England and Wales Patents County Court

IN THE PATENTS COUNTY COURT

		Claim No: PAT 06026
		07 February 2008

Introduction^[1]

1. This is an action for infringement of patent ? 2 363 635 C ('the Patent') in the name of Qual-Chem Limited ('Qual-Chem') entitled 'Method of introducing additives in steelmaking'. The subject-matter of the Patent concerns a process for 'killing slag'. The process is used in the manufacture of quality steel known as 'interstitial free' ('IF') steel. I understand that IF steel is used in the manufacture of inter alia, motor car bodies.
2. Qual-Chem is a family-run private company which specialises in the provision of products and services to the steelmaking industry. It is based in South Wales. The named inventors are Mr Keith Harris, the managing director of Qual-Chem and his son-in-law, Mr Stephen Bray, Qual-Chem's product development manager, both of whom gave evidence at trial.
3. Infringement is denied by the defendant, Corus UK Limited ('Corus') who have counterclaimed for revocation of the Patent on a number of grounds. Alternatively, if there has been infringement of a valid claim by Corus, Corus rely upon the provisions of the Patents Act 1977, s 64 ('Right to continue use begun before priority date' - see below §7)^[2].
4. Until early 2000, Qual-Chem had been doing good business with Corus supplying inter alia slag reduction products to it and conducting slag reduction processes for it at Corus' Port Talbot works. It was during this time and as a result of experimental trials at Port Talbot, that Qual-Chem developed the process the subject of the Patent. As will be appreciated when the Patent is discussed later on, Corus' process requires quantities of aluminium for the slag killing and this was initially supplied to Corus by Qual-Chem. After the patented process had been set up, at Corus' request Qual-Chem taught the process to a third party. This proved to be unwise from Qual-Chem's point of view as Corus soon dispensed with their services (and its aluminium products) and from September 2002 until August 2006, used a process which Qual-Chem believes infringes the Patent. As I understand it, that is the origin of this dispute.
5. Corus is in fact the same legal entity as British Steel plc ('BSC') which changed from being a public to a private company in 1999. In 2000 it changed its name to Corus UK Ltd and this year, following the take over of its parent, Corus Group plc, by Tata Steel UK Ltd, Corus Group plc also became a private company. Corus is thus part of the multinational Indian Tata Group. In this judgment, for convenience, I shall refer to all the defendant's past activities as if they had been carried out by Corus – though some were in fact carried out in its BSC days.
6. I have mentioned that there is a counterclaim for revocation of the Patent. First, as a result of the experimental trials in September 1999 to which I have referred, there is an allegation of lack of novelty by reason of prior use of the claimed process at Port Talbot, Corus alleging that the use of the process before the priority date was not confidential and that a number of its employees were involved in those trials with Qual-Chem's Mr Bray. Corus say that those employees were 'free in law and equity' to make what use they wished of knowledge thus gained during the trials.
7. These trials have a bearing on the s.64 defence mentioned above. Corus says that since the trials were carried out using its employees and were to some degree paid for by it they must properly be regarded as having been its trials and not Qual-Chem's: see Defence §§ 4-9. They were therefore 'bespoke' trials for Corus alone and if (which is denied) any infringement of a valid claim is established, Corus is entitled to rely on this section in its defence.
8. There is also an allegation of lack of inventive step in the light of three prior publications and common general knowledge alone:

(i) US Patent ? 5 332 199 ('Knapp')

(ii) Japanese Patent ? H01-294817 ('Kawasaki')

(iii) A paper by Bommaraju et al entitled "Ladle slag treatment for production of ultra-low carbon steels' (Steelmaking Confce. Proc. 1993). The author, a Dr Rama Bommaraju, was in fact Qual-Chem's expert in this case. Mr Michael Tappin, who appeared for Corus, did not make use of this paper at trial to attack the Patent and I need not therefore refer to it in detail.

9. There has been an amendment of the Patent after grant and partly as a result, there is also a multi-faceted attack based on 'added matter'. It was first said that the matter disclosed in the Patent as granted extends beyond that disclosed in the earliest application and thus, that the invention claimed in the Patent is not entitled to a priority date earlier than its international application date (18 September 2000). The allegation is also made that the C specification has illegitimately 'matured' as a result of amendment post-grant from a much broader process which was the subject of the granted Patent. The scope of these objections was somewhat reduced before the conclusion of the trial but what remained was very much part of Corus' attack on validity.
10. Needless to say all these allegations were denied by Qual-Chem, for whom Mr Roger Wyand QC and Mr Richard Davis appeared.
11. All the claims are method claims, the apparatus claims of the Patent as granted having been amended out after grant. Fortunately, it was agreed that both infringement and validity could be adjudicated by reference only to claim 1 of the Patent.

Some Introductory Technical Background.

12. There are two main processes for making steel: basic oxygen steelmaking using a basic oxygen furnace ('BOF') and the electric arc process using an electric arc furnace ('EAF'). A simple description of the two processes is set out in a Corus brochure annexed to the Report of Mr Peter Pope, Corus' expert witness^[3]. The proceedings are not however concerned with the primary aspects of steelmaking as such but with what is called 'secondary steelmaking', that is, with what happens to the steel after it has been 'tapped' from such furnaces into ladles – in this case, to be made into IF steel.
13. At the priority date, it was commonplace to treat the contents of the ladle, that is both the iron and its supernatant slag, with a variety of known conditioning additives depending on the desired quality and properties of the steel product required. This is part of what Dr Bommaraju has called 'quality improvement and product optimisation'. To a layman this might well seem part of the skill if not the alchemy of successful steelmaking. My own impression is that stoichiometric accuracy in the addition of such additives was not the norm, many of them being added by a shovel. As applied for, the patented process subject of the present dispute was directed to treating both the steel and the slag with additives but the C specification is concerned only with 'slag killing'. I shall come back to this when I consider the validity of the Patent under 'Added Matter'.
14. I would add that conditioning by additives was performed not only on steel destined to become IF steel, but on many other steels as well^[4].
15. Considering next the treatment of slags, various slag 'killing' or treatment methods were known before the priority date of the Patent. Dr Bommaraju, Qual-Chem's expert, classes these methods into four categories^[5]:

The addition of aluminium bearing materials (or other such reactant) to decrease FeO and MnO.

The adjustment of density and selection of constituents of the slag conditioning mix.

The use of carbonates to create mixing and stirring during reaction.

The use of lime as an additive.

16. In his Report, Dr Bommaraju has also set out in an uncontroversial way, a number of technical facts of a general kind describing a typical steel-making process^[6]. I believe it would be useful to reproduce some of this evidence at this juncture:

(20) Slag treatment or slag conditioning, as it is most commonly known, is the process to reduce the FeO and MnO contents of the ladle slag. In order to reduce the iron oxide content of slags, a reductant such as aluminium or calcium carbide must be added to the slag.…

(23) The tapping event in the steelmaking operation is significant. The tapping stream of liquid steel causes the entire air blanket around the ladle to heat up creating noticeable convective flow of cold air towards the hot tap stream and movement of lighter hot air from the ladle. The slag treatment process is an expensive extension of the steel-tapping step and adds considerable dust, smoke, and safety concerns.

(24) The addition of aluminium to the ladle slag is done right after tap. It is important to ensure the addition of the aluminium bearing slag treatment material at the very end of tap so that the added material reacts with the slag and not the steel. The calcium carbide with or without additives is added, most commonly, by dropping sealed cans into the steel during tapping. There is also gravity fed calcium carbide systems, wherein the material is delivered from sealed containers (with nitrogen atmosphere) via launders.

(25) The most common way to add the aluminium based slag treatment mix is from a bin where in the material is fed by gravity. In some steel mills, the slag treatment mixture is delivered by dropping the 'super sack' of aluminium containing slag-conditioning mixture.

(26) The effectiveness of the slag treatment process varies considerably from heat to heat depending on FeO and MnO present in the arrival ladle slag. In most cases, the actual composition of the BOF slag is not available prior to slag conditioning. The addition amount is typically a standard quantity to bring the FeO and MnO contents down to manageable levels before processing the heat at either the LMF or the Vacuum Degasser. The effectiveness of the slag treatment is measured only after a few minutes of processing at the LMF/ Vacuum Degasser."

