict.

689    Fourth, Mr Bembrick’s recollection of what occurred in respect of Benedict was limited, perhaps understandably given that it occurred many years earlier. Mr Bembrick confirmed that he did not have access to any documents other than those annexed to his affidavit, he could not recall whether he had written those documents, he did not have any independent recollection of the dates of the testwork, and he did not have a clear recollection of the equipment used and instead deferred to site reports.

690    Fifth, it became apparent that Mr Bembrick had tried to reconstruct, many years after the event, what he thought might have happened at Benedict. When first asked about his evidence in relation to water testing, his evidence was as follows: “You did not test the water, did you?---I did”. But when the basis for this evidence was tested, Mr Bembrick said:

You’ve just made that up on the spot? …I can’t 100 per cent confirm whether I took water samples and measured them and – and wrote down the measurement, but, in terms of experience, I would assume that it would be something that you would actually observe.

691    Indeed, Mr Bembrick confirmed that much of his evidence concerning Benedict was prepared in the following way:

I was talking to our – the – the lawyers, SNF lawyers, and, as I was talking with them, they kept on posing different questions and asking me questions of trying to remember what happened, and then ... say okay. Well, hang on a minute. Now I’m putting a bit more time and effort into – you know, thinking back – there’s – I mean, there’s snapshots that I can recall. I can just see them in my mind ... was one of them, the – getting the boat over, stuff like that, the dredge, Sydney tunnel, but, as you get asked more questions, more information becomes available. It’s just natural.

692    This was reconstruction undertaken by Mr Bembrick many years after the relevant events, in response to questions posed by the solicitors for SNF as to what might have happened.

693    Sixth, the weight of evidence suggests that the process implemented by Mr Bembrick for Benedict bore little resemblance to the process claimed in the opposed applications. The process was not a co-disposal process. Further, the process did not appear to achieve improved rigidification. It would seem that the relevant flocculated fines were not standing on deposition and it was the escape of sediment which mandated the building of a weir.

694    Finally, and in any event, there is no evidence that any other person in the field had any awareness of the testwork undertaken by Mr Bembrick. To the contrary, his evidence was that the work was confidential to Ciba.

The OreBind process

695    Let me now discuss the OreBind process. Nalco’s use of the OreBind process is an example of a flocculant supplier practising SDITB before the priority date.

696    The OreBind process was promoted by Nalco for use and used in the period 1999 to the priority date in Australia for the treatment of a wide range of mining tailings and waste materials. The OreBind process was the subject of a PowerPoint presentation distributed by Nalco to prospective clients, and a product sheet for the flocculant used in that process.

697    Nalco also recommended that its OreBind process could be used in conjunction with co-disposal processes before the priority date. This is confirmed by the evidence of Mr Holtzman, who was informed by a Nalco representative that the OreBind process was suitable for co-disposal.

698    Although the experts are in dispute as to whether improved rigidification was achieved in the various uses of the OreBind process, it is not in dispute that the OreBind process involved SDITB in which the flocculant was added in aqueous solution.

699    The evidence shows that SDITB was undertaken using flocculant added in aqueous solution according to the OreBind process at the following locations:

(a)    Ernest Henry;

(b)    Boral Stapylton;

(c)    Wemen; and

(d)    Bunbury albeit only as a laboratory test.

700    Now before I proceed further I should make some observations about the quality of the evidence led. Much of the evidence adduced by SNF at trial concerned alleged prior uses of secondary dosing or co-disposal processes, dating back as far as 1980. In many instances, the relevant witnesses sought to give their evidence largely or exclusively based on their recollection of events occurring decades ago. There are a number of difficulties.

701    Where the evidence adduced of alleged prior acts is the oral evidence of witnesses based on their recollection of events occurring some years ago, such evidence is to be treated with caution. The difficulty with such evidence is that even if witnesses are seeking to be truthful in giving their evidence, it is difficult to reach the necessary level of confidence with respect to the accuracy of such evidence.

702    In Fieldturf Tarkett Inc v Tigerturf International Limited (2014) 317 ALR 153, Jagot J identified difficulties with prior use evidence concerning events occurring some time ago. Jagot J reiterated “the cautions expressed about witnesses trying to recall details from many years before, particularly when the recollections are affected by knowledge of what is now relevant” (at [95]). She rejected some of that evidence for the following reasons.

703    First, there was a substantial period of time between the alleged prior use and when the witness first gave evidence regarding the prior use: “The significance of it should not be underestimated. It is an extraordinarily long time after the events in question. The risk of incorrect recollection and reconstruction instead is thus heightened” (at [96]).

704    Second, the evidence of prior use had been given in circumstances where the witness was aware of the claimed invention: “Again, the significance of this also cannot be overlooked. All of Mr Rooks’s evidence in 2004 and 2006 involved an attempt to recollect things relevant to the 904 application, the parent of the 2004 application… Accordingly, it must be accepted that Mr Rooks was never in a position to avoid the effects of hindsight” (at [97]).

705    Third, the evidence of prior use had changed over time: “It tends to reinforce the overall impression of reconstruction and forcing new information to fit within a framework first identified in 2004” (at [98]). The changing nature of such evidence placed it “on a very uncertain footing” (at [101]).

706    Generally, it appeared that much of the evidence before her merely involved a reconstruction of events which had occurred, with knowledge of the patent, rather than true recollection. And as her Honour (at [103]) said:

The problem in evidentiary terms is when the distinction between recollection and reconstruction (the latter being another completely normal human process which we all use all the time in everyday life) is not recognised, particularly when, as FieldTurf said, the reconstruction is carried out “with full knowledge of the target, being the claimed invention”.

707    The problems I now face in assessing analogous evidence are not dissimilar to those her Honour discussed. Let me turn then to the OreBind process.

708    Before discussing particular mine sites, there are five preliminary points to make about the relevance of the OreBind process to SNF’s allegations of obviousness.

709    First, SNF does not assert that the OreBind process itself was common general knowledge. Nor does SNF assert that any particular prior use of the OreBind process constituted s 7(3) information against which inventive step is to be tested. Accordingly, the OreBind process can only be relevant as a working example of how a person skilled in the art might at the priority date have attempted as a matter of routine to implement a tailings disposal solution.

710    Second, it is apparent from many of the contemporaneous documents produced by Nalco under subpoena that the OreBind process was not as at the priority date a routine tailings disposal solution, but rather, a confidential process the subject of ongoing research and development. In this respect, the evidence established the following matters.

711    Nalco had in December 1999 required a mine site to execute a five year confidentiality agreement in respect of its trial of the OreBind process. In this respect, it was common for Nalco to execute confidentiality agreements (at least in respect of the alumina industry) regarding its trial work and it was important to Nalco to keep the testwork confidential. Indeed, the evidence of Mr Woolley was that Nalco “were very serious on technology and protecting what they had developed”.

712    In July 2000, Nalco informed Iluka that its intent was to “establish a research project aimed at the development of chemical additivities for tailings disposal in the Mineral Sands Industry”. The trial work subsequently undertaken by Nalco regarding the OreBind process formed part of that research project.

713    In November 2000, Nalco sought to identify the “Technical Research Program Aims” to “determine the viability of a coal tailings dewatering process” utilising the OreBind process. Nalco identified necessary “areas of research” and repeatedly referred to its trial work as “research work”.

714    In March 2001, there was much discussion within Nalco about the OreBind process. As BASF correctly submitted, that discussion makes plain that Nalco was engaged in confidential research and development work and did not consider the OreBind process to constitute a routine tailings disposal process. For example, Karl Danenbergsons, an equipment engineer employed by Nalco, wrote that:

In a project such as this that involves new technology, most companies would run this project through their R & D department prior to handing over to Marketing and then Sales. We don’t have the luxury of having an R&D department that encompasses all the necessary people to do the job so it must be approached in a cross-functional way with a dedicated project manager planning, leading, organising and controlling the selected project team.

715    In June 2001, Nalco informed Boral, in respect of the trial work relating to the OreBind process at the Stapylton quarry, that “we are developing the process as we go” and proposed that Boral pay a monthly “process development fee”, in recognition of the fact that the trial work “will be expensive and very time consuming of Nalco manpower and resources”.

716    In December 2001, Nalco recorded research and development time and equipment charges in respect of its OreBind process trial work and claimed “R&D Tax Credits” for that trial work.

717    I agree with BASF that the trial work undertaken by Nalco in respect of the OreBind process cannot be used as a proxy for the state of the art at the priority date or the type of routine steps likely to be taken by a person skilled in the art. To the contrary, it represented a lengthy research project, undertaken at various mines subject to confidentiality obligations, over a number of years. Indeed, the evidence of Mr Schmidt was that he could not recall any fully scaled up implementation of the OreBind process prior to 2002 and that, in the period 1999 to 2003, it was a process in development with a lot of different people having different views as to what it was. Mr Schmidt further confirmed that, although he did not require his clients to sign confidentiality agreements, he did not expect his customers to disclose what he told them regarding the OreBind process.

718    Third, it is apparent that Nalco considered some form of mechanical dewatering to be essential to the OreBind process.

719    The trial work undertaken by Nalco at the Bayswater mine illustrates the focus on mechanical dewatering. Mr Woolley confirmed that the OreBind trial work undertaken at Bayswater involved pumping the treated material to a dewatering mechanism for mechanical dewatering. The documents produced by Nalco reveal the following matters.

720    In January 2000, Nalco undertook a preliminary evaluation of the OreBind process “to review the efficiency of the dewatering process”. The process used was to dose the underflow at two stages and “the resulting underflow material was then passed to the relevant dewatering device”. Ultimately, Nalco recommended “to evaluate two additional dewatering unit processes”.

721    In May 2000, Nalco reported on the “dewatering devices utilised to effect water removal from the conditioned underflow material” and undertook a “review of each dewatering device utilised”. It was concluded that “[t]he Orebind chemical program effectively conditioned thickener underflow for a subsequent water removal process” and Nalco recommended that “an inclined screw with free water drainage zones along the stream length, together with a compression zone be evaluated”.

722    In November 2000, Nalco identified the second aim of the OreBind research program as to “[d]efine the efficiency and selection of a dewatering process with review of vibrational screen and screw compression dewatering”.

723    In July 2001, Nalco considered that “the best way to de-water the Orebind slurry is through gentle squeezing”, there was “extreme doubt that the proposal of using a standard (sand) screw classified will actually be a success”, and that “a slow speed, screw type de-watering device is still the best option”.

724    In February 2002, further trial work was undertaken at Bayswater. Again, that work was concerned with “Investigating the Dewatering devices”. It was reported that:

With limited success in getting the dewatering devices to work I spent the final day investigating something a little left field from the project outline. The tailings underflow was treated and allowed to bypass the dewatering devices and collected in a sump. By allowing the material some extra time and mixing, a dense Orebind material was formed. This material readily dewatered, with its own weight compressing and releasing a clear centrate. The sump acted as a small batch thickener…

725    In other words, Nalco did attempt the “left field” idea of bypassing the mechanical dewatering step. Nevertheless, Nalco did not pursue this. Rather, it was recorded that “[u]nfortunately the sump contents do not withstand a centrifugal shear, with the material returning to thickener underflow consistency” and it was recommended that “[t]he dewatering device could be almost anything if the OreBind material is prepared well enough”.

726    I also agree with BASF that the focus on mechanical dewatering at Bayswater reflected a more general focus on the use of mechanical equipment as part of the OreBind process. In the minutes of a “Development Project” meeting concerning “Orebind” on 23 March 2001, it is recorded that “[d]e-watering equipment would also be required in the majority of cases”. Mr Schmidt confirmed this was generally true across different applications. The minutes further record a discussion of “the de-watering options” and it is recorded that “[a]ttendees cast a vote as to their preferred, top three de-watering methods”. Further, in an email exchange within Nalco in 2001, it is recorded that “[t]he dewatering process needs further refinement to give the results we are looking for – a spadable product” and that “[w]e need to consider the next step – getting a dewatering process up and running, to prove the OreBind process”. It is then suggested that Nalco “develop a relationship with a dewatering company” and “design a plant scale dewatering process”. Further, following some initial trial work at the Boral Stapylton quarry, Nalco recommended that it test “a sand screw for its suitability as a dewatering unit”, because the existing process was “costly and inefficient”. Nalco recommended the use of a screw classifier “to de-water the underflow slurry with free water being drained to the existing Sludge Pit”.

727    Similarly, Mr Schmidt agreed that at least prior to 2002 the OreBind process had not been worked on a scaled-up basis without dewatering devices being used.

728    Accordingly, and in my view, to the extent that Nalco’s experimentation with the OreBind process reveals anything relevant to the state of common general knowledge at the priority date, it reveals that there was an appreciation within Nalco that satisfactory dewatering of a treated slurry normally required the use of mechanical dewatering equipment.

729    Fourth, the evidence before me established that Nalco did not before the priority date implement a plant scale treatment program using the OreBind process without a mechanical dewatering step. To the contrary, and as discussed in detail below, the four specific uses of the OreBind process upon which SNF has sought to rely in this proceeding constituted one laboratory test (Bunbury), two discrete trials, which were conducted over a very short period of time and never scaled up (Ernest Henry and Wemen), and one implementation of a process in which Nalco recommended the use of a mechanical dewatering device (Boral Stapylton).

730    It would seem that despite Nalco experimenting with the OreBind process over a period of more than three years leading up to the priority date, it was still unable to successfully implement that process at any mine without the use of mechanical dewatering equipment, let alone as part of a co-disposal process.

731    Finally, some of the affidavit evidence adduced by SNF regarding the OreBind process was unreliable. The primary deponent adduced by SNF regarding the OreBind process was Mr Schmidt, a former employee of Nalco who had personal experience of trialling and promoting that process.