17. Mr Pope, Corus' expert, says that before the priority date, additives were either put into the ladle by hand (i.e. by shovelling or throwing) or more often gravity fed, via a chute and hopper or by a conveyor. There is also no dispute about this.
18. Metallic aluminium in a variety of physical shapes either alone or mixed with other additives had been used before the priority date for conditioning the contents of the ladle, both steel and slag, after pouring. On the other hand, aluminium powder, though commercially available, was the subject of prejudice for use in conditioning as it tended not to have a useful life in the aggressive environment immediately above the surface of the slag ('flaring off').
19. In chemical terms, the principal beneficial reaction which takes place when aluminium is added to the slag, is this:

2Al + 3FeO = Al2O3 + 3Fe

The Patent

20. The Patent bears unmistakable evidence of substantial application of scissors and paste; as a result of amendment after grant, it makes tiresome reading^[7]. This is evident from the poor English, the numerous gaps in the text, incorrect statements of 'preference', needless repetitions, bad punctuation and so on. Nonetheless, its teaching is clear enough: it is directed to an allegedly novel method for the secondary treatment of steel by the pneumatic treatment of slag (which is present in the ladle once the steel has been poured) with shaped aluminium additive. In other words, it is directed to and claims a method of 'killing' the slag.
21. The short preamble refers to prior art methods of conditioning steel by (for example) the gravity feed of additive from a hopper or by the use of an injection lance for its introduction. A US patent ('Yamada^[8]') for the conditioning of steel using an injection lance directly above the 'runner' from the furnace (i.e. vertically downwards with very little divergence of the injected stream^[9]) is mentioned in this context. The Patent later mentions that such lances are 'expensive and short-lived' (p 3, l6)^[10].
22. What is proposed is this: 'Metallic aluminium' additive in the form of 'pellets, tablets or briquettes'^[11] is pneumatically fired into and just below the surface of the slag in the ladle by a 'pneumatic conveying outlet', often referred to at the hearing (and in the Patent) simply as 'a gun'. The aluminium particles are conveyed and propelled from a nozzle by means of compressed air. The gun is conveniently mounted on existing plant structures and is fired at an inclined angle to the horizontal surface of the slag by an operative so that a divergent spray of aluminium particles penetrates into and remains in the slag. This method of treatment both enables the entire surface of the slag to be showered with particles and allows the resultant exothermic conditioning reaction within the slag to occur. The operative will 'tailor' the adequacy of penetration of the particles to his satisfaction by judicious balancing of the air pressure, the density and shape of the added particles and the quality of the slag. A number of benefits (including a cost benefit) are stated to result from the implementation of the proposal.
23. To my mind, this short patent is not difficult to understand. There is but one technical term which is used: 'killing slag'^[12]. Otherwise, the words used in the body and claims are ordinary English words which despite the eloquence of counsel, are I think, used in their ordinary sense. Before setting out claim 1 however, I should perhaps go into a little more of the detail of the disclosure as there has been a dispute concerning the meaning and scope of certain words in claim 1.
24. There are three 'process parameters' which the operative has to take into account in operating the process (p. 12, ll1-6):

(a) the particle size and density of the aluminium additive

(b) the thermal updraft from the molten metal in the ladle, and

(c) the desired penetration depth of the aluminium particles into the slag'

25. First, the Patent says this about the use of the angle of fire of the gun (p3, lines.9-16 and p 4, lines. 1-12):

The central axis of the stream is one about which the stream diverges to form a substantial divergent conical stream of pneumatically conveyed particular additive which impinges upon the slag in the form of projectiles. The first angle [of the gun] may be substantially horizontal or at an acute angle to the horizontal: it should not be vertical….When the central axis is at an acute angle to the horizontal, the additive may be directed toward the surface of slag in the receptacle. When the stream including the additive is directed towards the surface of the slag….the additive is preferably conveyed to reach below the aforesaid surface' [Emphasis added]. ^[13]

26. The following passages relate to the additive, whose density and composition are 'process parameters' for the operative to take into account (p 5, ll. 11-29):

'The additive is in the form of tablets, pellets or briquettes. The density and composition of such tablets, pellets and briquettes may be tailored in order to penetrate to predetermined depths in the slag at a predetermined rate. This enables the additive to be tailored to perform specific reaction requirements at specific depths and times. ..

The predetermined specific density of the particulate additive can ensure that the particles penetrate into and remain in the slag rather than descending into the liquid iron below) but resist flaring off on the surface.'

27. Regarding the additive itself, the Patent says this (p7, line.20- p8. line.28):

The additive comprises a multiplicity of shaped elements (i.e. tablets briquettes or pellets) which comprise metallic aluminium. Such shaped elements preferably comprise compressed divided material which form self-supporting individual elements. It may be beneficial to include calcium carbonate in the shaped elements….The shaped elements may include one or more non-aluminium materials preferably arranged to have a conditioning influence upon the slag. …Such additives may be bound in the shaped elements as divided material…in certain embodiments they may be distributed throughout a shaped body predominantly of aluminium'.

28. The proportions of such additives depend on 'application requirements' ( p13, lines l4-5). The density of the pellets etc is considered on p13, lines 9-15:

'The density of the relevant tablets, pellets or briquettes is pre-selected to meet the required performance characteristics.'

and examples of particle density are given.

29. The delivery pressure and velocity of the air

'can…be tailored depending on the 'sinkage' requirements of the additive being delivered and the upward thermal currents being encountered above the molten iron…'

and numerical examples of pressure and discharge rates are given.

30. Finally, this is what is said about the 'nozzle' (p6, ll19-22):

'Preferably the conveying outlet comprises a nozzle, preferably a diverging nozzle arranged to induce a diverging outlet stream which fans or diverges outwardly in a direction away from the nozzle.'

in this case however, no numerical indication of the required degree of divergence is given. In spite of the language used, this is not in fact a preference.

31. The Patent includes three diagrams but these do not I think materially add to what I have set out above, so I need not include them in this judgment.

Claim 1

32. The integers of claim 1 may conveniently be split up as follows:

A A method of killing slag on the surface of molten steel present in a receptacle

B in which additive material comprising metallic aluminium in the form of pellets, tablets or briquettes

C is gas conveyed to the slag in a conveying gas

D in an inclined divergent stream from conveying apparatus spaced above the slag surface

E the conveying gas pressure being tailored with respect to the conveyed additive material to cause the conveyed additive material to penetrate into and remain in the slag

G said conveying gas being air.

The Expert Witnesses

33. The expert witnesses played an important part in this case and I shall introduce them without further ado. Both experts were well qualified and both 'came as teachers'. I found their evidence to be both useful, frank and transparently independent.

Qual-Chem's Expert: Dr Rama Bommaraju

34. Dr Bommaraju is a metallurgical engineer. Having obtained a master's degree in physical metallurgy from Benares Hindu University (1977) Dr Bommaraju received practical experience as a trainee engineer in India and then, in the continuous casting of steel in the USSR. He then moved to Canada where he received his doctorate at the University of British Columbia (1988).
35. Dr Bommaraju entered the steelmaking industry in the USA and since 1988 occupied increasingly important positions within the US steel industry both as a steel maker and as a teacher. For the last 7 years he has run a consulting company, DML Steel Tech. Regarding the present subject matter he said this:

" I started working on slag treatment of ladle slags during my appointment as a senior engineer for process technology improvement at Inland's^[14] operations…I ran several trials of various recipes of slag treatment mixtures in various shapes for [IR] steels.'^[15]

Corus' Expert: Mr Peter Pope

36. Mr Pope is a self-employed steel plant operations consultant giving technical advice and training for EAF and secondary steelmaking facilities. He has a degree in metallurgy and has held various positions from shift manager to technical manager with Co-Steel Sheerness Ltd, Kent at various times.
37. Of the two experts, I have held Dr Bommaraju in rather higher regard for two reasons: first, he has had a broader variety of general steelmaking experience than Mr Pope and more importantly, he has had actual experience of killing slag – as the paper cited as prior art by Corus (and now abandoned) shows^[16]. Moreover his experience in steelmaking was more in the context of the BOF process wherein the subject of the Patent was born and developed^[17]. Also, as will be seen, I found Mr Pope's position on obviousness in the light of the common general knowledge alone, to be unsatisfactory.

The Skilled Addressee

38. The Patent is deemed to be read and understood by the skilled addressee who possesses the common general knowledge of the art. The concrete qualities attributed to this notional and important person in patent infringement cases, has frequently been commented upon and I have no need to repeat them here. In the end, I do not believe that there was much between counsel as to who was to be regarded as the appropriate skilled addressee in this case.
39. In my view, Mr Tappin's characterisation of the skilled addressee is about correct. He said that the skilled addressee in this case is a person involved in process and technical management of a steel plant and in particular, a person with hands-on experience of primary and secondary steelmaking processes. The skilled addressee is therefore essentially a practical person. Picking up on a memorably colourful phrase used by Dr Bommaraju, Mr Wyand suggested that he was a 'Regular Joe who makes steel'. To me, such a person would be more likely to be encountered in a Steinbeck novel than as the skilled addressee in a patent infringement action. The addressee in this case is in my view a relatively experienced worker in the industry, and not a novice. He is not a metallurgical scientist. And he of course possesses no inventive capacity whatever.

Common General Knowledge

40. In paragraphs 12-19 above I have set out some of the relevant common general knowledge about which there seems to be no dispute. There was nevertheless a dispute about the full extent of the common general knowledge since as I understand it, Corus contend that the contents of the Patent itself fell entirely within the common general knowledge at the priority date.
41. In particular, Mr Pope's evidence was that pneumatic 'guns' were in widespread use at the priority date for performing a number of tasks within a steel works (for example, for coating surfaces) and in a plant the size of Port Talbot (which I guess, having been there myself, must cover several square kilometres), that is indeed likely be so. Moreover, I have no doubt that guns of this type were ordinary articles of commerce at the time, and at trial some trade marks for such equipment were mentioned. However Mr Pope could not point to any evidence of their being used to 'inject' anything during any steelmaking operation: T4/571-572.
42. To gain some independent, contemporaneous insight here, it is interesting again to mention Dr Bommaraju's 1993 paper which Corus initially relied on in their counterclaim. The slag -killing solution proposed by its authors was to add the aluminium using two chutes, one each side, and to rely on gravity : see T4/492.
43. I have reviewed the evidence on this topic^[18] and am clearly of the view that the use of pneumatic delivery equipment of the sort under consideration in this case was not part of the common general knowledge of the skilled worker at the priority date.