732    It was asserted in the affidavit of Mr Schmidt that “[t]he Orebind process was promoted by Nalco for use and used prior to May 2003 in Australia for the treatment of a wide range of mining tailings and waste materials, including coal, gold, copper, mineral sand, tin, silver and aluminium tailings…”. But the evidence of Mr Schmidt in cross-examination was that:

(a)    his own experience was much more limited than that range;

(b)    to the best of his knowledge, there had not as at May 2003 been a fully scaled up use of the OreBind process in respect of any of those materials; and

(c)    apart from Ernest Henry, which he described as “a pretty primitive filter trial”, he did not think there had been any commercial use of the OreBind process from beginning to end at any mine prior to May 2003.

733    Further, Mr Schmidt gave his affidavit evidence after he had viewed a promotional video made by Nalco in 2011 regarding the OreBind process, which he assumed to be the same as the process he had promoted in 1999 and which he relied upon to assist him to describe the OreBind process. He also gave his evidence describing the OreBind process as a combination of his recollections over the whole of the period of time that he worked for Nalco (namely, 1999 until 2009). Indeed, he also confirmed that it was “potentially, partially correct” that “a lot of the detail” contained in his affidavits regarding the OreBind process was gained by him during the period of 1999 to 2009 and was not known to him in 1999.

734    It seems to me that the evidence of Mr Schmidt was affected by hindsight and reconstruction. And his memory is likely to have been influenced by the contents of the 2011 video, events occurring between 1999 and 2009, and generally reconstruction.

735    Let me now turn to particular mine sites.

Ernest Henry

736    The Ernest Henry mine is a copper/gold mine located in Queensland where the OreBind process was trialled in October 1999.

737    As I say, SNF relied on the evidence of Mr Schmidt, who on behalf of Nalco implemented the use of the OreBind process at Ernest Henry, the work recorded in a report dated 28 November 1999 prepared by Mr Schmidt and Mr Strickland relating to the Ernest Henry trial (the Ernest Henry trial report) and also the evidence of Dr de Kretser.

738    Let me set out some extracts of this report.

739    The introduction stated:

Introduction

Test work was carried out by Nalco to determine the effectiveness in a small scale continuous test of our Tailings Stacking aid.

The results are best summarised by the photos below, showing a untreated sample and a treated sample. This illustrates clearly that the Nalco stacking aid can effectively increase the stacking angle of deposited tailings.

Other advantages seen were:

    Improved water drainage / solids drying.

    Greater stability.

    Homogeneous solids.

    Reduced channeling.

Based on these results, further test work is justified to quantify the process improvements and assess the economics.

740    The treated tailings results were reported as follows:

Initially a couple of spotter tests were done to assess the effect of the reagent addition on the slurry characteristics and the operation of the test rig. From these tests it was determined that:

    A significant increase in the angle of repose of the deposited tailings slurry was achievable.

    The deposited solids were stabilised, and dewatered significantly quicker, leading to rapid development of compressive strength.

    Slurry viscosity increased with reagent addition, and there was a limit to the dose rate the rig could operate with.

    Some unactivated reagent was found in the treated slurry indicating that results to be measured could be achieved with a lower dose rate of reagent under better reagent addition technique and improved mixing.

741    It was then said:

Treated run # 1 High Dose rate

This dose rate was considered the maximum the rig could work with and inefficient use of reagent was expected to occur. The resulting deposit of solids had a significantly higher angle of repose than the blank (estimated at 2.8 degrees, 4.6 in 100 grade). This is clearly seen in the difference in the nature of the deposits formed as seen in the photographs.

Observations made during the run:

    Flow rate of slurry reduced relative to blank, indicating higher pumping energy requirement.

    Slurry distribution even and concentric around deposition point, with a significantly higher angle of repose relative to the blank (including the formation of a “lip” on the outer end) and no segregation visible.

    Water release visible around edge of solids.

    Unactivated reagent visible in released water, indicating insufficient post addition mixing.

    The slurry was judged to be overdosed with reagent for given test conditions.

742    So, an emulsion was used “[u]nactivated reagent visible”. Further, it would seem that there were no fines in the water.

743    The post run observations were in the following terms:

Post run observations:

    Free water release from the solids was good, though there was a component of interstitial water which was slower to release, thought to be caused by viscosity from unreacted reagent in the water. The outer edges of the solids dried rapidly as the rate of water release slowed. No ponding occurred on the top of the solids, with the solids on the top drying out.

    When a segment of the solids was removed using a shovel, it showed good stackability. The area where the slice was acted to release water trapped with the solids close to the wall, indicating the free water had been trapped by the drying outer edges.

These pictures illustrate that the deposited solids had good integrity, as shown by the way they “stacked” at the end of the segment, where the solids had been pushed up.

744    It is not in dispute that at Ernest Henry the trial involved the following:

(a)    A stream of the tailings underflow post the thickener (measured at 67% solids) was diverted into a hopper and a second dose of flocculant was added into the hopper. The treated tailings were fed through a 20 metre long hose to simulate pipeline mixing and residence time and then discharged to the corner of a large concrete tailings bund.

(b)    The flocculant was OREBIND 3676, a water-in-oil emulsion, activated by mixing with water.

(c)    The flocculant was added at two different doses and measured against a blank control test.

(d)    When treated at both the low dose and the high dose, the material on deposition onto the floor of the bund had the appearance of cottage cheese.

(e)    Clear water broke free on deposition of the slurry onto the deposition area.

(f)    The resulting deposit of solids had a significantly higher angle of repose than the blank (estimated at 2.8 degrees, 4.6 in 100 grade).

(g)    When a segment of the solids treated with the high dose was removed using a shovel, the material showed good stackability.

(h)    One hour after running the shovel through the deposited tailings, the release of further clear water released from the solids was evident.

(i)    The solid stack of material built up on the deposition area was hard and rigid within 2 days. Indeed Mr Schmidt stood on the stacked material at the end of the trial, which was six days after deposition.

745    Now Dr Farrow accepted the following propositions as SNF points out:

(a)    The material that was deposited was thickener underflow.

(b)    The thickener underflow had a solids concentration of 67%.

(c)    The addition of the flocculant to the thickener underflow improved the rigidity of the deposited material, and the treated material appeared to be demonstrably more rigidified than the untreated material.

(d)    The treated material had a high yield stress that would allow it to beach on deposition.

(e)    The increase in stacking angle was consistent with extra structure having been introduced into the deposited tailings.

(f)    The achievement of a significant stacking angle with no segregation visible was consistent with a material that had been flocculated and further rigidified.

(g)    The pictures in the Ernest Henry trial report showed clear water release and the reference in the report to the fact that there was no segregation visible was a reference to no fine particles remaining in the released water.

(h)    The references in the Ernest Henry trial report to the release of free water from the solids being good was indicative of a material which had been flocculated and could hence release water and on deposition, increased its yield stress and had a higher angle or an increase in its slope or stack angle.

(i)    Figure 12 in the Ernest Henry trial report demonstrated that the material had a rigid structure which allowed water to drain away from the solids.

(j)    The enhanced rigidity was dose related and that an increase in dose produced a significant increase in the angle of repose (stacking angle) of the material.

(k)    The observation that the deposited solids were stabilised and dewatered significantly quicker, leading to rapid development of compressive strength, was consistent with enhanced rigidity.

746    Nevertheless Dr Farrow’s position concerning Ernest Henry was that:

(a)    it was not clear whether the flocculant was added as an aqueous solution or as an emulsion;

(b)    the material deposited did not result in improved rigidification; and

(c)    the process was not a co-disposal process.

747    As to the question of aqueous solution, the evidence of Mr Schmidt was that the flocculant was added in the form of an aqueous solution although I must say that it would appear to me that his memory was imperfect on this aspect. But in any event, as SNF points out, irrespective of whether or not the flocculant was 100% activated as an aqueous solution, Mr Schmidt’s evidence is that he attempted to flip the flocculant from an emulsion into an aqueous solution for the purpose of the trial. The OreBind product sheet also referred to the use of the G series or polyfeed flocculant preparation plant, which was used to invert or flip neat emulsion flocculant.

748    SNF says that Mr Schmidt’s evidence should be preferred over Dr Farrow’s conjecture regarding the form of the flocculant. It says that Dr Farrow’s evidence was based primarily on the fact that the flocculant was added prior to the pump. But Dr Farrow was not aware of the process conditions at the Ernest Henry mine, including the strength of the pump used. Mr Schmidt explained that the pump used was a low power centrifugal pump which produced relatively low shear, allowing the flocculant to be added to the tailings in aqueous solution before being pumped.

749    As to the question of improved rigidification, Dr Farrow asserted that the rigidification achieved at Ernest Henry was not “the same extent or degree as the patent contemplates when it speaks of improved rigidification”. But Dr Farrow accepted (as did Dr de Kretser) that improved rigidification was an improvement compared to the rigidification achieved in the deposition of untreated tailings. SNF says that the evidence establishes that the process used at Ernest Henry improved the rigidification of the material on deposition in the manner claimed.

750    In any event, so SNF submitted, Dr Farrow accepted that improved rigidification as taught in the opposed applications could have been achieved at Ernest Henry if Mr Strickland and Mr Schmidt had introduced the flocculant effectively and in a greater amount.

751    Now as to co-disposal, SNF had to accept that the process used at Ernest Henry was not a co-disposal process in that there was no combining of coarse and fine streams. Nevertheless it contended that the evidence establishes that:

(a)    the slurry at Ernest Henry comprised both coarse and fine materials; and

(b)    the Ernest Henry trial report notes that in respect of the run using the maximum dose rate that the equipment could handle, there was no segregation, which indicated that the material comprised coarse and fine materials which were co-immobilised on deposition, but notes in respect of this run “Unactivated reagent visible in released water, indicating insufficient post addition mixing” and that there was “interstitial water which was slower to release, thought to be caused by viscosity from unreacted reagent in the water”.

752    Now in relation to SNF’s contentions I would note the following.

753    The only direct evidence adduced by SNF as to the trial work undertaken at the Ernest Henry mine in 1999 was that given by Mr Schmidt in 2017, approximately 18 years later. That evidence suffers from the difficulties identified earlier, including that Mr Schmidt gave that evidence after viewing and relying upon a 2011 video, and by reference to his experience at Nalco over the period 1999 to 2009.

754    In any event, Mr Schmidt agreed in cross-examination that the trial work undertaken at Ernest Henry constituted “a small scale testand described that work as “a pretty primitive filter trial”.

755    Further, the evidence in some respects supports the proposition that the flocculant was added to the slurry as part of the Ernest Henry trial work in the form of an emulsion.

756    The Ernest Henry trial report prepared by Mr Schmidt recorded that the flocculant was added to the slurry before it was fed into a centrifugal pump: “[t]he slurry was then fed into the suction of a centrifugal slurry pump alone with the stacking aid”. Mr Schmidt accepted in cross-examination that if the flocculant was added in the form of neat emulsion, feeding the slurry into the pump would “slightly” activate or flip the emulsion, whereas it would “somewhat” shear the flocculant if it was added as a solution. The report also referred to the fact that “unactivated reagent was found in the treated slurry”, which would seem to indicate that some of the emulsion added to the slurry had yet to be “flipped” into solution. Mr Schmidt accepted this might have occurred:

So that leaves the possibility, doesn’t it, that some of the emulsion has been activated in the process of going through the pump and the pipe and some of it hasn’t been activated at all?---That could be – that could be assessment, yes.

757    That evidence accorded with his other evidence in which he acknowledged that flocculant mixed in aqueous solution … was already active”. In other words, the reference to “unactivated reagent” in the Ernest Henry trial report is likely to be referring to unactivated flocculant in the form of an emulsion. Thus, Mr Schmidt gave the following evidence:

But you can’t rule out, can you, that there was unactivated emulsion that was added into the hopper and which became activated during the process leading up to discharge at the end of the 20 metre pipe?---You’re correct, I cannot.

758    Further, Mr Schmidt gave evidence as to how the flocculant was administered to the slurry as part of the Ernest Henry trial work. That evidence revealed that once the flocculant had been administered into the top of the hopper the following process occurred before the slurry was deposited into the concrete bund:

[I]t recirculates through the hopper, down the pipe, through the pump, 20 metres, back to the top of the hopper, and you do that two or three times till you’re satisfied it’s mixed, and then you let it come out of the hopper, through the pump, down the 20 metres, and discharge into the concrete bund?---Into your designated area. Yes.

759    Mr Schmidt confirmed that it would take “a couple of minutes” to recirculate the treated slurry. And he did not deny that this would mean that by the time that the slurry was recirculated two or three times, the flocculant and the slurry would have been mixed for approximately between six and up to 10 minutes. But this process bore little resemblance to the process of improving rigidification taught in the opposed applications, which involves the dosing of flocculant in the form of solution to a slurry shortly before discharge. Rather, as Mr Schmidt explained, the process at Ernest Henry involved “a relatively low shear environment compared to the shear forces that typically occur in the pipeline of a thickener underflow on its way to a deposition area often kilometres away”.

760    Further, as BASF points out, the approximately six to 10 minute mixing time used at Ernest Henry significantly exceeded the optimal 10 to 20 second mixing time for aqueous solution identified by Dr de Kretser, or the optimal mixing time of “a few seconds” identified for an aqueous solution by Mr Schroeter. It instead fell within the optimal mixing time for emulsion identified by Mr Schroeter, namely, two to 15 minutes. As BASF points out, it is difficult to reconcile the results achieved at Ernest Henry with the evidence of Dr de Kretser that “[e]ven a 15 second mixing time can be enough to result in over shearing of flocs formed from aqueous polymer addition”. It is likely that the only way that the results depicted in the Ernest Henry trial report could have been achieved, given the method of application and mixing times involved, was if the flocculant had been added as an emulsion.

761    But in any event, the objective of the trial work undertaken at the Ernest Henry mine was simply to restore the slurry to the form and degree of flocculation it had when exiting the thickener, comprising a series of discrete aggregates. Moreover, no attempt was made by Mr Schmidt to incorporate a co-disposal step into the process being trialled at Ernest Henry, in the sense of adding a separate stream of coarse material to the single stream coming out of the processing unit, which was made up of varyingly sized materials.