Construction: General Principles

44. There was no discernable dispute between counsel as to the correct approach to construction: I must read the documents of the case as if I were the skilled addressee. I need therefore only record that I was referred to Terrell (16^th Edn) §6-101-107 and to Kirin-Amgen Inc v Hoechst Marion Roussel Ltd [2005] RPC 169. The fundamental question is: What would the skilled person have understood the language, particularly the language of the claims, to mean?

"Construction whether of a patent or any other document is not directly concerned with what the author meant to say. There is no window into the mind of the patentee…Construction is objective in the sense that it is concerned with what a reasonable person to whom the utterance was addressed would have understood the author to be using the words to mean. Notice however that it is not, as is sometimes said "the meaning of the words the author used" but rather what the notional addressee would have understood the author to mean by using those words….What the author would have been understood to mean by using those words is not simply a matter of rules. It is highly sensitive to the context of and background to the particular utterance. It depends not only upon the words…but also upon the identity of the audience he is taken to have been addressing and the knowledge and assumptions which one attributes to that audience." ^[19].

The Integers of Claim 1

45. Integer A: 'Killing slag'.

The meaning of this integer was not actually in dispute but as it is a technical term I should record my understanding of it. This term refers to a reduction in the level of unstable oxides in the slag – principally FeO and MnO. There is no suggestion in the Patent (or in the evidence) that any degree of reduction needs to be achieved or that all (or any particular amount of) the slag in a ladle needs to be 'killed'.

46. Integer B: "additive material comprising metallic aluminium in the form of pellets, tablets or briquettes".

(a) I have experienced no difficulty in understanding the meaning of this integer in the context of the Patent. The words 'pellets, tablets or briquettes' are not terms of art and are not elaborated on (save as to density) in the Patent. The initial and overall impact the words made on me was this: that the additive should be aluminium or substantially aluminium, that it should have a definite, visible shape (in contrast to say, aluminium powder) and that the shapes need not be formed with any precision.

(b) On foot of the passages in the Patent relating to this integer, some of which I have set out above, Mr Tappin submitted that what the skilled man would understand by this integer is that the additive material in addition to containing aluminium, must also be adjustable both as to shape, density and composition. As to shape, he submitted that no deviation from pellets, tablets and briquettes was permissible and in addition, that such shapes must always be made from divided aluminium so that their density could be controlled. Solid aluminium which contains no additive, he said, has an invariable density (and composition) – and for that reason, must be excluded.

(c) I have mentioned that a numerical range of particle densities is actually provided in the Patent and the density of metallic aluminium happens to be about in the middle of this range, some particles for use in the process being heavier and some lighter than that of metallic aluminium^[20]. Moreover, it is agreed that the aluminium is the component in the additive which performs the principal reductive reaction which is at the heart of 'killing' slag: see §17 above. Not surprisingly therefore, metallic aluminium is nowhere said to be excluded from the patented process

(d) In my view Mr Tappin's construction of integer B far from being 'purposive' verges on the perverse; this is not the way the skilled reader would understand this integer, the more so as (see above) the use of aluminium shapes for this purpose formed part of the common general knowledge. This submission will therefore be rejected. I also record that my initial impression (see above) in the light of the evidence has not changed either.

47. Integer D: "inclined divergent stream"

(a) Mr Tappin agreed that inclined requires the centre of the stream to be at an angle to the vertical. In addition, for the job to be done, it cannot be horizontal or even nearly so. But, he asked, by how much is the divergence of the inclined stream to be if it is to fall within the claim? He submitted that the degree of divergence required is such that the need for 'scanning' with the pneumatic gun is substantially eliminated; the particle footprint must substantially cover the entire surface of the slag in the ladle – 'in one go'. In this context, the experts made related contributions which I shall record.

(c) First, in understanding the degree of divergence, Dr Bommaraju was understandably referred by Mr Tappin to the US patent (Yamada^[21]) mentioned in the preamble to the Patent. He was asked questions about Figure 1 of Yamada, in which some divergence of the material being discharged from a nozzle is shown. The narrative in the Patent on the other hand requires that there should be 'very little divergence of the stream'. Dr Bommaraju stated that the relevant wording in the narrative and claims of the Patent was readily understandable. He said that the dimension HL in Yamada's Figure 1 is only a matter of 'a few millimetres'^[22]. In addition, he considered that the skilled reader would regard Yamada's Figure 1 as being schematic.

(d)In relation to Yamada's Figure 1, Mr Pope invoked the notion of what he called 'natural divergence' which, he said, always occurs in such circumstances. However, he hardly explained what 'natural divergence'^[23] was and seemed to ignore both the small dimensions involved in Yamada and the purpose of divergence in both Yamada and in the Patent. The issue may be resolved without much difficulty in my view, by considering the context of the use of the word in the Patent.

(e) First, the Patent clearly contemplates nozzles giving different degrees of divergence: see para 30 above. 'Natural divergence' (whatever that my be), and hence a divergence greater than a 'natural divergence' would I think, come into the reader's mind. Furthermore, I can nowhere find any suggestion that the degree of divergence must be so great that the particles cover all the slag at once without some movement of the gun ('scanning') Though this is desirable (p. 12, ll 15-16) it is not necessary (p.11, ll following). Contrary to Mr Tappin's submission, I do not believe that the skilled reader would think that the divergence contemplated must be so great as to eliminate the necessity for 'scanning' with the gun .

(f) The proper construction of this integer is evidently one of degree, this being an evergreen topic in patent infringement cases. As to that, it has often been said that the phrase in question must be read purposively, that is, in the light of the object of the invention: see inter alia, Henriksen v Tallon Ltd [1965] RPC 434. Furthermore, when one is dealing with claims making use of words of degree, it is I believe, all too easy to overlook the Protocol on the interpretation of Article 69 EPC, the last sentence of which requires the claims to be interpreted in a manner which

'combines a fair protection for the patentee with a reasonable degree of certainty for third parties.'

(g) I therefore reject Corus' submission on this integer. The court can and will make up its mind (invoking that of the notional addressee) as to whether a particular relative limitation falls within the claim in a particular case or not. It is a matter of fact and degree.

48. Integer E: " the conveying gas pressure being tailored with respect to the conveyed additive material to penetrate into and remain in the slag"

(a) The scope of the word 'tailored' was what gave trouble. Mr Tappin submitted that this clearly required the operative to exercise control inter alia over the gas pressure used depending on the circumstances (thickness of slag, particulate condition of the aluminium additive, the heat of the melt and thus the air updraft etc) so that the aluminium additive will penetrate into and remain in the slag. To achieve this, as I understood Mr Tappin, this integer requires the operative to exercise a more or less continuous monitoring and (if necessary) adjustment of air pressure to accord with in situ conditions; the air supply must be both first chosen and thereafter be controllable at all times, he submitted. He accused Qual-Chem of wanting to 'airbrush' out (or at least downplay) this integer, the requirement for 'tailoring'.

(b)The integer would not I think, be understood by the skilled reader in this way. I have already shown that there are a number of variables which have to be taken into account for the successful accomplishment of the slag-killing operation contemplated by the Patent. These are the air pressure, the shape, composition and density of the aluminium additive, the strength of air uplift over the hot ladle and the configuration of the nozzle of the gun. The suggestion that all these variables always have to be assessed, chosen, selected (i.e. the air pressure, the additive and the nozzle), and if necessary, varied either continuously or at least every time the process is used in a plant on every fresh batch of pig iron in the ladle, is to my mind, far fetched.

(c) Before commencing a slag-killing operation, the operative would no doubt have to choose (or be given) the particular aluminium additive to be used and perhaps, also a nozzle if one was not already fixed in the gun. He would note these and then select an appropriate air pressure and no doubt undertake some ranging shots with the pneumatic gun to ensure that the desired effect was achieved. The required penetration would hopefully then be achieved - perhaps even at once or at any rate, after some adjustment to one or more of the variables. As the evidence showed, success would to some extent actually be visibly apparent to the operative. Too low a pressure for example, might result in poor penetration with the additive flaring off (with bright light) on the surface of the slag. The desired effect might of course be achieved at once (which one would think, would be likely with an experienced operative). If for example, the operative's choice of additive was limited to what the process supervisor had given him – an 'off the shelf' product, say^[24], this to my mind would not sidestep the requirement for tailoring.

49. I therefore reject all Corus' submissions on the construction of integers B,D and E. I would add this: these submissions have I suspect, been chosen to chime with Corus' submissions on infringement –to which I shall turn after describing the alleged infringing process.
50. Before addressing other matters, I would however mention that Dr Bommaraju identified a number of advantages of using the process subject of claim 1 which I do not think were denied by Corus. These are:

"(a) The pneumatic force aiding the introduction of the reactant overcomes the upward draft of air on the ladle top.

(b) The pneumatic introduction of the reactant is quick and disallows the premature oxidation of the aluminium before entry into the slag medium.

(c)The injecting force, metered to a given slag condition, prevents unwanted entry of aluminium into the steel lying below the slag –thus preserving the oxygen content of the steel necessary for ensuing vacuum degassing operation.