762    In other words there was no co-disposal element in terms.

Boral Stapylton

763    Boral Stapylton was a sand quarry located in South-East Queensland where the OreBind process was trialled in August 2002.

764    SNF relied on the evidence of Mr Carl Buckland, who witnessed the OreBind process being performed in a trial at Stapylton and the implementation at Stapylton for several months after the trial. Mr Buckland also gave evidence about the OreBind PowerPoint presentation, which recorded the trial of the OreBind process at Stapylton.

765    Now as to this PowerPoint presentation, one slide headed “OreBind at Work!” noted that its features demonstrated:

    Granular mud structure

    Significant Beaching

    Delta formation and free water release

    Clear supernatant

766    But another slide headed “Where to from here???” listed the options as:

    Larger beaching pond

    Single sand screw

    Single sand screw / Rotary screen

     [T]wo sand screws in series

    Two sand screws in series / Rotary screen

767    Implicitly it would seem that the chemical treatment itself was not sufficient. Why else refer to “larger beaching pond”? Further mechanical dewatering devices were contemplated as being necessary.

768    I would also note that a prior Nalco memorandum of nine pages dated 18 June 2001 authored by Keith Gibbs including a specification prepared by a project engineer referred to the “first stage involves testing a sand screw for its suitability as a dewatering unit” (i.e. OreBind with a modification).

769    The specification stated as the background:

This specification details the equipment proposed for use at Boral Stapylton for the purpose of improving the handling of underflow solids.

Currently, underflow from the existing thickener is treated with anionic and cationic polymer prior to being pumped to a sludge pit. Settled solids in the pit are periodically excavated and loaded on to trucks whilst free-water is removed via a submersible pump. Boral staff have indicated that this is a costly and inefficient method of dealing with underflow solids and have expressed a desire for a process to be devised that will reduce costs and improve thickener operating efficiency.

770    The solution proposed was that the underflow circuit be redesigned to:

(a)    allow the thickener to run at maximum capacity;

(b)    remove the necessity for an excavator and operator during normal operation.

771    But the circuit diagram clearly shows a screw classifier mechanical dewatering device being added as a substitute to the digger.

772    It is not in doubt that at Stapylton, the process which was trialled and subsequently implemented, involved the following:

(a)    Thickener underflow was pumped through a pipeline into a silt pond.

(b)    Flocculant was added in aqueous solution into the tailings about 2 metres before point of discharge.

773    From sometime in 1999 until October 2000, Boral attempted to improve the tailings disposal process operating at the quarry. The existing process was problematic because tailings were deposited into a silt pond, and then excavated onto a truck every few days for disposal elsewhere, but the tailings were very watery and the silt pond was full of slimy water. None of the solutions trialled by Boral solved this problem.

774    In my view the process initially trialled by Nalco at the Stapylton quarry resulted in traditional settling and sedimentation of the deposited tailings, not improved rigidification, and was “costly and inefficient”.

775    First, it is apparent from the evidence given by Mr Buckland who participated in and provided the only direct evidence regarding the trial work at the Stapylton quarry, that the tailings discharged into the tailings pond did not have the relevant rigidified structure. Mr Buckland agreed that the treated tailings were “quite watery”, were “segregating” as they fell out of the pipe, “splattered” on the wall of the pond, and were “watery and segregating”.

776    Second, the process trialled at the Stapylton quarry involved the deposition of the treated tailings into a silt pond already containing water and the gradual build-up of those solids from under the water:

Now, just to clarify what happened to the water as part of this process, when it was being run, the water stayed in the silt pond; is that correct?---When it was – yes. That’s correct. Or some of the water went into the back of the truck when we emptied the silt pond…

And so what was happening then is you would deposit the material which had been treated into the pond full of water?---Yes. That’s correct. The pond was not full of water, but yes, there was always water left in the pond.

Yes. And after a period of time, the deposited solids would build [u]p above the surface of the water?---That’s correct.

And when they were first deposited, they would of course fall to the bottom of the pond under the surface water?---That’s correct.

777    Third, photographs recording the typical way the solids looked after deposition revealed that the discharged tailings were quite flat. Mr Buckland gave the following evidence:

[T]he spread of solids through approximately half the pond as depicted in the top left photo of 2018 would also be consistent, wouldn’t it, with a process in which there was a gradual settling of deposited solids?---That’s correct.

778    Fourth, the polymer was added to the tailings approximately two metres before discharge. The evidence of Dr de Kretser was that “typical pipeline velocities might have been between two to five metres per second”. So, if a typical pipeline velocity is assumed, this meant that the polymer had about half a second to one second of mixing time before the tailings were discharged, whereas the evidence of Dr de Kretser was that “in the case of aqueous addition, the peak level of flocculation (and therefore viscosity) would occur at a time of order 10 to 20 seconds”. Moreover, the short mixing time was compounded by the fact that the material was “dropped from a significant height”.

779    I accept BASF’s contention that these deficiencies i.e. point of addition of polymer and dropping tailings from a height into a pond with water, ensured that improved rigidification was not achieved. The tailings spread throughout the pond before building up above the surface of the water.

780    Further, to the extent that Mr Buckland asserted that the deposited tailings he handled were “a homogenous mix of the particles”, in cross-examination he confirmed that he only handled the solid (and not water) materials excavated, he did not conduct an analysis of the particle-size distribution of the excavated materials, and he did not test whether a sedimentary layer developed at the bottom of the pond. Moreover, to the extent that Mr Buckland asserted in his affidavit that he may have been able to stand in the pond, he confirmed in cross-examination that he never tried this. The deposited material might not have been able to support his weight.

781    Fifth, Mr Buckland confirmed in cross-examination that an essential component of the process was the use of a digger, which would excavate a combination of the solids and the water from the silt pond, and place the excavated solids and water into trucks for final disposal elsewhere as part of the process. This process was confirmed by Mr Buckland in cross-examination as being both costly and inefficient.

782    Ultimately, the option that Nalco recommended, consistent with its focus on the use of mechanical dewatering equipment, was “testing a sand screw for it’s suitability as a dewatering unit”. This required further trial work which, as Mr Buckland explained, would have only been necessary if Boral didn’t believe what was in place was sustainable or cost effective.

783    In essence, Nalco recommended that Boral use a sandscrew as part of the process being trialled, because the tailings deposited in the silt pond were not being adequately dewatered. In this respect, Mr Buckland gave the following evidence:

So the critical change between the existing process and the new process Nalco is recommending is the use of a screw classifier instead of the digger to excavate the material?---It appears that in the diagram, yes.

And I want to suggest to you that the reason why that’s being proposed is because the material when excavated by the digger wasn’t sufficiently or adequately dewatered, and that’s why a screw classifier is being proposed to replace it?---You could assume that, yes.”

784    This interpretation accorded with the recollection of Mr Schmidt that he had visited the Stapylton quarry and observed the use of the OreBind process at that quarry in conjunction with the use of mechanical dewatering equipment.

785    Sixth, no attempt was made by Boral or Nalco to incorporate a co-disposal step into the secondary dosing process being trialled at the Stapylton quarry.

786    Finally, the trial work undertaken at the Stapylton quarry did not constitute or evidence common general knowledge at the priority date. Nalco requested that it be paid a process development fee because it was developing the process. And Mr Buckland did not disclose the trial work to anyone outside of Boral.

Wemen

787    Wemen was a mineral sands mine in northern Victoria which was operated by Cable Sands as part of a joint venture where the OreBind process was trialled in August 2002. Further, the OreBind process was trialled in a laboratory at Cable Sands’ offices in Bunbury, Western Australia in May 2002.

788    There is a dispute as to whether the OreBind process used at these mines resulted in improved rigidification. But irrespective of whether improved rigidification was achieved, the evidence demonstrates that Nalco addressed the problem by adding a second dose of flocculant in aqueous solution to the thickener underflow close to the end of the outlet pipe. Further, Dr Farrow accepted that if the people implementing the OreBind process had wanted to achieve improved rigidification, they could have done so by increasing the dose at a suitable dose point.

789    I would note that there were some documents in evidence concerning Wemen.

790    For example, a site visit report dated 25 July 2002 authored by Ron Corcoran of Nalco appeared to disclose the simulation of a secondary flocculation process, with the flocculant in aqueous solution. It noted:

OREBIND Flocculation of Slimes Dam #1:

The trial commenced July 4th but encountered several delays with equipment problems and lack of dilution water, these problems were rectified and Orebind is now starting to show good progress.

About 30% of the volume in Dam #1 has been treated and, successfully de-watered. The return clean water is being pumped off to Dam #4 for reuse in the dredge pond.

The best steady state running is achieved with a constant flow of mud to the Orebind application point with a slimes load of 20-25% prior to flocculation. Mud flow of around 50-60 M3/hr and a polymer dosing pump setting of 60 Hz.

Maintaining a constant flow through the inline mixers is proving difficult due to the variability in the incoming mud. Constant vigilance on the flow readout is required and controlled manually opening or closing the knife valve, Bill seems to have mastered this. I’m sure he’ll recommend a good way to semi-automate it with in line density control to flow.

The flocculant primary solution is making up at a constant 1 %.

Water flow is 140 litres/min and polymer flow set at 1.4 litres /min.

Secondary dilution is now set at 350 litresd/min ~ 21 M3/hr = 0.1 % solution.

Dose rate is still variable and our last setting yesterday was 50 Hz ~ 40 Umin with a mud-flow 61 M3/hr and solids around 25%.

Dewatered runoff is rich in carry over flocc and is being reused as slurry dilution.

Based on the above numbers and an estimate of the volume treated the program cost is running around $4 to $5/tonne and above what our lab work indicated. However, cost efficiencies should come down as the application improves.

The dewatered slimes are beaching with an angle of between 15-20% and the mud solids after 2 weeks look fairly high (est. 80%)

I will start taking samples next week to get an average moisture content and indicative drying rate.

Finally, I recommend the mud be spread in windrows pointed down hill to maximise the dewatering and stop pooling.

791    A copy of this report was forwarded on to Heather Hutcheson on 5 August 2002.

792    I must say that the quality of the photos in the report was poor and difficult to verify against some of the report’s remarks.

793    It was noted in a later email that “solids after 48 hours [were] around 48% and 75% after 5 days” indicating enhanced and advanced rigidity. Further, in a later email after two weeks what was reported was “high 75-85% solids” and a “[b]each angle with cottage cheese flocc [of] around 20 degrees”.

794    Now SNF did not adduce any evidence from any witness with personal knowledge of the trial work undertaken by Nalco at Wemen. Indeed, the only fact witness who even referred to the Wemen trials, Mr Holtzman, confirmed that he was not involved in those trials and was not even aware of the trials when they were being undertaken. I should say that I do not have great confidence that the hearsay documents adduced in evidence regarding that trial work constitute an accurate or complete record of that trial work.

795    Further, to the extent that the results of some of that trial work were, after the event, communicated to Mr Holtzman, those results did not enable Mr Holtzman to successfully implement, as a matter of routine, a secondary dosing process achieving improved rigidification. To the contrary, the evidence of Mr Holtzman was: “And these emails certainly didn’t put you in a position where you could just go and implement the process at one of the mines?---No. There’s – no. There’s not a process there that would have worked for – for us in that situation”.

Bunbury

796    The laboratory work undertaken by Nalco at Bunbury was unremarkable. Mr Holtzman, who participated in that laboratory work, gave evidence that:

(a)    the testwork was limited to laboratory work and there was no field trial or trial of any process;

(b)    the treated underflow was never transferred to a disposal area and no attempt was made to deposit or stack such material on top of other treated material;

(c)    no attempt was made to combine the thickener underflow with coarse tailings as part of the laboratory work; and

(d)    it was not known or proved, after the laboratory test, whether the OreBind process would work.

797    The laboratory work at Bunbury involved treating material with flocculant in a beaker full of water, in which the treated material would sink to the bottom with the water on top, the treated material would have a flat surface, and then water from the beaker would be manually removed. As BASF pointed out, that is a classic settling process.

798    Further, to the extent that the first affidavit of Mr Holtzman purported to set out, in quite some detail, conversation(s) he had with a Nalco representative at the time of the Bunbury trial work, it became apparent during his cross-examination that he did not have any actual recollection of the specific conversation(s) alluded to:

And can you recall any particular conversation you had with him?---Not anything in particular. No…

So the extent of your recollection – it’s fair enough, given how long ago it was, was that you had a conversation where he said we’ve got this OreBind process, and we might be able to incorporate it into the co-disposal process you’re investigating?---Yes. Plus we did that laboratory testwork.

Other Nalco evidence

799    Let me deal with some other evidence concerning Nalco.

800    First, there was material before me concerning work done by Nalco for the Bayswater coal preparation plant. Nalco required Bayswater Colliery Co Pty Ltd to sign a confidentiality agreement, which it did. In relation to Bayswater, Nalco proposed two dosing points upstream of a mechanical dewatering device (see the report dated 8 November 1999).

801    A Nalco report dated 11 January 2000 in the executive summary and introduction stated:

A new development chemical program utilising Nalco’s Orebind® process aid technology was evaluated on Bayswater Coal Preparation Plant thickener underflow. The aim of the program was to remove free and intercellular water from the thickener underflow material and produce a high solids material capable of being conveyored with reject, rather than being pumped to a tailings emplacement area.

The first stage of the evaluation was undertaken in December 1999 with optimisation of both the Orebind® chemical program and the dewatering process being undertaken.

Initial results indicate an increase in underflow solids from 23% W/W to 52% W/W with further optimisation of the Orebind® treated solids dewatering process required.

The Orebind® process aid technology was developed to optimise the efficiency of the coal thickening underflow dewatering process. Developmental work was undertaken at the Bayswater coal washery during 1999 and a preliminary evaluation undertaken during December 1999.