(d) The tailoring of the injecting force also prevents the reactant prematurely burning off after falling and laying on top of the crusty slag's top layers."^[25]

The Alleged Infringing Process

51. What follows is an edited version of the Process Description [2/6] provided by Corus of the slag-conditioning process which was in use at their BOS plant at Port Talbot from September 2002 to August 2006 – after which the alleged infringement ceased in the light of Corus' adoption of a gravity-feed chute to introduce aluminium particles into the slag. The Process Description included three 'schematics' and also some photographs to illustrate what was happening. I have however included only Schematic B at the end of this judgment - as Annex A. Though it does not actually say so, I must assume that the process described in the Process Description sufficiently killed the slag and presumably, also yielded a satisfactory commercial product.
PROCESS DESCRIPTION

Aluminium Pellets

52. The aluminium pellets used in the process comprise pea sized 97% aluminium metal pellets formed into approximately teardrop shaped pieces by a melting and cooling process.

Conveying Gun

53. The Conveying Gun used in the process is shown in schematic form in Schematic A. The Conveying Gun comprises standard spraying apparatus that is and has been routinely used in the steel (and many other) industries for conveying a variety of materials in various processes for many years before September 1999.
54. Referring to Schematic A, the Conveying Gun comprises a batch fed container with a calibrated hopper placed above a container (a pressurised tank). The hopper and container are separated by a ball valve. A nitrogen compressor is attached and connected to the container. At the bottom of the container is a further ball valve separating the container from an air line. The air line comprises an armoured rubber hose approximately 6 cm in diameter that is fed with compressed air. The compressed air is taken from the general use compressed air line running through the plant. This air line operates at pressure which is uncontrolled save that it lies within the range of 3 to 6 bar. A nozzle is placed at the end of the air line, the nozzle comprising a 120 cm long generally straight steel tube of 6 cm diameter.
55. The nozzle is placed in an aperture in the wall of the surround to the BOS converter as shown in Schematic A. The pivots enable the nozzle to be pivoted horizontally and vertically allowing the operator to direct the stream of pellets to different areas of the surface of the slag. The nozzle is directed downwards onto the surface in the ladle, at a distance of approximately 5 metres from the surface (as depicted in Schematic A).

Process

56. The Process is used for slag conditioning, and in particular to remove oxygen from the Basic Oxygen Steel (BOS) carryover slag. The Process is used after conversion of the raw materials and scrap in the BOS furnace (also known as the converter) and once the molten steel has been "tapped" from the furnace into the ladle. A diagram of the steel conversion process is depicted in Schematic B. Once tapping is complete, the Process is applied to the slag layer floating on top of the molten steel in the ladle.
57. Referring again to Schematic A, the Conveying Gun is charged by the addition of 170kg aluminum pellets into the hopper. The upper ball valve is opened and the pellets are discharged into the container. The top ball valve is then closed. The nitrogen compressor then compresses the container with nitrogen. The pressure (approximately 3 bar) of the container used in the process is applied on all ladles and not adjusted or varied between them. Once the container is compressed, the lower ball valve is opened by the Process operator, the batch of pellets is discharged into the air line. The air pressure conveys the pellets along the air line to the nozzle.
58. The Process operator holds the nozzle as indicated in Schematic A. The pellets are discharged from the nozzle in a stream of pellets. The diameter of the footprint of the spray on the surface of the slag is approximately 600mm and, at any given time, the pellets generally impinge with slag extending over approximately 5-10% of the surface area of the molten steel in the ladle (the surface area of the slag in the ladle being approximately 12m²).
59. The Process operator views the slag surface through the nozzle aperture in the converter wall. Other personnel are able to view the process through an observation window. The Process operator directs or scans the stream of pellets across the surface of the slag, by moving the nozzle horizontally and vertically on its pivots, to ensure that the whole surface of the slag is covered by the pellets, and, if appropriate, concentrating the spray on areas of particularly thick slag.
60. The batch of 170kg of aluminum pellets is discharged in approximately 30 to 60 seconds. 170 kg aluminum pellets equates to about 0.12m³ of material. The pellets generally penetrate through the outer crust of the slag and remain in the slag layer. A small percentage of the pellets fail to penetrate into the slag and remain on the surface where they are subsequently burned away.

Infringement

61. In the light of the Process Description, Mr Tappin submitted that there were three integers of claim 1 which were not present in the allegedly infringing process. These are integers B, D and E which have already been considered under 'Construction' above.

Integer B: Aluminium pellets vs. Aluminium shot

62. A bag of Corus' additive is Court Exhibit 2^[26]. Since the Corus Process Description itself refers only to 'pellets' and since Corus themselves order their aluminium for the allegedly infringing process as 'pellets' [X1/8], Mr Tappin plainly had an uphill task trying to persuade me that Corus were using 'shot' and not 'pellets'. As the case progressed, it became clear from the experts that these particles could just as well be called 'shot' or 'pellets': see Bommaraju T3/464, Pope T4/566.

"I mean, I have no problem if you want to call these pellets. It does not make any difference to me."^[27]

63. Under 'Construction', I considered and rejected Mr Tappin's argument that the use of pure aluminium does not allow one to vary the density of the additive and must therefore be excluded from claim 1.
64. In the light of the foregoing, Corus' 'shot' falls within this integer.

Integer D: ' in an inclined divergent stream'

65. There is no doubt that the gun and the 'stream' issuing from IT is 'inclined': see Annex A. Corus' nozzle was 60mm in diameter; the diameter of the spray at the point when it reached the slag was 600mm. In terms of the nozzle diameter, this is an increase of 1000%.
66. The argument on this integer centred largely on what the experts said about the specification of Yamada which features in the preamble to the Patent. I have already considered these arguments.
67. In the light of these facts and my views on construction of this integer, I have no doubt that in Corus' process, the aluminium additive is conveyed into the slag in an inclined divergent stream.

Integer E: 'Tailoring'

68. The Process Description states that in carrying out the alleged infringing process at Port Talbot, the plant's 'in-house' compressed air line was used –without either antecedent numerical verification or adjustment of pressure. The compressed air in the works line varied somewhat but was at a pressure of 3-6 bar but was otherwise constant. Working the process by making use of the pressure available on this line was found to be sufficient to cause adequate particle penetration and thus, satisfactory slag killing – which was of course the object of the exercise. Thus, said Mr Tappin there was no 'tailoring'.
69. I have already rehearsed the relevant construction arguments on this integer. They are directed essentially to the necessity said by Corus to be latent in it to be able initially (and indeed continuously) to monitor and if necessary, control the relevant parameters so as to achieve the necessary initial and ongoing 'tailoring'.
70. In relation to the process in issue, Corus must have initially chosen inter alia what pressure to use – though I cannot find evidence as to exactly what happened. This choice would have taken into account the shape, constitution and density of the aluminium additive to be used – and the other variables as well – the positioning of the gun and the nozzle configuration, etc. In the same way, they chose to make use of the compressed air available on site since, AS I understand it, it was found to be adequate for its purpose 'straight off', without adjustment of pressure by say, a regulating valve in the compressed air line. That was both convenient and obviously sensible. The choice implicit in the word 'tailored' in claim 1 does not I think, demand anything in the nature of a formal pre-production and/or ongoing research programme; it is a case of 'first try it and see'; if one is lucky first time, so be it. But that does not eliminate choice on the part of the process operative. In the process described in the Process Description, there was, in my judgment, 'tailoring' within the meaning of claim 1.

Conclusion

71. For the foregoing reasons, if claim 1 is valid, it has been infringed by the Corus process.

Validity

Self-anticipation by prior user : Qual-Chem's tests at Port Talbot

72. It is agreed that the process trials that led to the application for the Patent were conducted at Corus' Port Talbot works in three stages. The first took place from 9 August till 6 September 1999. Ten days later, Qual-Chem made their first application for the Patent, the earliest priority date being 16 September 1999. Then, on 2 October 1999 there were more experimental trials. The third trials took place in December 1999. It is common ground that the second and third trials (only) did make use of the method subject of claim1.It is also common ground that all these trials were witnessed only by employees of the parties to this litigation. The trials were devised by and were all under the control of Qual-Chem's Mr Bray – and after Mr Bray's feedback, by MR Harris as well.
73. There was a good deal of evidence as to what happened at these trials but basically, Qual-Chem contended that at the first trials, the process failed because most of the aluminium additive did not penetrate the slag but flared off at its surface instead. The method of claim 1 was therefore not used. If that is so, this attack must obviously fail in limine because the other trials took place after the priority date. If on the other hand, Corus establish that the patented process was used at the first trials, a further question arises as to whether an obligation of confidence nevertheless existed such as to make it unconscionable for an employee of Corus who knew about these trials, to have communicated his knowledge to a third party. This is a question of law and fact. It is common ground however that, as in so many such situations, there was never an express agreement between the parties as to confidentiality.
74. It is further contended by Qual-Chem that none of the alleged disclosures was an enabling disclosure and that in any event, they did not constitute a public disclosure.
75. Corus accepted that the burden was upon them to prove that these trials form part of the state of the art within s 2 of PA'77, that is, that at least one person was free in law and equity to disclose the relevant information: Terrell 16^th Edn § 7-12.
76. It was also, I think, accepted that by its nature, the process claimed in claim 1 had to be trialled so as to establish that it worked, that trials could only be carried out in a steel works and that at the time, the only practicable steel works to do it in the UK, was at the Port Talbot works.

The Facts: Witnesses

77. Qual-Chem called two witnesses to deal with this aspect of the case, Mr Keith Harris and his son in law, Mr Steve Bray, the named inventors in the Patent. Mr Haris is the managing director of Qual-Chem whilst his son in law Mr Bray was its product development manager.
78. The trials in question were conducted on site by Mr Bray with input from Mr Harris, assisted also by sundry employees of Corus. It was suggested that as Mr Harris was physically not present at the actual trials, his evidence on these matters was of little relevance. I disagree. Mr Bray kept his father in law continuously au fait with the trials and with progress generally – and on a contemporaneous basis. Mr Harris thus provides strong corroborative evidence of Mr Bray's evidence. In a small family company like Qual-Chem, one would indeed expect a close working relationship of this kind.
79. I found both Mr Harris and Mr Bray to be sound witnesses of fact with good memories.
80. Corus' sole witness on this matter was Mr George Govan, the Plant Manager at their BOS plant at Port Talbot. At the time of these trials, he was manufacturing manager with responsibility primarily for efficiency and safety. Not surprisingly therefore he had to make it his business to know what was going on in the plant and this, he said, included what Qual-Chem were doing in 1999.
81. In my assessment, Mr Govan gave evidence of a most relevant fact relating to the issue of confidentiality. By 1999, he already knew Qual-Chem and a number of its staff since they had been BSC's and later Corus' suppliers of materials (including aluminium) for use in the secondary treatment of steel since the mid-1970's^[28]. It seems that they were in fact all on friendly, first-name terms. Corus (the biggest steel maker in the UK), became interested in a 'collaboration' with Qual-Chem into slag conditioning with aluminium in early 1999 after encountering some difficulties in this regard^[29]. This is what in due course led to the trials – in which Corus of course took a considerable commercial interest. There was he said, "an open dialogue between us at Corus and Qual-Chem"^[30]. The evidence as to the parties' close commercial relationship was corroborated by Messrs Harris and Bray.
82. When the relevant trials took place in around mid-1999, in addition to Mr Govan, the following Corus employees seem also to have been to varying degree involved: Mr Gerry Bekta (a steel technologist), Ms. Helen Taylor (who documented the trials), Mr Ian Gittings (who was present at a number of the meetings) and Mr Bob Colquhoun, who was Mr Govan's superior. Whilst their evidence would have been most material, none of them were called. This emerges both from the evidence and the disclosure documents. Take for example, the position of Gerry Bekta who it was said (like Mr GoVan) was at the time thinking of pneumatically gunning the slag with solid aluminium^[31]. The fact that Corus called only Mr Govan to give evidence as to what occurred at these trials (and of what people were thinking), was a matter of comment by Mr Wyand.
83. When the detail of what occurred some seven years ago is of undoubted probative value (as it is in many prior user cases) and where no reason is advanced as to why a particular witness to those facts has not given evidence for the party bearing the burden of proof (as here), I think Mr Wyand's criticism is justified – but is not of course, determinative.
84. I found Mr Govan to be a reserved witness, the quality of whose evidence struck me from time to time as being rather partisan when it came to sensitive matters. In relation to the conflict of evidence over what happened at the first trials, I therefore prefer the evidence of Mr Bray to that of Mr Govan.

What actually happened at the first trial?

85. I shall go first to the contemporaneous record of events taken from the disclosure in the case, the most important item of which is a Qual-Chem document [5/61]^[32]. From its language this document was written for the benefit a third party which not surprisingly, turned out (but only during the trial) to be Corus itself^[33]. It is headed 'Private and Confidential' and contains a record of inter alia what was agreed to be the result of the first trial. Under 'Result' it is recorded that there was 'Excellent coverage and exothermic reaction. Very fluid slag….' Later we read the following:

'Our most recent trial has been the application of our new aluminium composite onto the surface of the slag as the ladle leaves the rear of the converter. The material is pneumatically accelerated into the slag where it can react in an exothermic mode within the slag. We currently await the chemical analysis of the slag samples …to quantify the benefits of this slag conditioning practice.'

86. In assessing the impact of this document I have tried also to take into account the contemporary commercial background. Fortunately, Mr Govan has provided a sufficient summary of what was happening in his witness statement: 3/3/§§ 27-33. Basically, as I have previously mentioned, Qual-Chem had (and was still having) a good commercial relationship with Corus supplying it with varieties of materials for steelmaking. They had however recently lost a lucrative contract for the supply of aluminium 'dabs' for slag treatment to a third party competitor.
87. In early 1999, Corus' Mr Govan and Mr Betka visited Qual-Chem to see if Qual-Chem could help to improve their existing slag treatment with a new Qual-Chem product which they had heard about^[34], being a pressed tablet or briquette made from aluminium scrap which was apparently cheaper than what they were using. This was in fact the origin of the patented process. As Mr Govan records^[35]:

"Qual-Chem were very keen to find a buyer for this product and in view of the cost savings for us, we agreed to work with them on it."

88. Initially, Mr Govan records being sceptical about the new Qual-Chem product but during July 1999, Qual-Chem came up with a briquette "which we agreed could be trialled on IF grades at the plant." Trials indeed later took place and there seems to have been a good deal of technical interaction as a result. But, said Mr Govan, the Qual-Chem product was not proving to be as good as the shot they were already using. By August however,

" as a result of discussions between Qual-Chem and us, it was agreed that a spraying method would be trialled using the Qual-Chem slag conditioning aluminium pressed tablet".

and this led to the first trials with which I am concerned in this case.

89. What then happened from Qual-Chem's point of view is well recorded in the witness statements of Messrs Harris and Bray and the cross-examination thereon.
90. First, it is clear that it was Mr Bray who was in charge of what was being done on site, was being assisted by operatives from Corus. It is also clear that Mr Govan knew what was going on in some detail throughout. Mr Bray reported on a daily basis to Mr Harris and the two of them seem to have undertaken some collaborative thinking about this slag killing problem. They had two variables to manipulate: the nature of the aluminium additive and how to get it into and to react with the slag. After some initial abortive thought as to how best to get the aluminium to react with the slag, Mr Harris hit upon the idea of using a pneumatic applicator. He had previously used such a machine (from a company called Gunform Ltd) in relation to the spraying of refractory linings for tundishes. His thinking was to try somehow to adapt this equipment to achieve an even distribution of the Qual-Chem aluminium pellets over the surface of the ladle. He discussed this with Mr Bray and the equipment was sourced and in due course tried out at Port Talbot by Mr Bray. As Mr Harris put it:

"We were left to get on with things."

91. Basically (and very much in brief as there was much evidence on this) these witnesses record that the first trials met with some success – but it was a qualified success only. There was indeed some local spread – but on the other hand, the reductive reaction only took place within such cracks and fissures as there were in the surface of the slag. What Mr Bray saw is clearly stated in the evidence: 'islands of slag were still untreated'; this was 'quite promising' but was nonetheless, technically unsatisfactory: T2/269. The nozzle spread the aluminium but lacked the power to penetrate.
92. It then occurred to Mr Harris and Mr Bray that if this idea was to have a practical future, the aluminium had to be forced ('gunned' as Mr Harris said) more efficiently and reproducibly into the slag layer. This had to be done by adjusting ('tailoring') the various parameters such as the nozzle, distances, the aluminium particles etc, 'depending on application requirements', as the application for the Patent later stated. Mr Bray's first attempts at spraying the pellets were partially successful only but (and this is important) Messrs Harris and Bray got the insight from these first trials that they were onto a potentially viable process for killing slag. But more trials were needed. Acting on their insight, they applied for the Patent (16 September 1999) and actively prepared for the second and third trials which followed – and which according to both of them they considered to be necessary to justify their initial enthusiasm. That was their thinking at the time and I accept this evidence.
93. Based on lessons learnt from the first trials, in the trials which followed aluminium pellets were fired into the slag from various heights at different angles and so forth. The method and apparatus indeed proved to work.
94. Returning now to the Qual-Chem 'Private and Confidential' document mentioned in §85 above [5/61], the premature optimism there expressed is understandable. Not surprisingly, Mr Bray was vigorously cross-examined on it: see T2/ 269-286. What emerges is this. Mr Bray continued to assert that he thought that Qual-Chem were onto something that would most likely prove to be effective – and in commercial context, would be a welcome outlet for the bulk sale of their new aluminium pellets. But they were not home yet. I am satisfied that this document must be read in this way. The enthusiastic tone of the document can I think be explained more as the result of Qual-Chem's commercial hopes than as an indication or scientific record of what happened.
95. For the record I am not convinced with Mr Govan's explanation of what happened at the first tests as the result of what he called "some natural suction of the pellets into the slag." where the cracks were: T2/300. I have seen no reference to suction at the slag surface occurring in any written material in this case. But some pellets did of course fall into the cracks and reacted there.
96. In the light of this, in my judgment, the first trials did not make use of integer E of claim 1. There had been no 'tailoring' and the objection of lack of novelty fails on this basis alone.

The parties' relationship.