The Orebind® process aid chemical program consists of two chemical programs that effect release of water from the high rate thickener underflow stream. A line was run from the pressurised thickener underflow line to the Orebind® equipment and approximately 20 - 30Umin of underflow solids were conditioned with the Orebind® chemical program during the evaluation.

Once the underflow material has been conditioned with the Orebind® process aid program, water removal was undertaken to maximise the solid liquid separation efficiency.

A Contra Shear rotary drum screen (0.5mm aperture) and a vibrating Honert screen were utilised to remove the water liberated from the coal tailings material after application of the Orebind® program.

The preliminary evaluation was undertaken to both determine the most efficient Orebind® process aid dose rates and also to review the efficiency of the dewatering processes.

802    The conclusion stated:

The Orebind® chemical program effectively conditioned thickener underflow for a subsequent water removal process. Substantial free water was still evident within the conditioned material exiting the Honert and Contra Shear screens and there exists an opportunity for further and more efficient dewatering.

The Contra Shear rotary drum screen was not suited to the dewatering application although it may have effected more efficient dewatering had the screen aperture size been larger or the solids detention time been greater. The Honert screen also did not remove all available water during the screening process.

It is recommended to evaluate two additional dewatering unit processes. These are:

    A CMI slow speed centrifuge, currently used for evaluation purposes at the CSIRO Ryde Dewatering Technologies Group laboratories

    A Contra Shear screw press

The Orebind® chemical program will be utilised during the next evaluation at the dose rates determined during the December evaluation. This dose rate, together with the program cost performance will be optimised and quantified once a suitable dewatering method has been selected.

803    Clearly mechanical dewatering was contemplated.

804    In a further report dated 12 May 2000, reference was made to the works in the second stage of the trials with various mechanical dewatering devices having been used, namely, a Contra Sheer rotary drum screen, a Vibrating Honert screen, an inclined rotary screw press and a variable speed drive basket centrifuge. In a yet further report dated 21 November 2000, reference was made to further mechanical dewatering devices to be reviewed.

805    In a much later internal email circa 2002 authored by Stephen Kelly to which Mark Crosbie responded, he said the following:

Following the conclusion of the OreBind trial at Bayswater in 2001 due to plant closure over the christmas period I will share my observations on the project. From this point I need some input regarding the direction of the project.

The OreBind project to date has proven two things;

The chemical program works, I would say we are 95% there, and further chemical optimisation will only improve the process slightly.

The dewatering process needs further refinement to give the final result we are looking for – a spadable product

At Bayswater I evaluated three dewatering devices; a high frequency vibrating screen and two cyclones. The screen worked to an extent, while the cyclones did not retain any solids as a cake product. Further investigation into the lack of success on the cyclones led to a better understanding of the process as it stands at Bayswater.

The tailings + chemicals needs a lot of mixing, and benefits from some quiescent mixing and pre-dewatering (excess surface water removal). The product being presented to the dewatering devices at Bayswater was not what we envisaged. This problem was not noticed earlier due to a lack of visual confimation in the scaled up process, where the OreBind material is sent directly to the dewatering device and only the cake is visible product. At the pilot scale the OreBind material was sent to an intermediate tank (an auger feeder) prior to the pilot dewatering device (visible product and centrate).

The “OreBind” material we want is a gravelly agglomeration that has a strength to it, this is the material we saw on the pilot scale and all the testwork has been done to acheive. At Bayswater the material was not at this stage, and had only produced slimy floes with no strength before dewatering. Being presented to the cyclone the material was splashed and splattered inside (visually confirmed). At the screen a proportion of the material was retained due to the low available screening area (poly screens vs wedgewire cyclones), this subsequently formed the “right” OreBind material from mixing on the water laden section of the screen and dewatered to a reasonable quality cake. The material that passed through the screen (slimy floes) was followed to a ground floor sump which contained about a tonne of the most perfect OreBind material, clear water discharge of the top due to self compression and a dense underflow from the sump discharge (acting like a secondary thickener). This material cannot be pumped as the shear liquifies the material back to thickener underflow consistency, but could be gravity or auger handled to a dewatering device.

From these observations one thing was made very clear - the OreBind material must be adequately prepared (and visually confirmed) to be adequate prior to dewatering. This may take one or more of the following;

better in-line mixing

feed conditioning box

pre-dewatering (e.g. sieve bend or wedgewire feed launder) prior to dewatering device.

I would recommend all three personally. However, at Bayswater the first option is the only one we can realistically purusue with current plant and OreBind dosing equipment. This leads to – what’s next.

806    Clearly there were problems with dewatering and that only a mechanical solution was being sought concerning dewatering.

807    Second, this theme for the necessity or a high preference for mechanical dewatering devices was the theme of the minutes of a meeting dated 23 March 2001 concerning a project for the Pacific Mining Group involving Nalco whose representatives were present at that meeting. The minutes recorded in item 3.2:

RW – Each person involved with Orebind has a different view on what Orebind actually is. MC, IP and KG all offered up their requirements for the criteria of: chemistry, equipment, control, current customer opportunity, current customer timing and perceived cost of flocculant and equipment. The details were stated as follows:

Chemistry: Dosing of solution anionic and solution latex cationic polymer in to underflow. It was acknowledged that Roger Strickland currently has an application at EHM where only neat anionic polymer is dosed.

Equipment/Process: In general terms, it was acknowledged that preparation equipment, dosing pumps and mixers were required in the majority of cases. Dewatering equipment would also be required in the majority of cases. It was acknowledged that Roger Strickland did not require preparation equipment or dewatering equipment in his current application but that would be the exception rather than the rule.

808    The minutes record that various dewatering equipment options were tabled with the “Screw” achieving the most votes.

809    All of the above material well demonstrates that OreBind was used or contemplated as being used preferably with a mechanical dewatering device, with any secondary flocculation upstream thereof. In my view, none of the Nalco OreBind evidence assists SNF to establish a lack of inventive step. The context for this process was more the use of mechanical dewatering devices. Moreover, co-disposal was not the central or significant theme.

Other mine operators’ investigations

810    SNF submits that other mine operators had undertaken laboratory and field trials before the priority date using SDITB processes including in co-disposal processes.

Yoganup

811    An example of SDITB being undertaken is the trial work performed by Iluka at Yoganup in 2000.

812    In 2000 a trial was undertaken by Iluka at its Yoganup mineral sands mine in Western Australia.

813    A technical report for stage 1 was prepared in October 2000 titled “Minesite Tailings Co-disposal with Flocculant Addition”; it was co-authored by Mr Brett Wroth and Mr J Warnock (the Yoganup report).

814    The Executive Summary set out the following:

1.0    EXECUTIVE SUMMARY

1.1    OBJECTIVES

This report details the test work proposed in and performed as Stage 1 of the project: Tailings Codisposal Using Flocculants (Wroth, 2000). Its objectives were to:

    Evaluate a tailings co-disposal option to lower the non-segregating threshold of deposited sand/slime mixtures by adding flocculant.

    Determine the effect of flocculant addition on the deposition beach angle (and its sensitivity to flocculant dosage) of the test mixes used in Stage 1.

1.2    SUMMARY

A test circuit was assembled and operated at the Yoganup mine site to test the objectives.

Samples of sand/slime mix were taken to determine solids content, mix ratio and flocculant dose rate for each test.

Table 1.1: Test results.

Test #

mix ratio

(wt. % sand/slime)

solids

(wt. %)

flocculant dose

(g/t)

beach angle (degrees)

max.

min.

1

-

-

-

9

4

2

81/19

58

855

18

16

3

76/24

60

0

8

5

4

77/23

33

1544

7

4

5

77/23

54

478

17

10

a – dose equivalent of active flocculant (grams dry powder) per tonne of dry slime.

1.3    CONCLUSIONS

It is concluded that:

    Adding flocculant to a segregating sand/slime mix will lower the segregation threshold of the deposited mix.

    Adding flocculant to a non-segregating sand/slime mix will increase the beach angle of the deposited mix.

    The flocculant dose rate used appears to exceed that required to produce a measurable effect in the deposited non-segregating sand/slime mixes.

815    A description and diagram of the test circuit was the following with flocculant in aqueous solution being used:

Test Circuit

Stage 1 of the project: Tailings Codisposal Using Flocculants (R00013) was basically a proof of concept exercise to test the effect of flocculant addition on stacking (and its sensitivity to flocculant dosage) of sand/slime mixes. It used a circuit with controlled input of dry sand and slime. This enabled close control of the sand/slime mix ratio and the solids concentration of the mix presented for pumping.

The circuit used consists of a dry sand stockpile feed onto a conveyor discharging into a mixing tank. Controlled flows of slime (thickener underflow) and make-up water also feed into the mixing tank. The mixing tank includes a dual impeller agitator to mix the sand/slime batches. Slurry (particle) suspension is maintained by re-circulation with a Warman centrifugal pump. The 3 m3 batches of sand/slime mix were pumped with a mono pump to the discharge point where dilute flocculant was mixed with the discharge.

Figure 3.1: Schematic of the test circuit used in Stage 1.

816    The following results were reported together with their interpretation:

5.0    RESULTS

5.1    Measurements

Results obtained from each test run are reported in Table 5.1.

Table 5.1: Results of tailings co-disposal tests.

Test #

mix ratio

(wt. % sand/slime)

solids

(wt. %)

flocculant dose

(g/t)

beach angle (degrees)

max.

min.

1

-

-

-

9

4

2

81/19

58

855

18

16

3

76/24

60

0

8

5

4

77/23

33

1544

7

4

5

77/23

54

478

17

10

a – dose equivalent of active flocculant (grams dry powder) per tonne of dry slime.

5.2    Observations

Images 4 (Test 2) and 5 (Test 3) illustrate the differences observed between the deposited mixes with and without flocculant addition respectively.

General observations of the deposited sand/slime mixes:

    Initial dewatering of the stacks was rapid and clean. No segregation or slime washout was observed.

    The initial dewatering rate appeared higher for mixes with flocculant addition.

6.0    INTERPRETATION

6.1    Test 1

Test number one was the first trial of the circuit after its commissioning on water only. No measurements of the sand, thickener underflow or water inputs were made. Its purpose was to confirm the ability of the test circuit to operate on the expected mixes and conditions planned in Stage 1 of this test program. In this respect the test was successful.

6.2    Test 2

Previous test work on Yoganup samples suggested the maximum achievable beach angle, of a non-segregating 88/12 wt.% sand/ slime mix, was 3 to 4 degrees at 58 wt.% solids (Williams, 1999). Test two (81/19 wt.% sand/slime at 58 wt.% solids) indicates that a significant improvement in beach angle may result from flocculant addition to the depositing mix.

The Williams (1999) prediction of beach angle takes into account the following:

    Sand/slime mix ratio, solids concentration and rheology.

    Rate and depth of tailings deposition over time of a plant scale trial.

    Void shape and size.

    Geological formation bounding the void (permeability).

    Hydrogeology (level and variation of water table).

Whereas, this test work is only small scale and estimates the effect of sand/slime mix ratio, solids concentration and flocculant addition. It is expected that at a large scale the beach angles reported (in this test work) would not be achieved.

On a plant scale the beach angle of deposited tailings is dependent on the tailings (sand/slime) slurry characteristics and deposition rate.

The sand/slime mix ratio and wt.% solids determine the segregation threshold and rheology of the tailings slurry. A non-segregating tailings slurry is necessary for it to stack with a planar beach angle. A concave beach results from segregating slurries, as hydraulic sorting will deposit sand at the beach head (discharge point) and allow slimes wash out to the toe.

The rate of deposition and slurry rheology (resistance to flow) determines the thickness of the depositing layer (of the tailings slurry) and hence the beach angle of the stacked tailings.

The effect of slurry rheology and deposition rate on beach angle was not addressed in this test work.

6.3    Tests 3 and 4

Tests three and four have similar sand/slime mix ratios, 76/24 and 77/23 wt.% respectively. They differ in that Test 3 (60 wt.% solids) is a non-segregating mix, whereas Test 4 (33 wt.% solids) is a segregating mix. It has been shown that, for Yoganup mine site, all sand/slime mix ratios below 35 wt.% solids are segregating (Williams, 1999).

The use of flocculant in Test 4 prevented the deposited mix from segregating. However, there is no significant difference in the beach angle achieved in Test 4 (with flocculant added) compared to Test 3 (with no flocculant added).

It is concluded that:

•    Adding flocculant to a segregating sand/slime mix will lower the segregation threshold of the deposited mix.

6.4    Tests 3 and 5

Tests three and five have similar sand/slime mix ratios (76/24 and 77/23 wt.% respectively) and solids content (60 and 54 wt.% respectively). Above 50 wt.% these mixes are non-segregating, the addition of flocculant (Test 5) clearly shows an improvement of beach angle.

It is concluded that adding flocculant to a non-segregating sand/slime mix will increase the beach angle of the deposited mix.

6.5    Tests 2 and 5

Tests two and five have sand/slime mix ratios (81/19 and 77/23 wt.% respectively) and solids content (58 and 54 wt.% respectively).

The addition of flocculant in Test 5, almost half that of Test 2, had no significant effect on the measured beach angle of the deposited mix.

It is concluded that the flocculant dose rate used appears to exceed that required to produce a measurable effect in the deposited non-segregating sand/slime mixes.

817    Dr Farrow accepted that:

(a)    the trial involved SDITB where flocculant was added in aqueous solution in a co-disposal process;

(b)    a coarse stream was added to the fines stream which had been treated with flocculant in the thickener;

(c)    the process involved adding the flocculant to the combined coarse and fine streams just prior to the deposition point;

(d)    the trial was to evaluate the benefits of that process to assist in avoiding segregation of coarse and fine materials;

(e)    the intent of the work was to evaluate whether adding flocculant to the co-disposed tailings could lower the segregation threshold;

(f)    the trial was to determine the effect of flocculant on the beach angle of the deposition;

(g)    the tests involved adding the flocculant to tailings with a solids concentration below 35% (these had been segregating);

(h)    the tests also involved adding the flocculant to tailings with a solids concentration of 54% and 58% (which was a high solids concentration tails slurry);

(i)    when flocculant was added beach angles of 17 and 18 degrees were achieved; these were “very high beach angles”; the addition of flocculant brought about a significant improvement in beach angles;

(j)    when the flocculant was added, initial dewatering of the stacks was rapid and clean water was released; this was consistent with the formation of a heaped geometry;

(k)    when the flocculant was added, there was no observed segregation of sand and slimes;

(l)    in a plant trial the beach angles may not be as high as 17 or 18 degrees but a heaped geometry would still be expected to be achieved;

(m)    in using the process in the field, the process would be optimised to suit the particular mining conditions;

(n)    in the trials, the desired result was achieved at a dose of 478 g/t (grams per tonne); and

(o)    there was a marked contrast between the nature of the treated and untreated material; the untreated material was less rigidified, had a lower beach angle and was not a highly structured or flocculated material.

818    A photograph of the deposited tailings extracted from the Yoganup report is reproduced below:

819    Now I would make the following observations.

820    First, the Yoganup report is marked with the notation that “This is a confidential document produced for internal Iluka use only”. There was no evidence that Ms Herzig (of CRL) or Mr Clarke or Mr Cigulev (of Iluka) were even aware of the existence of that report at the time of the trial work at Yarraman in late 2002. Nor was there any evidence that any person outside of Iluka had read the Yoganup report before the priority date. Indeed, although Ms Herzig had obtained a copy of the Yoganup report in around mid-July 2003 she considered the document to be confidential and did not annex a copy of the report to her affidavit. Ultimately, a copy of the Yoganup report was only obtained through a subpoena issued to Ms Herzig during trial.

821    Second, the report was a “technical report” written by Brett Wroth, who was employed by Iluka as a “research scientist” within the “R&D department”. The report recorded the results of confidential research being undertaken by the R&D department of a substantial mining company.

822    Third, and in any event, the Yoganup report recorded that the results obtained in the trial undertaken would not scale up: “Whereas, this test work is only small scale and estimates the effect of sand/slime mix ratio, solids concentration and flocculant addition. It is expected that at a large scale the beach angles reported (in this test work) would not be achieved”. Further, the report only revealed that the use of flocculant reduced the segregation threshold marginally.

823    Fourth, as the substantial and expensive trial work subsequently undertaken by Ciba, CRL and Iluka between 2002 and 2004 revealed, the Yoganup report fell short of providing a tailings disposal solution to Iluka.

824    Now I have also reviewed various 2003 documents concerning Yoganup, but I do not consider that they assist SNF. They are after the priority date. Moreover, I would observe the following.

825    In July-August 2003, a scope of work was done for a large scale trial at Yoganup involving a process described as Viscosity Modified Blended Co-Disposal (ViMCoD). In a version of this scope of work in August 2003 prepared by Dr Clarke it was said in the introduction and the aims:

1 INTRODUCTION

Viscosity Modified Blended Co-Disposal (ViMCoD) involves adding a specially formulated polymer to a mixture of sand and fine tailings. The mixture is then discharged into the final tailings impoundment. The polymer is intended to prevent segregation and enhance drainage.

ViMCoD has been implemented at the Yarraman mine at CRL’s operations, but is only required there on an intermittent basis. The technique has not yet been used continuously at any mineral sands operation.

Iluka has for many years investigated tailings disposal by Non-Segregating blended Co-Disposal (NoSCoD). This is an enhancement of thetechnique in use at most sites (except in the SW Operations Division), where sand and fines are disposed of together, but then tend to segregate in the impoundment. The segregation leads to increased costs and planning difficulties.

It has been shown that eliminating solar drying and using NoSCoD would substantially reduce costs in the SW. ViMCoD may offer the same benefits but without the need for capital expenditure on sand dewatering and with a much more rapid consolidation and rehabilitation cycle.

The ability to prove up ViMCoD in a probably shorter time frame is also important. There is an immediate need to decide on what tailings disposal technique will be used at Yoganup West. The Douglas project also needs a proven tailings disposal· technique.

A large-scale trial of ViMCoD is a pre-requisite to potential adoption for production.

2 AIMS OF A TRIAL

The following are the aims of a large-scale trial of ViMCoD:

1)    Determine the time to develop a trafficable surface and the effects of depth on that time.

2)    Confirm that the modified sand:fines mix remains non-segregating when discharged at production rates.

3)    Determine the medium and long-term consolidation behaviour (time to consolidate and final volume) of the modified tailings mix.

4)    Determine the effects of the reagent on long term drainage rates, required to dewater beyond the end of the consolidation period.

  5)    Determine the mixing time and condition required.

  6)    Determine the effects of varying sand:fines ratio.

  7)    Determine the effect of varying water content.

  8)    Determine the effect of dose rate.

9)    Confirm that the process is operable at full scale, and how outcomes compare to those determined in the laboratory.

826    It was also said in the opening to section 3.11:

3.11 Effect of No Pre-Thickening

A very important question for the Douglas project is whether pre-thickening of the fines is essential to successful application of ViMCoD.

The lack of pre-thickening might affect ViMCoD in three ways:

    By greatly increasing the water flow rates, it may prevent stacking of the tailings. This might have the effect only of flattening the beach angle; or it may break up the floc structure and lead to complete failure of the method.

    Because the fines have not been pre-agglomerated, the viscosity modifier reagent may not be effective. This has not been a problem in laboratory testwork, but needs to be checked at full scale.

    The higher water content may change the reagent addition rate.

It is not possible to fully test ViMCoD without pre-thickening, at production scale, as installed pipeline sizes are to small. However, the following tests will be carried out…

827    Now clearly this was all research work. Further, they were exploring the need for two stages of flocculation.

828    Further, on 24 October 2003, Ms Herzig produced a Trial Report – Part 1. She reported that segregation had occurred in the Yoganup trial (section 2.2.5). In section 2.2.6 she said:

2.2.6 Reducing Segregation

Segregation that arises from remobilisation of fines by bleed water may presumably be reduced by reducing the amount of bleed water. That can be achieved by increasing the density of the tailings discharged into the void, by dewatering the sand over dewatering screens ad potentially using paste thickeners – or possibly Finlay or Jadair Settlers.

However, if these measures are taken, it is likely that in many cases the mixture would be non-segregating without the added expense of the rheology aid. It would then be necessary to ask if the other benefits of ViMCoD justified the expense.

Further dewatering would also eliminate segregation that arose from low reagent additions, provided that the mixture was naturally non-segregating. If the mixture were not non-segregating, then dewatering would not fully address this cause of segregation.

829    She then listed substantial further work (section 4) that would be required “to optimise the concept of ViMCoD”, and that a full production trial was recommended.

830    I would also note that as part of her report she said (section 2.2.2):

One way to eliminate the problems of segregating co-disposal is to make a non-segregating mixture. In many cases this can be done by dewatering the sand and/or the fines streams. However, when the clay content is low (the actual percentage depending on clay mineralogy), the required % solids is high and can result in a mixture which is difficult to pump. When the clay content is high, it is relatively easy to make a pumpable non-segregating mixture, but the material is then slow to consolidate.

Ciba developed their rheology aids to avoid these problems. The rheology aid is added to the pipeline containing the mixed tailings and increases the yield stress sufficiently to prevent segregation, without needing further dewatering. If the rheology aid is added near the end of the pipeline, there is negligible effect on pumping costs.

The degree of segregation occurring with VimCoD is therefore an important criterion of success, since the avoidance of segregation was one key driver for developing the process.

831    I do not see how any of this later work really assists SNF to show a lack of inventive step.

Douglas

832    I would also note that Iluka material concerning the Douglas operation in June to September 2003 was in evidence before me concerning the use and effect of Ciba’s rheology modifiers.

833    In a technical report dated 20 June 2003 Mr Scammell reported on a test program using these modifiers. In summary, the DPW-1-1329 was said to be “the most effective powder grade rheology modifier for treating combined thickener underflow and coarse tails”. He recommended:

We recommended that DPW-1-1329 be assessed on site at Yoganup when the Douglas tails disposal trials occur later this year.

Further development work is needed to better understand the mixing requirements for treating the unflocculated slimes in the combined tails sample.

Further dewatering test work can be performed once the tailings disposal method is confirmed.

Alternative Ciba liquid rheology modifiers may prove even more efficient than DPW-1-1477 and may warrant ongoing test work at both Eneabba and Yoganup sites.

834    In a scope of work dated 11 August 2003 for the Douglas Ore Piloting Trial prepared by Iluka it was stated in section 6.1:

6.1 Viscosity Modified Co-disposal (ViMCoD) Demonstration Trial

Ciba Speciality Chemicals recently developed some viscosity modifiers that have been designed to increase the viscosity of sand and fine slurry mixes to such a point that the mix is suitable for stacking and rapid dewatering. The technique has been named ViMCoD by Iluka.

This technology is sensitive to both Ciba and Iluka. All information contained within this scope on ViMCoD should be treated as confidential and not disclosed to any third parties as per the pre-existing Non-disclosure Agreement signed between Iluka Resources and Amdel Limited on June 26th 2003.

835    On 19 September 2003 in a technical report authored by Mr Scammell he recommended:

6.0 RECOMMENDATIONS

We recommended that DPW-1-1329 be assessed at the AMDEL site when Douglas ore is scrubbed.

We recommended that lluka consider the addition of DPW-1-1329 direct into a tailings pipe as an alternate disposal method to that originally proposed by Basin Minerals.

Due to the simplicity of this disposal method a mobile discharge pipe may be an appropriate means for disposal of the treated tails stream.

836    On 25 September 2003, Mr Scammell sent an email to Ms Herzig reporting that “[t]he powder rheology modifiers were diluted down to 0.25% then added to the slurry”.

837    In my view none of this material after the priority date substantially assists SNF to show a lack of inventive step.

Old Hickory

838    Dr Clarke also gave evidence that Iluka had also undertaken some trial work of SDITB in a co-disposal process at Old Hickory in the United States at least as early as January 2002.

839    There was scant material before me concerning Old Hickory.

840    In an Iluka memorandum dated 29 April 2003 the following was said:

SITUATION

The Virginian operations of Iluka mine ores with high clay content, in the range 10 – 40%. The tailings disposal system is designed around blended co-disposal of sand and fines, but historically, control of the system has been poor and serious problems have been encountered due to segregation of the clays.

COMPLICATION

With recently improved tailings system control, it was expected that non-segregating co-disposal would be achieved, but this has not occurred. If truly non-segregating mixtures are segregating on discharge, this would have serious implications for implementation of non-segregating blended co-disposal (NoSCoD) at Iluka. A visit to the Virginian operations to carry out testwork on site was therefore authorised.

At the same time, a potential opportunity was recognised to conduct a large scale controlled trial of NoSCoD in Virginia. A further aim of a visit was to evaluate whether and how this could be achieved.

841    It was also said:

7 RELEVANCE OF A TRIAL TO DOUGLAS

It is clear from the data collected during this testwork, together with that determined by Golder, that the Virginian clays are very different in behaviour to the Douglas and Yoganup clays. Atterberg Limits also show that the Virginian clays have much lower plasticity. They appear to dewater more easily certainly than Douglas clays, they result in much higher non-segregation limits (i.e. higher densities required) and it would not be surprising if they consolidate and drain more rapidly.

In these regards, the Yoganup clays are much more similar to the Douglas clays and in fact plot almost on the same curve for segregation limit vs fines content.

A trial of NoSCod at Yoganup would therefore probably give results closer to those that would be achieved at Douglas.

However, the aim of the large scale test is not to directly mimic performance at any one future site, but to demonstrate that NoSCod can be achieved consistently in practice and that the results can be predicted with adequate accuracy. In that respect, a trial at Concord should be as effective as at Yoganup.

A trial at Concord also has the benefit of being in a high clay deposit in rich farmland, where swell is a major issue. This does help to highlight at least some of the issues that will be faced at Douglas.

842    Further, in an email dated 4 August 2013 from Mr Scammell it was pointed out that there was a difference between mineral sands tailings and alumina tailings, and therefore the testwork needed to be different.

843    Nothing concerning what was occurring at Old Hickory assists SNF to establish a lack of inventive step.

Beenup

844    BHP implemented SDITB at Beenup from 1996 to 1999. Flocculant was added in aqueous solution 10 metres before the discharge point and at the discharge point. Dr Farrow witnessed SDITB being performed at the Beenup mine. Dr de Kretser was also aware of SDITB at the Beenup mine, although he did not personally visit the mine.

845    But I agree with BASF that the process trialled at the Beenup mine bore little resemblance to the process claimed in the opposed applications.

846    First, although it was stated in the affidavit of Mr Cigulev that “[t]he process used at Beenup was a co-disposal process whereby separated waste streams of coarse and fine materials were recombined for disposal” and also that “[t]he process used at Beenup was a co-disposal process”, this evidence was incorrect.

847    Mr Cigulev confirmed, when cross-examined, that the process implemented at Beenup involved removing sand, rather than adding it:

So what you did was you didn’t add sand to the fines; you took sand away; is that correct?---In this instance, yes…

[T]he process was inefficient and sand was in there due to that inefficiency. So it wasn’t a conscious act. In fact, we consciously tried to remove the sand, but we – the device we had to do that, it was not an efficient device.

Yes. Thank you. So that you didn’t at Beenup take a stream of slimes or clay, a separate stream of sand, and combine them, did you?---Initially, we did and we found that it didn’t work…

But you accept that there was not a conscious combination of sand added to a stream of fines?---No. In – in – yes, there was no – in fact, we were trying to remove the sand. So there was no conscious – you know, decision to add sand

848    Second, the process trialled at Beenup was a traditional settling and sedimentation process. So much was apparent from the evidence of Mr Cigulev to the following effect:

And, here, you describe depositing the treated materials into a tailing pond filled with water. Do you agree?---Yes, that’s correct.