97. In the light of my findings concerning the first tests, I need go no further; anticipation by the prior users pleaded, fails. Nonetheless, as some time was spent at trial on the issue of whether all these experimental tests were in any event carried out under circumstances importing an obligation of confidence upon Corus ( and thus in any event outside the area of public disclosure of the invention), I shall next consider this issue.
98. In this context, I was referred by counsel to a number of well-known authorities relating to breach of confidence: Pall v Commercial Hydraulics [1990] FSR 329, Strix v Otter [1995] RPC 607, and Kavanagh Balloons v Cameron Balloons [2004] RPC 5. I was also referred to prior user cases from the EPO: for example, T 782/92.. These and other cases in the same field depend to a degree on the facts of the commercial relationship in question.
99. Mr Bray was in no doubt whatever that from his perspective what Qual-Chem were doing was confidential to the parties: see T2/254-255 and 247-248. Mr Bray also said that when third parties visited the site of the trials, the guns were dismantled and removed: T2/249.
100. I would just mention the marginally relevant fact that though Corus did themselves make use of the claimed invention (see above), there is no evidence that they told any third party what slag killing process they were using.
101. In my judgment and in the light of the approach given in these authorities and of the facts relating to the contemporaneous dealings involving the parties, I have no doubt that Corus and its employees were not 'free in law and equity' to make whatever use they wished of any information gleaned from any of the Qual-Chem experimental trials which took place in 1999. In coming to this conclusion I have been impressed by a number of factors:

(a) The long, ongoing 'open and honest' and generally cordial relationship which existed between Corus and Qual-Chem: see Govan T3/359.

(b) The need for such tests to be conducted and the fact that, in the United Kingdom, they could only have been conducted at Port Talbot.

(c) The fact that the trials were devised and supervised by Qual-Chem.

(d) The fact that the trials arose as a result of a practical need on Corus' part for a better and cheaper method of slag killing, a fact which itself was not of course, in the public domain and in reality was commercially sensitive.

(e) The curious fact that Corus at one time (2003) contended that it had a right to be nominated co-applicant for the Patent: see Corus' letter 5/259-260.

Were the trials an enabling disclosure?

102. Finally under this head, Qual-Chem contended that the trials, even if falling within the process of claim 1 and free from any bonds of confidence on Corus' part, did not constitute an enabling disclosure of the method subject of claim 1, In essence, this boils down to 'What did the Corus personnel (particularly Mr Govan) see?^[36] What did those present see? What did they know?
103. There is a good deal of conflicting evidence here, the parties' respective memories and beliefs from some seven years back being quite predictable. In this connection, it stands agreed of course that Mr Govan knew about most of what Mr Bray was doing. It is also common ground that the fumes generated during the addition of conditioner may have impaired clear vision.
104. Mr Wyand's submission was: "Yes but Mr Govan is guessing with regard to integer E. He never really looked at what was happening. He is guessing". Mr Govan's riposte was: "I did look from time to time but in any event, I knew perfectly well what Mr Bray was trying to achieve and he did achieve it. It was being done for Corus, after all".
105. I have reviewed the evidence and have come to the conclusion that Mr Govan knew and saw enough for that to be an enabling disclosure.

Obviousness

The Law

106. There was no issue between the parties as to the applicable law. The correct general approach to obviousness is that set out in Windsurfing v Tabur Marine as re-ordered by Jacob LJ in Pozzoli v BDMO [2007] EWCA Civ 588 at [23], namely:

"(1) (a) Identify the notional "person skilled in the art"

(b) Identify the relevant common general knowledge of that person;

(2) Identify the inventive concept of the claim in question or if that cannot readily be done, construe it;

(3) Identify what, if any, differences exist between the matter cited as forming part of the "state of the art" and the inventive concept of the claim or the claim as construed;

(4) Viewed without any knowledge of the alleged invention as claimed, do those differences constitute steps which would have been obvious to the person skilled in the art or do they require any degree of invention?"

The documentary citations

Japanese application ? H01-294 817 (Kawasaki) (1989)^[37]

107. The version of Kawasaki used in Court was an awkward translation from the Japanese. Overall, Kawasaki's proposal is to kill slag using aluminium powder as the reducing agent by blowing it with gas onto and into the slag via a lance. The carrier gas must be inert. Kawasaki clearly has the same overall approach to slag killing as that of the Patent – but the detail is quite different.
108. For the purpose of assessing inventive step, one follows the structured approach laid down in Pozzoli/Windsurfing. One posits a skilled man with all his attributes, positive and negative, reading the citation in question at the priority date. He obviously has no knowledge of the Patent and is interested in improvements to known methods of killing slag – with aluminium. What does Kawasaki tell him? Would he find Kawasaki to be of any practical assistance? If so, would he make any changes to Kawasaki and what changes would he make? My criticism of both experts' approach in relation to Kawasaki is the same: they have already read the Patent and this has I think affected their evidence.
109. Kawasaki first describes two prior art methods for killing slag ( pp 71-72) both involving the addition of aluminium in the form of 'lumps', bar, shot or wire. Whilst this treatment is effective, Kawasaki says that it has two drawbacks. The reaction of aluminium shapes with ferric oxide is slow and in addition, the use of shapes cannot deoxidise the slag evenly^[38]. Kawasaki tackles this problem in three ways:

(i) Instead of solid aluminium shapes he proposes the use of aluminium powder^[39];

(ii) Kawasaki blows the aluminium powder over, onto or under the slag using a 'top-blowing lance', and nitrogen or argon are identified as being suitable inert carriers.

(iii) A more even distribution of the aluminium powder may be procured by movement of the top-blowing lance. Thus the lance may be bent and rotated (Fig 1b) or it may be tilted back and forth on a central axis over the surface of the slag (Fig 1c).

Note however that the lance is fixed centrally above the contents of the ladle: see Fig 1.

110. In relation to the slag to be treated, the lance may be placed just above it (Fig 1 (a) (b) and (c)), within the slag itself (Fig 2), and for blowing into the slag, 'immediately below the slag layer'. In each case, the lance may be rotated or tilted back and forth for a more even application. Kawasaki suggests a number of working permutations of which the principal one of interest to the experts (and to counsel) was that of moving the lance just above the surface of the slag. This fixation of interest must I think have come about as a result of reading the Patent. I shall call this Method 1.
111. Practice of Kawasaki's method leads we are told to accelerated reaction times and to a more even distribution of the deoxidant as already mentioned^[40]. Note that: 'by sealing above the slag layer with inert gas the oxidation loss is reduced which is to oxidise the added reducing agent by air…' The translation becomes particularly poor at this juncture but as I understand it, this is a reference to the potential premature loss of the aluminium powder before it does its deoxidising work. As Dr Bommaraju said in evidence, that in air, the powder would 'catch fire' in the aggressive environment above the ladle. Evidently the problem would be particularly troublesome using Method 1.
112. To overcome this problem, Kawasaki then proposes a sealed vessel (Fig 4) for use when the top-blowing lance is neither immersed in slag nor immediately below it. When Method 1 is used, the sealed vessel is flushed with inert gas to prevent the aluminium powder oxidising on the surface of the slag^[41]. This benefit is reflected in claim 5 of Kawasaki. This teaching would not I think be glossed over by the skilled reader.
113. In my judgment, a fair reading of Kawasaki teaches the following:

(i) Do not use aluminium shapes of any kind: use aluminium powder instead;

(ii) Inert gas must be used as carrier in all cases

(iii) Rotate or tilt the centrally placed lance to secure a more even distribution of the aluminium powder.

(iv) To prevent the powder catching fire using Method 1, kill the slag in a closed ladle flushed with inert gas.

(v) According to the common general knowledge, there was a prejudice against using aluminium in powder form because of flaring (see above). This teaching enables the benefits of using aluminium powder to be used nevertheless.

114. To get Kawasaki within the focus of claim 1 for the purpose of their attack on obviousness, Corus' evidence begins with a substantial 'if'. If one were to ignore the plain teaching of Kawasaki and not use aluminium powder and inert gas, one would inevitably use aluminium shapes and air as the carrier instead. Not surprisingly, Dr Bommaraju agreed: T4/548-549. But such a reading comprehensively abuses the very point of Kawasaki. In view of this, Mr Tappin's submissions on this document are quite artificial and must be rejected.
115. But even if such a transformation of thought were likely, this still does not bridge the gap between this prior art and the claim. Whilst one of the lance configurations is 'inclined' and another is intermittently so, none I think, provide a divergent stream. The even distribution of the aluminium powder is achieved not by the blowing of entrained material so as to create a divergent stream but by the rotation or tilting of the delivery means^[42]. In addition, as Mr Wyand pointed out, there is no teaching of 'tailoring'.
116. In my judgment, the validity attack based on Kawasaki must fail.

US Patent ? 5 332 199 (Knapp) (1994) ^[43]

117. This US patent proposes a method of treating molten metal in vessels with (i) 'conditioning gas' and (ii) 'fine grain solid materials'. This is done by using two water-cooled lances, the first supplying the conditioning gas (e.g. oxygen) and the second being used for the pneumatic supply of the fine grain solid material. The two lances are maintained at an angle to each other in V-shaped configuration. The lances are also set at an inclined angle to the horizontal, both pointing down to the surface of the slag and molten metal at the same angle.
118. The conditioning gas in the first lance travels at supersonic speed and impacts with the air and solid material entrained therein exiting from the second lance, completely diverting it in the direction of flow of the conditioning gas. The air or inert gas which entrains the solid material is thereby dissipated into the ambient atmosphere (Col 2 ll 30-34). The conditioning gas and the fine grain solid material then penetrate through the slag layer (item 8 in Figs 1-3) into the liquid metal.
119. The relevant 'fine grain solid materials' are carbon or slag forming agents such as lime, bauxite and dolomite and/or metallurgical agents for treatment of the molten metal ( Col 2, ll 37-41).
120. Knapp is put forward by its inventors as being an improvement on their earlier patent (US ? 4 986 847) which is described in its first two columns: see particularly Col 2, ll 60 – Col 3, line 20. One object of the invention is to avoid excess of air during the treatment process by inserting the lances through a hole in the furnace wall (Col 2, l 67 – Col 3, l 4).The treatment of metal proposed by Knapp appears to occur before pouring that is, within the furnace itself. The furnace described is an EAF furnace ( Col 3, ll 52-53) where there would be little or no crust on the slag: T4/612. Not surprisingly, the 'layer of slag (8)…is not shown in greater detail' : Col 3 l 57. Col 4, ll14-18 reads as follows:

' The angle a at which the two lances 117 and 119 extend with respect to the horizontal plane is about 50°; that ensures that the fine-grain solid materials and gases which are introduced by way of the lances….can satisfactorily penetrate through the layer of slag 8.'