Now, that tailings pond or dam was a water-filled dam analogous to a thickener, wasn’t it?---Yes…

[T]his tailings dam was a large dam, with a high wall. You would agree with that?---That is correct, yes. I would say that. Absolutely.

Yes. And it was filled with water?--- Yes. Correct.

So it would take a long time for solid material to build up from the floor of the dam to the top of the wall?---Not long enough, in hindsight, but, yes, it would.

Yes. And how it would work is that the tailings discharged into the dam would settle down to the floor of the dam and gradually pile up?---That’s correct, yes

849    That evidence made plain that the tailings area was full of water, like a thickener, with the tailings deposited into that water gradually rising up. To the extent that stacking occurred, this was because, as Dr Farrow explained, it was well known that sand in tailings will accumulate close to the discharge point and form stacks.

850    Third, the process trialled at Beenup did not work. Now Mr Cigulev gave evidence that there were many reasons why the mine closed, but he confirmed that one of them was tailings disposal. Indeed, when Dr Farrow visited that mine in 1998, during the period when the process implemented by Mr Cigulev was in operation, what he observed was a tailings disposal process with as he described it “very poor operating performance”.

Londonderry

851    The mine operator at Londonderry implemented SDITB adding the flocculant in aqueous solution at two points in the tailings pipe between the hydrocyclone and the deposition area from 1980 to 1985.

852    Mr Woolley took photographs of the deposited tailings. Mr Woolley’s photograph that I have reproduced showed the deposited tailings approximately one week after deposition had ceased:

853    There were a number of problems with the evidence adduced by SNF as to what happened at the Londonderry mine between 1980 and 1984.

854    First, the evidence adduced by SNF was given by the same two witnesses, Mr Woolley and Mr Coleman, who had previously given evidence regarding the work at Londonderry in the 2008 proceedings. But their evidence in the present proceedings had some significant differences to the evidence they gave in the 2008 proceedings.

855    A comparison of the affidavits given by Mr Woolley and Mr Coleman in the 2008 proceedings with those given in the present proceedings reveals that the present affidavits given by Mr Woolley and Mr Coleman contain a significant amount of further evidence and detail concerning events occurring in 1980 and 1984 respectively.

856    Second, both Mr Woolley and Mr Coleman confirmed that they had given this new evidence without obtaining any further materials which assisted them to recall events occurring more than 30 years earlier and even though Mr Coleman had understood that it was important to provide detailed evidence in the 2008 proceedings. This makes this new evidence of problematic probative value.

857    Mr Woolley and Mr Coleman visited numerous mine sites in the decades after their work at Londonderry. The evidence of Mr Coleman was that he was “making fifty or sixty visits a year, possibly more than that” to mine sites over a 40 year period and “one of those mine sites was Londonderry”. The evidence of Mr Woolley was that he had visited “a number of mine sites”, not all of which he could remember, because “age takes one’s memory, to some extent”. In these circumstances, I cannot have great confidence that Mr Woolley and Mr Coleman were not in giving evidence before me unintentionally reconstructing what they thought might have occurred at Londonderry between 1980 and 1984 rather than truly recollecting what happened at that mine.

858    Let me dwell on this risk of reconstruction.

859    Insofar as Mr Coleman was concerned, the evidence established that he had between giving his evidence in the 2008 proceedings and these proceedings, reviewed photographs of trial work undertaken by Ciba at CRL, photographs of trial work undertaken by Ciba at Ardlethan, the Bulga video and the 2011 OreBind video, and given detailed evidence regarding those photographs and videos before the delegate in 2014. But when Mr Coleman was cross-examined about this before me, he had no recollection of viewing the CRL photographs or the Ardlethan photographs and he was adamant that “I’ve never seen a – an Orebind technology”, even though he had given six pages of affidavit evidence concerning the relevant video before the delegate.

860    Insofar as Mr Woolley was concerned, he could not explain how he had recalled some aspects of the new evidence included in his affidavit in the present proceedings (“I don’t know. I’m being honest. I don’t know”). Moreover, he confirmed that other aspects were based solely on his recollection of events occurring more than 37 years ago and conversations he may have had with Mr Coleman over the years. Further, Mr Woolley did not prepare, and was not able to identify the origin of, the diagram annexed to his affidavit or the text appearing in that diagram, which purported to depict what was described in his affidavit as “the Londonderry Process I installed in 1980 at the Londonderry Mine”(emphasis added). To the contrary, his oral evidence was:

So you were just provided with the document at tab 22.2, and all of the text appearing in that document - - -?---None of that’s mine. I – I – I – this is all – I would say this is all – would have been done by probably Ron, or it’s Ron’s – it – certainly I didn’t do this. It was just put in the affidavit to … represent the process out there.

861    It is likely that Mr Woolley and Mr Coleman were before me reconstructing what they thought might have happened at Londonderry by reference to photographs, videos and documents they had subsequently reviewed, and conversations with one another, and by reference to the numerous mines they had visited since 1980 and 1984 respectively.

862    Further and as BASF points out, the unreliable nature of the new evidence given by Mr Woolley and Mr Coleman was demonstrated by various inconsistencies between the evidence they gave in the 2008 proceedings and that which they gave before me.

863    In the 2008 proceedings, Mr Woolley gave evidence that the tailings at Londonderry were deposited into a water filled area and the solids sunk to the bottom under the water:

And so the water sits in that pond and the discharge falls into that and the solids sink to the bottom?---Yes, and the water is obviously moving on because you’ve got a large surface area with a dam. The flow is naturally flowing on—

864    But in his affidavit in these proceedings, Mr Woolley deposed that the “tailings were deposited from the outlet pipe onto the sloped side of the tailings dam wall” and that “further treated material was deposited from the outlet pipe onto the solids which remained on the dam wall”. Not only was that evidence inconsistent with the evidence he gave in the 2008 proceedings, but it was inconsistent with the evidence given by Mr Coleman in these proceedings, who explained that:

The – the surface water that would be squeezed to the top would run off the dam, most of it, but then the dam would then be let – let sit for a while, and there would be a bit of water on top of the dam after that dam was no longer being used, which would dry out first, and then – the material underneath it would – would open up and crack and dry very quickly.

865    Further, in the 2008 proceedings Mr Woolley exhibited a diagram of the process which he said that he installed at Londonderry in 1980. In the present proceedings, he annexed a different diagram of the same process. There were a number of differences between those diagrams concerning the depiction of the pump relative to the hydrocyclone and a lack of clarity concerning the function of the relevant pump, and the depiction of the region of the dosing of flocculant. The same inconsistencies existed between the two diagrams put into evidence by Mr Coleman in the 2008 proceedings and these proceedings. Further, the diagram exhibited by Mr Coleman had a further inconsistency, namely, it depicted five different flocculant addition points (D1 to D5), rather than the three in the diagram in the 2008 proceedings (D1 to D3). There was a note on the later diagram, which was not on the earlier diagram, stating:

Flocculant Addition Points D4 and D5

These final points were changed whenever the pipework to the outlets changed position. The locations of D4 and D5 shown were typical for the O5 outlet.

866    Further, in the 2008 proceedings Mr Woolley described the contents of a conversation he had had with Mr Ernie Upton in 1980. In the present proceedings, decades after that conversation apparently occurred, and years after he gave his evidence in the 2008 proceedings, Mr Woolley provided a different description of the detail of that conversation, including that Mr Upton wanted “to improve the rate of water recovery at the Londonderry Mine” and that he “needed to clear the solid materials from the deposition area and to recover the water from the tailings dam more quickly”.

867    Further, in the 2008 proceedings Mr Coleman gave evidence that he had walked on the surface of the dam at Londonderry six weeks after deposition. In the present proceedings Mr Coleman deposed that “[a]fter 2-3 weeks the area could be walked on by humans” and that “I walked on the dried tailings only a few weeks after the outlet pipe was moved to a new area”. In cross-examination, Mr Coleman acknowledged that the new evidence he had given was wrong:

Your recollection in 2010 was, wasn’t it, Mr Coleman, that you walked on the dam after six weeks?---That’s correct.

And you didn’t suggest to the court when previously asked questions about this that you’d ever walked on it any earlier, did you?---On reflection, I don’t know that I did walk on it at the two or three weeks. I walked out to the dam, and I could see that you could walk on it and people were walking on it…

Page 12, the very first line?---I’m sorry. I walked on it – I said a few weeks. I should have said “six weeks” there, shouldn’t I, to be more consistent.

868    In the 2008 proceedings, Mr Coleman gave evidence that he thought that in the disposal area at the Londonderry mine “there were some coarser particles at the bottom, some finer ones at the very top”. In contrast, in the present proceedings Mr Coleman deposed that the deposited material was “a homogenous mix of coarse and finer particles” and that he “did not observe the build-up for a sedimentary layer of sand”. Not only was this inconsistent with his oral evidence in the 2008 proceedings, but it was also inconsistent with his evidence in cross-examination that he would expect some of the coarse sands materials to “escape the agglomeration process” and to “fall to the bottom a bit quicker”.

869    Further, and in any event, the process implemented at Londonderry was not a co-disposal process.

870    The evidence of both Mr Woolley and Mr Coleman was that a hydrocyclone was used at Londonderry to separate the sand (the coarse fraction) from the waste, which was “fine and watery”, and that it was the waste that was treated with flocculant and disposed in the tailings area. In this respect, Mr Woolley confirmed that the whole objective of the Londonderry process was to separate the sand from the fine and watery waste. As a result, Mr Coleman confirmed that the waste being treated at Londonderry was “much more watery” than thickener underflow, indeed, he likened it to “a thickener feed” and described it as “a relatively thin liquid solution”.

871    Moreover, both Mr Woolley and Mr Coleman gave evidence that the waste treated at Londonderry was a “solution of approximately 3% solids”. This very low level of solids concentration made it problematic that it would have achieved improved rigidification as taught in the opposed applications. As Dr Farrow explained in some detail, the process claimed in the opposed applications requires a sufficiently high initial concentration (as often is achieved in a thickener underflow) so that there is the prospect of being able to create a network structure through the addition of flocculant. That evidence of Dr Farrow accorded with the teaching in the opposed applications that “[t]he fine tailings or other material which is pumped may have a solids content in the range 10% to 80% by weight. The slurries are often in the range 20% to 70% by weight, for instance 45% to 65% by weight”. The evidence of Dr Farrow was also confirmed by the practical experience of Mr Schmidt, who gave evidence that:

In your experience, what is the lowest solids concentration in thickener underflow where you’ve effectively achieved rigidification?---I’m just trying to think. There has been some very low ones…I’ve got a feeling it’s somewhere in the 30 per cent to 40 per cent, I’m thinking. I’ve got a feeling it’s alumina industry that the lowest underflow densities I’ve seen

872    The evidence of Dr Farrow on this point was also confirmed by the findings of Mr Bembrick in the ACARP report, who wrote after a two year research program that:

The Underflow density needs to be greater than 35% wt/wt solids (SG > 1.20) for a suitable beach to form. Treating slurry with low initial solids content causes the slurry to move further away from the discharge point increasing the chance of floc shear and clarified water contamination (no beach).

873    Similarly, Dr de Kretser recognised that it would be advantageous to have a significant solids concentration rather than a minimalist solids concentration. Further, it is to be noted that when Iluka in October 2000 undertook confidential research into the use of flocculants to inhibit segregation many years after Mr Woolley and Mr Coleman had undertaken their work at Londonderry, Iluka did not trial the effect of secondary flocculation on tailings with less than 33 wt% solids.

874    Accordingly, and as Dr Farrow said:

[T]he low solids concentration (3 wt%) is insufficient for a networked structure to be formed since the particles are not in sufficient close proximity. In the case of Londonderry, the solids concentration was ~3 wt%, but this included a sand fraction, which means the number of particles per unit volume, and hence their respective proximities, would be even less favourable for network formation …

875    Further, Mr Woolley confirmed that the process at Londonderry resulted in “a watery discharge” because “it’s only 3 percent up the pipeline”, that there was a “watery depression” at the point of discharge, and that there was a watery channel running away from the point of discharge. Mr Coleman confirmed that little flocs compressed together as they settled in the tailings area at Londonderry, that water on top of the deposited material dried out first, and that the material underneath “would open up and crack and dry very quickly” thereby aiding evaporation. Mr Coleman considered this process was “pretty similar” to the flocculation process used in a thickener.

876    In my view, all of this evidence also accorded with the three photographs of the Londonderry process, which depict a very watery discharge of material that spread in a flat form over a large tailings disposal area as it cracked and dried out. This bears little resemblance to a process of improving rigidification as taught in the opposed applications.

Other evidence

877    There was other evidence of other trial work of SDITB which had problematic value. Apparently Ms Herzig attended a lecture by Professor Ralston about his work on SDITB at mineral sands and red mud mines in Western Australia testing flocculant added in solution in a co-disposal process, deposited onto a slope and achieving stable deposition of co-disposed tailings. I must say that I found this indirect and hearsay evidence of little value.

878    Further, apparently Dr de Kretser was also aware of Rio Tinto undertaking trials of SDITB adding flocculant in both aqueous solution and powder form at its Labrador City mine in Canada before the priority date. Further, by the priority date a plant trial of SDITB was apparently being developed by Rio Tinto for commencement in 2003. But again the form and content of this evidence meant that it had little probative value.

879    Further, SNF has also relied upon the prior art publications referred to later in my reasons in relation to s 7(3), which it says disclose the use of SDITB processes including in co-disposal processes. For example, SNF says that each of the processes described in the Pearson patent (1996) and the Condolios patent (1982) involved SDITB in co-disposal processes adding the flocculant in aqueous solution. I will discuss these later.

880    Further, SNF says that the opposed applications acknowledge that the practice of SDITB adding flocculant in aqueous solution was known before the priority date. But the opposed applications teach that those attempts had proved to be unsuccessful.

881    In summary, none of the above evidence in my view assists SNF to show a lack of inventive step.

(e)    Ciba’s trial work

882    Mr Bruce Caine QC for SNF submits that evidence of an inventor’s conduct is plainly relevant and admissible for answering questions relating to inventive step. For present purposes I am prepared to accept that evidence of an inventor’s conduct can be of secondary significance. As I noted in BlueScope Steel at [38]:

Fourth, there are examples of trial judges having treated as relevant and admissible, on the question of inventive step, evidence of what the inventor actually did. I do not need to dwell on these examples as I am prepared to accept for present purposes that what an inventor did or thought may have secondary significance or relevance. And indeed when one analyses such examples, it seems that the inventor’s evidence was used more to fortify the primary evidence and analysis on inventive step, consistent with the characterisation of secondary significance.

883    Accordingly, I can and should have regard to the evidence about how and why Ciba went about implementing SDITB at various mines including Yarraman, Sandalwood and Ardlethan and the reactions of senior people within Ciba when Mr Scammell and Mr McColl reported on their laboratory trial work at Yarraman.

884    Now Mr Bellwood, Mr Scammell and Ms Beveridge give evidence regarding their understanding of the work performed by Mr Adkins, Mr Gallagher and Mr Parker prior to August/September 2002 in relation to SDITB adding both powder and solution. Their evidence was that the work revealed that adding flocculant in solution in SDITB would be ineffective, they did not believe it would be possible to achieve effective mixing of the flocculant if added in solution near the end of the outlet pipe, and they were surprised by the results achieved in the laboratory testwork by Mr Scammell and Mr McColl.

885    But SNF says that it is clear that the understanding of Mr Bellwood, Ms Beveridge and Mr Scammell regarding Ciba’s prior trial work did not reflect the understanding of those persons directly involved in that work.

886    Now as at August 2002, Ciba had only tested SDITB adding the flocculant in solution at two mines in laboratory trials at:

(a)    Suriname in September 2000; and

(b)    an unknown location in the US in 2000.

887    Further, as at August 2002, Ciba had not undertaken any sighter trial or plant trial work testing SDITB which involved adding flocculant in solution. The only sighter trial and plant trial work of SDITB by this time, involved adding flocculant in powder form, and then only in relation to red mud tailings created in alumina mining where there were typically long transit times.

888    Ciba undertook laboratory trial work of powder and solution at Suriname in September 2000. A report dated 8 September 2000 was prepared by Mr Adkins and Mr Parker (the Suriname 2000 Report).

889    At Suriname, the flocculants tested were added both as a powder and as an aqueous solution. The testing was evaluating the influence of dose and residence time. Mr Bellwood acknowledged these are two important variables when evaluating the performance of a flocculant in SDITB, because the residence time has an important effect on whether or not the yield strength is degraded over time.

890    At Suriname the residence time for the tailings from the thickener to the deposition area was in the order of 60 minutes. It was common for tailings to encounter transit times of this order in the alumina industry. Tests were undertaken to simulate pipe residence times of between 10 and 137 minutes. Yield strength was measured at 10, 60 and 90 minutes. There was no testing of at mixing times less than 10 minutes.

891    The Suriname 2000 Report highlighted that the addition of powder was the product of choice where it was desired to add the flocculant at a point where the treated slurry would be in transit for approximately 60 minutes. As recorded in the report, the powder product had “a strong resistance to shear degradation and, as such, offers an extended effective working environment”.

892    The Suriname 2000 Report recommended dry powder for use at Suriname principally because of its ability to withstand shear degradation over long transit time.

893    Now Mr Scammell understood from the Suriname 2000 Report that some of the trials at Suriname involved adding the flocculant in aqueous solution and being subjected to 10, 60 and 90 minutes mixing times. Further, he understood that 10 minutes was the shortest time that was tested, and that if shorter periods had been tested, the test may have highlighted more benefits of the solution addition method.

894    With respect to the 90 minutes mixing time designed to simulate a residence time of 90 minutes, Mr Scammell accepted that if flocculant in aqueous solution was mixed with the tailings for 90 minutes, the yield stress and the slump height of the tailings on deposition would be affected significantly, as the Suriname 2000 Report showed. Mr Scammell accepted that he would not have been surprised that adding a flocculant in solution then subjecting it to 90 minutes of mixing time would produce a poor result. Mr Scammell accepted that the Suriname 2000 Report did not disclose to him that solution addition was ineffective, but rather that solution addition was not as effective as powder addition in extended transit times of 10, 60 and 90 minutes.

895    Similarly, Mr Bellwood gave evidence that someone reading the Suriname 2000 Report who wished to flocculate tailings and who was aware of the way aqueous solution performed in a pipeline would know that the transit time of 60 to 90 minutes would cause the flocs to rapidly form and rapidly degrade.

896    In summary, the laboratory tests undertaken at Suriname in 2000 demonstrated that powder was more effective than aqueous solution where there was a transit time of at least 10 minutes after the flocculant was added. But there was little in that work that taught anything about the comparative performance of the addition of powder versus solution where the transit time was less than 10 minutes after the flocculant was added.

897    I would also just note at this point that the Adkins paper presented at the Minerals, Metals and Materials Society annual meeting in Washington in February 2002, included results from the trials conducted at Suriname in 2000. Further, results from those trials appear to have been referred to in the Dymond paper.

898    Let me move further forward in the chronology. In October 2000 and May 2001, Ciba undertook laboratory trial work of SDITB adding the flocculant in powder form at Alcoa Pinjarra.

899    A range of tests were conducted using various mixing regimes to disperse the powder into the slurry, ranging from (alone or in combination) repeated cylinder inversions (a minimum of 10), high shear mixing for up to two minutes and tumbling for up to 120 minutes.

900    Mr Bellwood accepted that:

(a)    the documents recorded the results of slump tests conducted on the last washer underflow from Pinjarra on 11 October 2000 and May 2001;

(b)    the results were measured after mixing times after 15, 30 and 60 minutes; and

(c)    there was no test of flocculant added in solution.

901    These tests were clearly undertaken to simulate the long pipe line residence times at Pinjarra. There was nothing in the laboratory trial work at Pinjarra that taught anything about the comparative performance of the addition of powder versus solution. There was a comparative reference to brine, but this is not on point.

902    The following evidence was given by Mr Bellwood:

MR CAINE: If you go forward to 794, you will see that again, they’re Pinjarra slump tests, this time on 12 February 2001?---Yes.

And you will see there’s an AW50 number there?---Yes.

That’s a reference to – if you turn the page to 795, it’s Alclar W 50?---Yes. That’s- - -

And then if you go to 796, you will see some more slump tests on Pinjarra slurry in May 2001?---Yes.

Now, may I take it that – sorry. Were you aware of any of those slump tests in either October 2000, February 2001 or, in the last case, May 2001?--- No. I wasn’t.

If you go to 795, you will see that what is plotted in the graph, as we see on the righthand side, is that product in three forms: dry, in a PEG slurry and in brine. Do you see that?---Yes.

We’ve spoken about the slurry form of the product. Using calcium brine is the third method, is it not?---I – I know that that’s – from – from my subsequent knowledge of things that were evaluated at that time, I know they were looking at a calcium – calcium chloride or calcium bromide brine as a potential carrier.

Yes?---So I’m happy to assume that that’s what’s meant by “brine” – “brine” in this case.

And I will put to you for the moment – and we can deal with it later on – that using calcium brine is something that’s only useful in the alumina industry?---Well, the – the – the – the calcium has a strong potential to affect the overall solubility of these polymers, and so it would only be useful if there was something that would completely remove the calcium from the system when it was added and so the calcium potentially then wouldn’t inhibit the ..... solubility of the polymer.

And we see in the graph at 795 that the best-performing product was Alclar W50 when it’s added dry?---Again, it depends upon how you define best-performing product. But certainly it acts quicker than the PEG - - -

Yes?--- - - - and the rheology is – is slightly more sustainable on the long-term.

And you will see that the transit time that was considered – from the raw data above the graph – was at three or four points: zero, 15, 30 and 60 minutes?---Yes.

And what was not tested there is the use of any product in the form of an aqueous solution?---That’s correct, yes.

903    Further, laboratory and field trial work was conducted at Point Comfort, Texas being:

(a)    initial laboratory trial work (on an unstated date);

(b)    three “side stream” trials conducted in November 2000, February 2001 and May 2001;

(c)    “full scale tests” conducted in late 2001; and

(d)    slump tests on Point Comfort underflow conducted in April 2002.

904    Mr Bellwood accepted in relation to the laboratory trial work at Point Comfort that the tests involved the addition of flocculant as powder and there was no testing of the flocculant added in aqueous solution.

905    It is clear that the field trials at Point Comfort prior to May 2001 involved the addition of flocculant in powder form. Further, it would seem that the photographs in the Dymond paper make it clear that the trial work in the photographs was carried out at Point Comfort prior to May 2001. The presentation refers to the flocculant being added in a PEG slurry (i.e. in powder). The presentation also discloses that in the trial, the flocculant had 10 minutes in transit to the deposition area prior to depositing the tailings.

906    The Adkins paper also discloses the results of trials at Point Comfort. Mr Bellwood accepted that the Adkins paper disclosed that the optimum dose point for powder addition in the conditions tested was a residence time of 60 minutes, being a transit time frequently encountered in the alumina industry in the treatment of red mud slurries, where the deposition area was a distance from the refinery. Mr Bellwood accepted that the Adkins paper was referring to the addition of flocculant in powder “under conditions of extended conditioning”, which was a reference to extended transit time.

907    Now the Point Comfort trial work:

(a)    was directed to the Gallagher process which was designed for long transit times and the tests were consistent with evaluating the performance of flocculant over long transit times (10 to 90 minutes), which was suitable for powder addition but not solution addition; and

(b)    there was nothing in the trial work undertaken at Point Comfort that taught anything about the comparative performance of the addition of powder versus solution.

908    Further, there was laboratory trial work on red mud in 2000 in the USA. A three page Ciba document headed “Document 478” was originally discovered in the Canadian proceedings in respect of the Canadian patent. This document apparently records the results of laboratory trial work undertaken by Mr Gallagher in the US in 2000 on red mud. This document is the only Ciba document recording the results of any trial of flocculant added in aqueous solution in which the results were measured after mixing times of less than 10 minutes. The highest viscosity recorded for the aqueous solution was higher than the highest viscosity recorded for the powder flocculant.

909    Dr de Kretser’s evidence was to the effect that Document 478 discloses the following matters:

(a)    first, the addition of the same dose of aqueous and powder flocculant into separate samples of red mud slurry;

(b)    second, the mixing of the flocculants in the samples for times between 1.5 minutes and 60 minutes and measuring changes to the viscosity of the slurry over time after addition; those mixing times equated to transit times of between 180 metres and 7.2 kilometres based on a typical flow rate of 2 m/s;

(c)    third, the highest viscosity recorded for the aqueous solution being measured at 1.5 minutes, but no measurements of viscosity being taken for shorter mixing times; Dr de Kretser expected that the peak level of flocculation for aqueous solution, and therefore viscosity, would have occurred within 10 to 20 seconds;

(d)    fourth, the highest viscosity recorded for the powder flocculant being measured at 4 minutes; the highest viscosity recorded for the aqueous solution was higher than the highest viscosity recorded for the powder flocculant; and

(e)    fifth, after a period of mixing, the viscosity decreasing for both powder and solution.

910    Mr Bellwood’s evidence in relation to Document 478 was to the effect that:

(a)    it records the results of a laboratory test plotting viscosity against agitation time;

(b)    the agitation time was to simulate transit time in the pipeline;

(c)    a comparative evaluation of flocculants added in solution and powder was undertaken to determine their relative performance in maintaining the viscosity of a slurry;

(d)    the flocculants and the material being treated were agitated by a four blade lab mixer for the amounts of time recorded on the horizontal axis;

(e)    the flocculant added in solution achieved peak viscosity at 1.5 minutes;

(f)    the flocculant added in powder achieved peak viscosity at 4 minutes;

(g)    the flocculant added in solution achieved a peak viscosity higher than the peak viscosity achieved with powder;

(h)    the peak viscosity built up by the solution addition was lost as the agitation (or transit time) increased; and

(i)    the graph did not suggest that flocculant added in aqueous solution was ineffective.

911    Dr Farrow gave consistent evidence and accepted the following:

(a)    The test work involved adding the same dose of aqueous and powder flocculant into separate samples of the slurry and measuring the viscosity of the two suspensions as a function of mixing time after the flocculant had been added.

(b)    The reduction in viscosity over time was caused by the shear forces caused by continued agitation (the lab mixer).

(c)    It was known that when flocculant was added in aqueous solution, it flocculates very quickly “in a matter of seconds”.

(d)    The peak level of flocculation for aqueous solution would be at shorter times than 1.5 minutes and that it could be in the order of 10 seconds depending on mixing efficiency.

(e)    The difference in the two peak viscosity values was consistent with the more rapid development of structure where the flocculant was added in the form of an aqueous solution.

(f)    The graph demonstrated that significantly higher viscosities could be obtained by adding flocculant in aqueous form than for adding a flocculant in powder form, provided the mixing time was sufficiently short.

912    In summary, the difference in the two peak viscosity values was consistent with the more rapid development of structure where the flocculant was added in the form of an aqueous solution. The graph demonstrated that significantly higher viscosities could be obtained by adding flocculant in aqueous form than for adding a flocculant in powder form, provided the mixing time was sufficiently short.

913    Further, as I have indicated, the Dymond paper presented in May 2001 at the PTT conference alluded to Ciba’s trial work at Point Comfort and Suriname. As I have said, the paper was given on 10 and 11 May 2001 in Pilanesberg, South Africa and titled “Beyond Conventional Flocculation”. Dr Farrow attended the conference.