121. The function of the lances is described in Col 4, ll 27-47. This and other passages show that Knapp relates to a method for treating molten metal only, in particular by the oxidation of the metal with oxygen travelling at a supersonic speed. It does not relate to a method of killing slag – which is of course associated with a reductive process. The slag layer is noted but the entire tenor of Knapp and the prior art mentioned is only for the treatment of molten metal. According to my reading of the experts' evidence and the background material to which I have briefly referred, in my view, the only reason why Knapp is put forward is because it shows the pneumatic conveyance of materials in another steelmaking process. Adding materials to condition on the one hand, slag and on the other, the metal below it may occur in the same circumstances (or even at the same time) but are separate steelmaking activities. One may want to carry out one but not the other. Integers A-C of claim 1 are not therefore present.
122. Moreover, though what is conveyed by air or inert gas is indeed conveyed in an inclined stream from a conveying apparatus spaced above the slag surface, in the light of the passages noted from Col 4 of Knapp, I do not think that anything but minimal divergence of the stream would arise, let alone strike the skilled reader as being desirable. On the contrary, the fast-moving stream of oxygen is being targeted to a 'locally delimited area of molten metal 5 in a 'direction which remains substantially unchanged' as it leaves the outlet 123 of the conditioning gas lance. In my judgment, integer D is only present to the extent that inclined delivery means are present.
123. Furthermore, there is no question of the gas pressure being 'tailored' with respect to the conveyed additive material to cause it to penetrate and remain in the slag. Integer E is therefore not present. Finally, the conveying gas may be air (Col 2. l 26) but it is not this gas which conveys the fine grain solid material but rather the supersonic conditioning gas; the conveying gas is dissipated.
124. Turning to the evidence, the burden is of course on Corus to make good their case on Knapp. In the first place, I was not very impressed with Mr Pope's overall assessment of Knapp: T4/614-615. On a number of assumptions, it is no doubt possible to make the six or seven changes to Knapp which would be required to bring it within claim 1 of the Patent and this was indeed put to Dr Bommaraju^[44]. However, I am not convinced that the skilled person would be likely to be moved to make all these changes and further, that if he did so, he would not then be exercising an inventive capacity.
125. The attack on the Patent based on Knapp therefore fails.

Obviousness based on Common General Knowledge

126. I have already considered the issue of common general knowledge: see §§40-43 above. It is true that a number of the integers of claim 1 both individually and collectively could be said to form part of the common general knowledge of this aspect of steelmaking. Thus, using aluminium in various physical presentations to add to the ladle so as to condition steel and to kill slag thereon was commonplace. Such conditioners were added by gravity - chute or shovel, as I understand it. I have however rejected the proposition that everything within claim 1 was common general knowledge at the priority date, which appeared to be the basis of Mr Pope's view about claim1. Though I have no doubt that machines for the pneumatic delivery of various materials were not only commercially available at the priority date (for example for use in the building trade), and were even likely to be found somewhere or other on Corus' extensive Port Talbot site, there is no evidence that until Messrs Harris and Bray hit upon the idea of using such machines to condition the ladle contents by 'gunning' aluminium shapes into the slag layer, anyone had done so before^[45]. Indeed the very fact that such machines were (as Mr Pope says) 'hanging about' at Port Talbot but used for other purposes, seems to me, if anything, to enhance the likelihood of an inventive step having taken place.
127. I would however go further. Allegations of obviousness based on common general knowledge alone have to be approached with caution as they tend to be tainted with hindsight: see per Pumfrey J in 3M v ATI Atlas Ltd [2001] FSR 514 at 529^[46]. In my view, that is particularly true in this case.

Conclusion on obviousness

128. The method claimed in claim 1 may not be a great advance in the process of conditioning slag^[47]but that is not the question when it comes to assessing inventive step. I have reviewed the evidence in relation to each citation and consider that an inventive step was taken in relation to the method of claim 1. The attack based on obviousness therefore fails.

Loss of Priority and Added Matter

129. A number of points have arisen under this head: see Grounds of Invalidity, paras 3 and 4^[48]. However, Mr Tappin has not proceeded with a number of the specific issues pleaded in the light of the way the case has developed.

The facts

130. It will first be necessary to read and understand the relevant specifications in a purposive manner, using for this purpose, the eyes of the skilled person at or about mid-1999.

The earliest priority document

131. The earliest priority document is dated 15 September 1999 (? 992 1818 at 1/3^[49]). It concerns the use of a pneumatic gun to introduce 'conditioning material' into the contents of the ladle, the contents of which are both slag and 'molten steel and/or iron material'. The contents are referred to collectively as 'molten material': see p.6, l 25 – p.7. The passage on p.2 ll13-17 refers specifically to the conditioning of the slag using the gun. The use of aluminium elements for this purpose is also disclosed: p.4, lines 11-16.
132. On p2. ll 29-31 one reads the following:

'The delivery pressure of the conveying gas is therefore tailored depending on the 'sinkage' requirements of the additive/conditioning agent.'

133. The gun ('apparatus') is described on page 5 (as a second aspect of the invention) and is the subject of the only diagram in the document. I notice that the word 'tailored' is used in several places in connection with using the gun to condition the ladle contents.
134. In my view the skilled reader would at once appreciate that this teaching relates to conditioning of the 'molten material', either steel or slag, using the gun described. The 'sinkage' of the particulate conditioner from the gun is tailored to the sort of conditioning one has in mind.

WO/01 20047A1

135. I shall next consider the Patent as granted but before post-grant amendment: [1/2]. The prosecution citation US 4601749 (Yamada-see above) may have caused the application to be revised. Corus has supplied [at 1/3.1], an emotive comparison between the text of the priority document [1.3] and the Patent as granted [1.2], deletions being struck through in red and additions being shown in blue. The comparison shows that there has been a substantial change of direction of the invention as a result of the amendment. The Patent (before post-grant amendment) began thus:

'The present invention relates to a method of introducing additives during steelmaking either to molten iron in a ladle or the like or to molten iron while the latter is being poured'. [Emphasis added]

136. On page 4, lines 7-12, the following is stated:

'When the stream including the additive is directed towards the surface of the molten iron in the receptacle, the additive is preferably conveyed to reach below the aforesaid surface penetrating through slag or other surface covering thereon.'

137. The additive may be in the form of 'tablets, pellets, briquettes or powder' (p.5, l.12) and the use of inter alia aluminium for this purpose is maintained.
138. My Wyand submits however that though the thrust of the disclosure has undoubtedly shifted to the conditioning of steel, the use of the same gun to condition slag alone has not been abandoned. He relies in particular on the following passage, on p5 lines 25-29 which has been lifted from the priority document (p.2, ll13-17):

'The predetermined specific density of the particulate additive can ensure that the particles penetrate into and remain in, the slag (rather than descending into the liquid iron below) but resist flaring off on the surface.'

139. There is another passage which specifically refers to killing slag on the surface of the steel (at p. 7, ll 26- 31) and in addition, a method claim, claim 15 is directed to conditioning either steel or slag. The experts were (perhaps unnecessarily) asked questions as to their understanding of these passages: see for example, Dr Bommaraju T/3 432-433 and 433-434. Ultimately however the decision as to the meaning of the various passages is for the Court.
140. In my judgment, a fair reading of the Patent as granted would tell the skilled reader that though the main purpose of the method taught is to condition molten iron or steel in the ladle, another, though certainly less prominent purpose, is to be able to kill the supernatant slag alone – using of course the very same gun - should one wish to do so. I would add that the use of additives in the ladle to condition steel or to kill slag formed part of the common general knowledge (see above).

The Patent as amended after grant: the C specification

141. I have already fully considered the contents of the C specification in issue under 'The Patent', see § 20 above. Corus have again provided a colour comparison to highlight the differences between the Patent after grant and after amendment: [1/2.1]. In brief, the Patent has now taken another sharp turn but not I think in an altogether different direction. It is now directed to the conditioning or killing of the slag alone – and not of the molten iron/steel. Mr Tappin submits that the Patent as amended now relates to quite a different invention since the Patent as granted is concerned with the conditioning of molten iron - and he refers to pages 1 and 2 of the Patent as granted in particular. I need not elaborate on this since the core observation of Mr Tappin on the thrust of the teaching of the Patent as granted is plainly right. But as noted above, the use of the gun to condition slag, though relegated to a minor position in the narrative, was never abandoned. But is that sufficient to sustain Corus' objection under 72(1)(d) of the Act?