914    The paper reported on “New Disposal Techniques”. One of them was “Dry/Convex/Dome Stacking” (i.e. not sub aqueous). The benefits were said to be to minimise disposal area, improve liquor release and eliminate sand segregation. It was said that dry stacking was “not a new technique in itself, but a polymeric technique has been developed to enhance the materials that can utilise the benefits”. Reference was then made to the fact that for a slurry to stack (heap/dome) it required a threshold viscosity usually related to its yield stress. It then said:

It might be anticipated that thickener underflows would [inherently] be beyond this threshold. However, many slurries arise from a flocculation process and this delicate consistency is often lost through the higher shear forces of pumping to disposal.

915    The following was then reported:

New Disposal Techniques

For Dry/Convex/Dome Stacking

    Polymer treatment system builds paste structure

    Acts in transit to disposal area

    Soluble polymer added in dry form

    Predictable performance

    Optimum viscosity at point of stacking

    Increases stacking angle/liquor release

    Minimum capital outlay

A polymer treatment has been developed to enhance slurry structure such that an increased viscosity is provided at the point of stacking.

An agitation stage is required but utilises the prevailing pipeline mixing conditions.

Soluble polymer - not an absorbant [sic: should be adsorbant] - as liquor release is required at stacking.

Added dry- no extensive pre-dissolution or dilution of the slurry.

Gives controllable response to dosage and conditions.

Achieves the rheology required at the point of placement.

Yields increases in slump [angle] and liquor release is maintained.

Does not require a high capital installation.

This graph shows a typical viscosity response to polymer dosage and pipeline transit time. Applied as a solution the 200g/t treatment is not only impractical but does not achieve the same improvement. The effect is shallow and the viscosity quickly breaks down.

916    I would note that several points can be made.

917    As seen from the graph, what was presented as favoured was adding polymer (i.e. the flocculant) in dry form (as per the top three lines on the graph).

918    Further, it was reported that the solution treatment was not only impractical but did not achieve the same improvement (as per the bottom red line on the graph) as adding polymer in dry form.

919    But it is important to note that results in terms of yield stress were not reported for any period less than 10 minutes transit time except zero. In other words, for less than 10 minutes transit time, in terms of any increase in yield stress, adding polymer in aqueous solution may have had considerable advantages over any addition in dry form. As I have said in discussing Document 478, in evidence were graphs which showed improved performance (viscosity) of the aqueous form of polymer over the powder form at 1.5 minutes or less including obviously 10 to 30 seconds. In other words, all else being equal, shorter transit times (less than 1.5 minutes) favoured the aqueous form over the powder form in terms of increased viscosity and correspondingly higher yield stress. To put this in context, assuming a typical flow rate in a tailings pipeline was about 2 metres per second, a transit time of 1.5 minutes would correspond to a pipeline transit distance of 180 metres.

920    The Dymond paper presented on a technique developed by Ciba involving the following:

(a)    the addition of flocculant to thickener underflow;

(b)    one of the aims of the process being “eliminating sand segregation”;

(c)    enhancing the slurry structure such that increased viscosity was provided at the point of stacking, resulting in increased stacking angle and the release of liquor;

(d)    measuring the development of yield stress in the material comparing the addition of dry powder at various doses including comparing dry powder and solution; and

(e)    measuring the development in yield stress after mixing times of 10, 60 and 90 minutes.

921    The paper concluded that the equivalent dose of flocculant “applied as a solution…does not achieve the same improvement. The effect is shallow and the viscosity quickly breaks down”. But as I say and I do accept, the asserted benefits of powder over aqueous assumed that there were long transit times for the treated tailings after the point of addition. This is apparent from the results recorded in the Dymond paper that:

(a)    when measured at 10 minutes transit time (equating to 1.2 kilometres assuming a flow rate of 2 m/s), flocculant added dry developed a greater level of yield stress (of 55 Pa) and maintained that yield stress through to about 60 minutes transit time after which the yield stress started to reduce; and

(b)    after 10 minutes transit time, the yield stress of the material treated with solution was only 20 Pa and it slowly broke down thereafter.

922    Mr Bellwood accepted that:

(a)    the Dymond paper does not say anything about the effectiveness of solution at shorter residence times than 10 minutes;

(b)    the results do not reveal whether, when the flocculant was added in solution, it reached peak viscosity between zero and 10 minutes, and then declined to the point that is recorded at 10 minutes; and

(c)    the peak yield stress for solution may have been a lot higher than the peak yield stress for flocculant added as a dry powder.

923    Further, there are laboratory trials referred to in the Gallagher patent. There are four examples in the Gallagher patent all of which are laboratory tests which simulated transit times of 60 minutes by gently agitating the treated material in a lab tumbler for 60 minutes. There are no examples of field trials.

924    Now the Gallagher patent asserts that flocculant added as a powder is far better than flocculant added as a solution, suggesting that comparative trial work had been undertaken by Ciba. But I do accept that none of the examples in the Gallagher patent involved testing flocculant added in solution. Indeed, it can hardly be said that it teaches in favour of use of the aqueous form. Clearly it teaches in favour of the dry form.

925    The Gallagher patent pointed out an important difference between solution addition and powder addition. The viscosity of the material treated with powder does not increase or decrease as rapidly as material treated with flocculant in solution. Accordingly, the Gallagher patent made clear that one difference between dry addition and solution addition was the time required to obtain the rigidifying effect and the time in which it would break down.

926    Now Ciba’s then understanding of the significance of residence time in the use of powders and solutions is also reflected in an email on 19 November 2001 from Mr Parker to Mr Cameron and others at Ciba in which Mr Parker noted in relation to the Gallagher process that “the whole idea is just to rebuild yield stress that is lost through pumping when the flocculated structure is broken down”. Mr Parker noted that: “With dry addition we have continuous dissolution over 30-60 mins thus always have floc available to rebuild yield stress”.

927    What may be gleaned from the above material that supports SNF’s case? In my view, not that much. On any view the Gallagher patent was hardly teaching towards flocculant in aqueous form. But clearly this material demonstrated that the aqueous form may be preferable for very short residence times.

928    Let me at this point say something about Ciba’s “Standard Method of Test”.

929    The fact that Ciba’s trial work just discussed was directed to demonstrating the efficacy of the Gallagher process in the context of long transit times is reflected by Ciba’s “Standard Method of Test” (SMOT) for the use of Rheomax Enhanced Tailings Disposal, the first draft of which at least was issued on July 2002.

930    The SMOT sets out two different test methods for testing flocculant added to tailings inline either in powder form or as an aqueous solution.

931    In the case of powder addition, the method prescribed mixing the flocculant with the tailings and subjecting it to high intensity mixing for one minute, followed by subjecting the tailings to ongoing shear for a range of durations to simulate shear within a tailings transfer pipeline.

932    Contrastingly, for aqueous solution the recommended method involved adding the flocculant into a beaker containing the tailings, and then pouring the material backwards and forwards between another beaker to mix the flocculant into the slurry. Dr de Kretser’s evidence was that the beaker pour method for adding flocculant in aqueous solution was extremely gentle when compared to the high intensity mixing prescribed for adding the flocculant in powder form. This was accepted by Mr Scammell. Mr Scammell’s evidence was that:

(a)    he performed beaker pour tests when evaluating the addition of flocculant in aqueous solution to belt press filter feed before the priority date;

(b)    consistent with the SMOT, those beaker pours involved taking a sample of the slurry and then mixing it with the flocculant that was being used in the belt press, and then tipping the mixture from one beaker to another a specified number of times in order to simulate mixing; the mixture was then tipped onto a cloth in order to record the drainage of water from the material; and

(c)    consistent with the SMOT, the beaker pour method was a low shear way to mix the flocculant in aqueous solution with the slurry, and that the extent of shear imparted on the mixture was determined by the number of beaker pours.

933    Let me now turn to the question of Yarraman.

934    As at April 2002 Ciba was supplying flocculant to the CRL Yarraman mine under a long term contract. That flocculant was added to the thickener in an aqueous solution. Mr Scammell was the Ciba sales representative responsible for the CRL account.

935    In April 2002 Ciba initially set out to market the supply of flocculant as powder for use in the Gallagher process to CRL at Yarraman. As part of that exercise, in May 2002 Mr Scammell and Mr McColl undertook some laboratory tests on site at Yarraman adding a dry powder flocculant to thickener underflow. The thickener underflow comprising fines was not combined with coarse material for these tests, although combined thickener underflow and coarse tails were used in later tests conducted in August and September 2002, which included a test of liquid form flocculant just prior to discharge.

936    In order to test the effectiveness of the powder, Mr Scammell and Mr McColl added the powder to the sample of the tailings and subjected it to mixing times of between 10 and 60 minutes and then measured the slump angle. The sample was subjected to high speed mixing at approximately 1,500 RPM using an “on-off mixer” after the powder flocculant was added to the sample. This process subjected the powder to a lot of mixing energy. Mr Scammell’s evidence is that this was the standard practice he was instructed to follow when testing flocculant added in powder form.

937    On 22 May 2002, Mr McColl sent an email to Mr Adkins and others including Mr Scammell in relation to the May laboratory testwork noting that the results of those initial lab trials did not show any significant difference between the blank and the treated samples, and asked whether the samples should be given more mixing time, as well as whether there was an issue with the use in the trials of a thickener sample which was a few days old.

938    Mr McColl noted that the samples were mixed for up to an hour, but that the realistic transfer time at Yarraman would only be 10 to 20 minutes, being the time it took for the tailings to travel from the concentrator to the discharge point. The initial proposal was to add the flocculant as powder to the tails suction side of the pump in the concentrator on board the dredge. Mr McColl was concerned that mixing the samples for more than an hour would not be representative of the transit times at Yarraman.

939    By 22 May 2002, Mr Scammell understood that there was a problem with using the Gallagher process at Yarraman, given the short residence/transit time that was available at Yarraman. Mr McColl asked whether the mixing time could be reduced by making use of a product with higher solubility. Mr Adkins noted in his response to Mr McColl that solubility of powder can be improved by reducing its particle size, however this was not likely to be economically viable for CRL due to the increased cost.

940    Now at the time Mr Scammell and Mr McColl undertook these trials, Mr Scammell was aware that flocculant added in aqueous solution was less effective than powder addition, but only when long residence times were involved. Let me move forward.

941    In August 2002, Mr Scammell undertook some “very rough” bench tests on site at Yarraman with one test adding flocculant in dry powder form (DPW-1-1332) and one test adding the flocculant (DPW-1-1067) supplied by Ciba for use in the thickener as an aqueous solution. Ciba had been supplying DPW-1-1067 to Yarraman in commercial quantities since around 2000 by reference to this DPW number. DPW-1-1067 was a modified form of Ciba’s commercial flocculant Magnafloc X125, which Ciba manufactured to compete with Nalco’s equivalent emulsion based flocculant. The tailings tested were a combined slurry of thickener underflow and coarse material.

942    Mr Scammell reported on the tests in an email to Mr McColl and Mr Adkins on 19 August 2002. The tests were conducted in early August 2002, some time before he sent that email.

943    In order to test the effectiveness of flocculant added in powder, Mr Scammell added the powder to the sample of the tailings at doses of 12, 31, 40, 53 and 105 g/t and subjected the tailings to manual shaking for two minutes and some spatula knife stirring, and then observed the stacking and water release.

944    In order to test the effectiveness of flocculant added in solution, Mr Scammell added flocculant in solution at a dose of 80 g/t and used only gentle spatula/knife stirring to mix the tailings for 30 seconds. Mr Scammell accepted that this was a low energy form of mixing (consistent with the SMOT), which was different to the mixing he subjected the powder flocculant to in his laboratory tests in May 2002.

945    Mr Scammell understood that in order to meaningfully compare the effectiveness of flocculant added in powder form compared to aqueous solution, it was necessary to subject the two forms of flocculant to different mixing conditions. In particular, he understood that it was necessary to subject the flocculant in aqueous solution to gentle mixing and to measure the results after a short period of mixing, whereas for powder it was necessary to subject the treated tailings to intense mixing and measure the results after a longer period of mixing time.

946    Mr Scammell characterised his decision to test flocculant in aqueous solution as “quite random”. His evidence is that the flocculant being used in the thickener at Yarraman was available to him, so he thought he would “give it a go”. Mr Scammell accepted in his oral evidence that at the time of making this decision he:

(a)    appreciated that the transit time at Yarraman was insufficient for flocculant added as powder to work;

(b)    already had a lot of experience in using flocculant added in aqueous solution in thickeners and belt press filters;

(c)    had particular experience in adding flocculant in aqueous solution in the thickener at Yarraman;

(d)    was aware of the difficulties with transporting, storing and administering dry powder at Yarraman, and that these difficulties were greater for powder than flocculant added in aqueous solution;

(e)    knew from his work of secondary dosing in the use of belt press filters that flocculants added in solution had to be added close to the end of the pipeline in order to minimise shear thinning; and

(f)    knew from his previous experience in using belt presses at Mount Thorley that to get flocculant in solution to work in secondary dosing was a matter of looking at the chemistry of the flocculant, the mixing conditions, and conducting test work.

947    I accept that the knowledge that informed Mr Scammell’s decision to switch the form of the flocculant from powder to aqueous solution reflects the knowledge that the person skilled in the art could have utilised in the same or similar circumstances.

948    Now the results for both powder and solution addition were positive. The flocculant already used in the thickener on site (DPW-1-1067), which was added as a 1% aqueous solution and then mixed gently for 30 seconds, delivered results equivalent to dry powder addition after mixing for two minutes of high intensity mixing. Mr Scammell wrote to Mr Adkins and Mr McColl reporting on the results of those tests on 19 August 2002. Mr Adkins responded to Mr Scammell’s email on 21 August 2002. Just to be clear about his position, Mr Adkins was a research project manager who worked predominantly in the laboratory in Bradford in the United Kingdom, was involved in the trial work at Suriname in 2000 at which Ciba trialled powder and solution addition, was one of the named inventors of the Gallagher patent and, as I have previously said, an author of the Adkins paper.

949    Mr Scammell regarded Mr Adkins as a person who was knowledge