Loss of Priority date

142. In the Grounds of Invalidity, para 4, it is said that the invention claimed is entitled only to a later priority date, being that of the date of the International application, 18 September 2000. Corus invokes s 5(2)(a) of the Act in this connection. It will be recalled that the word 'inclined' is used in integer D of claim 1 and is thus an essential part of the integer. It is common ground that the word 'inclined' was not used in the first priority document. However there is a single two-dimensional figure which forms part of it. The basis of the objection is that in the earliest priority document the inclination of the divergent stream was not mentioned and that this figure shows a divergent stream of additive playing vertically above the ladle and not being inclined to it.

The Law

143. I was referred to the classic authorities on this topic: Biogen v Medeva [1997] RPC 1 at 46-52, Pharmacia v Merck [2001] RPC 41 at [91]-103] and Unilin v Berry [2005] FSR 6 at [44]-[50]. The question is whether there has been an explicit or implicit disclosure in the priority document and not whether it would have been obvious to do something in a certain way.

The Facts

144. Much time was spent in evidence and argument as to whether this single figure was an elevation or a plan diagram. It is described (ambiguously in my view) in the application as a 'schematic side view' and there is a contrast within the narrative between the proposal and conventional gravity feed arrangements. The issue is whether this constitutes an implicit disclosure of inclined introduction of conditioning material.
145. I have re-read the evidence on this aspect of the case and, should the matter go further, record that I agree with Mr Tappin: I am not satisfied that the Patent is entitled to the earliest priority date. I say 'should the matter go further' because the issue is irrelevant in the light of my finding that none of the trials were conducted within the public domain.

Added Matter

146. I shall therefore pass on to what I regard as the more substantial objection under this general head viz that 'the matter disclosed in the specification of the patent extends beyond that disclosed in the application for the patent, as filed.': s. 72(1)(d) of the Act. The Patent was amended after grant – and hence the objection. The objection is pleaded at §3 in the Grounds of Invalidity but by the end of the trial, the objection reduced to that contained in § 3 (a), (b) and (e).

The Law

147. In Bonzel v Intervention (no 3), [1991] RPC 553 at 574, a leading authority on this part of the law, Aldous J stated as follows:

" The decision as to whether there was an extension of disclosure must be made on a comparison of the two documents read through the eyes of a skilled addressee. The task of the Court is threefold:

(a) to ascertain through the eyes of the skilled addressee what is disclosed both explicitly and implicitly in the application;

(b) to do the same in respect of the patent as granted, and

(c) to compare the two disclosures and decide whether any subject matter relevant to the invention has been added whether by deletion or addition.

The comparison is strict in the sense that subject matter will be added unless such matter has been clearly and unambiguously disclosed in the application either explicitly or implicitly.".

148. In Richardson -Vicks' Patent [1995] RPC 568 at 576, Jacob J summarised the rule against added matter in a single sentence:

" I think the test of added matter is whether a skilled man would, upon looking at the amended specification, learn anything about the invention which he could not learn from the unamended specification."

Application to this case

149. In 130-141 above I have set out my understanding of how the skilled person would read and understand the three documents presently under consideration. There have undoubtedly been 'about turns'. Nevertheless, the teaching relating to slag killing using a pneumatic gun which started with equal prominence to that of metal conditioning, then diminished to that of an incidental benefit or bonus and resurrected finally as the sole purpose of the method in the C specification, has always explicitly been present. Killing slag with the gun was disclosed at the start and never dropped. Moreover, the vagaries of the way these specifications were drafted have not in my view led to any unwarranted advantage to the patentee over this time.
150. Applying the relevant principles of law therefore, I have come to the conclusion that there has been no added matter as a result of amendment and that this objection has not been established.

Patents Act 1977, s. 64. The right to continue to use the process

151. I have held the Patent to be valid and that it has been infringed. Is Corus nevertheless entitled to rely as a defence on the provisions of the Patents Act 1977, s 64 (as amended)?^[50] I incorporate the section by reference into this judgment.
152. First, it is important to establish some facts. Before the trials started, Corus had a slag killing problem^[51]. They turned to Qual-Chem for help, the latter being (as Corus knew) experts in this field. As a result Qual-Chem alone devised the trials which (a) had to take place before anything could be proved anyway, and (b) could only take place at Port Talbot , being realistically the only site available in the United Kingdom where the trials could be carried out. Mr Govan seems to have been the senior liaison person at Corus for this purpose.
153. The question of who carried out the trials was specifically canvassed with Mr Govan in cross-examination. He corrected his evidence in this respect; it was Qual-Chem who carried out the trials: T3/352-353.
154. Thirdly, I have held that the first trial (i.e. pre-priority date) did not make use of the process of claim 1; that is therefore the end of the matter. So there was no exercise of the process of claim 1 by anyone before the priority date and s 64(1) does not apply.
155. Furthermore, since it was not Corus who conducted these trials anyway, s. 64(1) (a) and (b) does not apply for that reason also. Corus cannot have the 'practical protection to enable a man to continue doing what in substance he was doing before": Lubrizol v Esso [1998] RPC 727 at 770. In my view s. 64 affords no defence to Corus.

Conclusion

156. The Patent is valid and has been infringed. The s.64 defence fails. I shall hear counsel in due course on the form of order to be made in the light of this judgment and on the question of costs.

Note 1   References to the court bundles, dividers and pages thus 5/4/23. References to the transcript pages thus: T4/23    [Back]

Note 2   See Defence §9 [2/3].    [Back]

Note 3   4A/4/188 et seq.    [Back]

Note 4   See T4/560 and T3/390    [Back]

Note 5   Bommaraju ¶28. [4/1/11]    [Back]

Note 6   As will be seen, what follows is within the ‘common general knowledge’ of the steelmaking art.    [Back]

Note 7   See further under ‘Added Matter’, below    [Back]

Note 8   1/7    [Back]

Note 9   Patent p 4, l. 19    [Back]

Note 10   Page numbers refer to the pagination of the Patent and not of the bundle.    [Back]

Note 11   For convenience, called ‘particles’ from time to time in this judgment.     [Back]

Note 12   See § 40, post.    [Back]

Note 13   This passage illustrates how the draftsman has wrongly in fact stated that this is a preference whereas it is in fact a requirement of claim 1.     [Back]

Note 14   Inland Steel –Flat Products, East Chicago, IL     [Back]

Note 15   Report para 4 in 4/1.    [Back]

Note 16   See para 8 above.    [Back]

Note 17   The Patent is not however concerned with steels made by the BOF process alone.    [Back]

Note 18   Apart from the Bommaraju paper and Corus’ other prior art citations (Kawasaki and Knapp) the only evidence relating to using pneumatic apparatus in steelmaking was Mr Pope’s recollection of an old and obscure method (Aliva guns) not forming part of the common general knowledge: T4/ 600     [Back]

Note 19   Kirin-Amgen (supra) at § 32.    [Back]

Note 20   T4/566-570.    [Back]

Note 21   See 1/7. T4/481-482    [Back]

Note 22   See Yamada Col 4, ll. 7-9.    [Back]

Note 23   “ It is new jargon to me” : see T4/583-584.    [Back]

Note 24   Whose shape, density and composition had been established by say, the manufacturer. Cf the Patent p5, ll 11-15    [Back]

Note 25   Bommaraju ¶¶42-45. [4/1/19]    [Back]

Note 26   Exhibit 1 was a bag of Qual-Chem’s aluminium additive.    [Back]

Note 27   T4 /564. The evidence was that aluminium used in steelmaking comes in a wide variety of forms: T4/589    [Back]

Note 28   Witness statement: 3/3/40.     [Back]

Note 29   See Document 4A/19 and T2/305-319    [Back]

Note 30   Witness statement, 3/3/43    [Back]

Note 31   Govan Witness statement, 3/3/43.    [Back]

Note 32   Mr Harris did not know who wrote this document. Its date was put at August oR early September 1999    [Back]

Note 33   When Mr Bray was being cross-examined about this document there was confusion as to whether it was an internal document. Later however (see the cross-examination on Document [6/32]) it became clear that it was written for the sole benefit of Corus with whom (it will be recalled) Qual-Chem was collaborating. However, Mr Govan had no recollection of having seen the document before.    [Back]

Note 34   An account of what was happening from Qual-Chem’s point of view up to mid-1999 is set out in Mr Harris’ witness statement [3/1] at §§ 12-21. This evidence (which I accept) also well illustrates the close working relationship which existed during this time between the parties.    [Back]

Note 35   3/3/42    [Back]

Note 36   See for example T2/303 and T3/ 330    [Back]

Note 37   1/5    [Back]

Note 38   Trs. p 72 first complete para    [Back]

Note 39   Preferably having a diameter of > 10mm    [Back]

Note 40   p 73 second complete paragraph.    [Back]

Note 41   P 74 first complete paragraph.     [Back]

Note 42   See Mr Pope: T4/633    [Back]

Note 43   1/4    [Back]

Note 44   See for example T4/538-540 and the summary table of differences prepared by Qual-Chem’s counsel in their closing skeleton of argument, §55.    [Back]

Note 45   The most Mr Pope can say is on T4/572    [Back]

Note 46   Not cited by counsel.    [Back]

Note 47   Though as I have already noted it seemed to possess unchallenged benefits: see § 50.    [Back]

Note 48   2/3    [Back]

Note 49   I.e. the first priority document. It is I believe, common ground that the claims are not entitled to the second claimed priority date of 24 March 2000.    [Back]

Note 50   See §§ 3 and 6 above    [Back]

Note 51   See for example the contemporaneous documents at 4A/19 and 20 and 6/45.    [Back]