Scandiamant Aktiebolag, Appellant, v. Commissioner of Patents, 509 F.2d 463 (D.C. Cir. 1974)

George W. Whitney, New York City, with whom Henry Taliaferro, Jr., Washington, D.C., was on the brief for appellant.

Fred W. Sherling, Associate Sol. U.S. Patent Office, with whom Joseph F. Nakamura, Acting Sol., U.S. Patent Office, was on the brief, for appellee.

Before ROBINSON and ROBB, Circuit Judges, and MATTHEWS,^*  Senior District Judge for the United States District Court for the District of Columbia.

MATTHEWS, Senior District Judge:

Inventors Olle Lindstrom and Erik Lundblad filed an application for a patent on a method for preparing diamonds for use with grinding wheels. The claimed invention is the alleged discovery that the use of diamonds coated with metal in a prescribed thickness range in a resin bonded grinding wheel will improve the grinding results, i.e., the measure of grinding efficiency of the wheel, by as much as 200% to 300%. The thickness range for this metal coating is between 0.5 and 25 microns.

The Patent Office rejected the application, and the Board of Appeals affirmed that rejection.¹  The appellant, scandiamant Aktiebolag (Scandiamant),²  assignee of the named inventors, then brought an action in the District Court under 35 U.S.C. § 145 to compel the issuance of a patent. The District Court dismissed the suit, and this appeal followed.

The only issue involved in this appeal is whether the subject matter of the application in suit, as described in the finally rejected claims, is obvious in the light of the combined disclosures of the U.S. patent to Wentorf No. 3,125,418, issued on March 17, 1964, and the French patent to Soulard No. 1,142,688, issued September 20, 1957.

Of the finally rejected claims, numbered 29, 31--33, and 35--40, claim 29 is acknowledged to be illustrative and reads as follows:

'29. A grinding wheel having a hub portion and a diamond section secured on the hub portion, said diamond section comprising a body of cured resin bond having imbedded therein diamonds coated with a layer of metallic material, said layer having a thickness of between 0.5 and 25 microns.'³ 

At the time Lindstrom and Lundblad made their discovery, there were basically two types of diamond bonded grinding wheels: the metal bonded wheel and the resin bonded wheel. Natural or synthetic diamond was used as the abrasive in each type wheel. The resin bonded wheel had a greater cutting ability than the metal bonded wheel, but suffered the drawback of accelerated deterioration during the metal grinding process. The reason for this was the limited capability of the resin to retain the diamond during the grinding process.

In 1963, Lundblad, the General Manager of the Diamond Division of ASEA,⁴  was authorized to conduct research to determine the cause of the accelerated deterioration during use and develop a resin bonded wheel with an improved grinding ratio.⁵  Working with him was Professor Olle Lindstrom, head of the central laboratory of ASEA. Lundblad and Lindstrom planned the work, and they had staff and personnel to carry it out. As early as 1962, others in the field, including major suppliers of diamond and manufacturers of bonded diamond grinding wheels, were actively engaged, without success, in the same type of research and development.

Lundblad and Lindstrom discovered, in the course of their work, that application of a thin, metal coating in a thickness range between 0.5 and 25 microns around the diamond and incorporating such metal clad diamond in a resin bonded grinding wheel, resulted in increased bondability with the resin in which the metal clad diamonds are embedded. There was greater dissipation of the heat buildup in the diamond. There was reduced fracturing of the diamond when worked which, in turn, eliminated premature loss (pullout) of the diamond from the resin. Most significant, however, was the discovery that the grinding efficiency of the wheel was improved by as much as 200% to 300%. They filed their initial application for patent October 19, 1965.

The Patent Office, the Board of Appeals and the District Court relied (as does the Commissioner) on a combination of the Wentorf and Soulard patents--'the teaching of Wentorf when taken with Soulard'⁶  --which, it is contended, renders the discovery of Lundblad and Lindstrom obvious. All claims were rejected as obvious under 35 U.S.C. § 103.

The Wentorf patent relates to the synthesis of radioactive diamonds. By adding a radioactive metal along with the metals which are normally used in the synthesis of diamonds, Wentorf produces a diamond having in adherent relation to its surface a radioactive metal with a thickness range of 0.5 to 5.0 microns. The process produces diamonds differing from naked uncoated diamonds only in regard to radioactivity. The Soulard patent shows an abrasive tool which is characterized particularly in that each abrasive grain is contained in a spheroidal coating binding it to the bond in the tool. The thickness of the metal coating varies, depending upon the size of the diamond. By calculation the Soulard range appears to be 25.5 microns to 937.5 microns.

Interpreting Wentorf as disclosing a 'coating' process, it is the position of the Commissioner that it would have been obvious to one skilled in the art to combine the Wentorf disclosure of diamonds coated with a layer of metallic material coming within the thickness range recited in the claims and which is incorporated in an abrasive wheel, and the Soulard abrasive wheel which includes diamonds with a metallic coating combined with a cured synthetic resin bond to yield a grinding wheel substantially similar to that claimed. The Commissioner further maintains that appellant has not made a disclosure of criticality which would support patentability of the claims on that basis.

'(A) presumption of correctness accompanies the decisions of the Patent Office, * * * and substantially the same rule applies to the findings of fact of the District Court.⁷  What the prior art is and what the claimed invention is are questions of fact. However, whether the standard of obviousness applied to those facts is correct, is a question of law.'⁸ 

Patentability of an invention depends not only on novelty and utility, but Section 103 of Title 35 requires that the subject matter as a whole must not have been obvious to one skilled in the art.⁹  In applying Section 103, the Supreme Court instructs us that 'the scope and content of the prior art are to be determined; differences between the prior art and the claims at issue are to be ascertained; and the level of ordinary skill in the pertinent art resolved. Against this background, the obviousness or nonobviousness of the subject matter is determined. Such secondary considerations as commercial success, long felt but unsolved needs, failure of others, etc., might be utilized to give light to the circumstances surrounding the origin of the subject matter sought to be patented.'¹⁰ 

The mere existence of the basic teachings of a new claim in the prior art do not necessarily render a new claim obvious. '(E)ven if all of the elements used are old, a new result, an unexpected result, a far more efficient result, or a more economical result will satisfy the requirements of patentability.'¹¹  Likewise, a 'claim that combines prior art elements can embody sufficient invention to be patentable, if the claimed invention represents 'progress' from the prior art.'¹² 

Conversely, 'a change only in form, proportions, or degree, the substitution of equivalents, doing substantially the same thing in the same way by substantially the same means with better results, is not such invention as will sustain a patent.'¹³ 

At the outset, we hasten to acknowledge that, as a court, we are not persons having ordinary skill in the art. Consequently, we may consider the unchallenged testimony of appellant's two expert witnesses, Dr. Wentorf and Dr. Seal, as well as that of the co-inventor, Mr. Lundblad. The record discloses these witnesses to be exceedingly well skilled in the art.¹⁴  We are guided by their expertise as we turn now to our task of applying the principles of law in the case before us.

Looking first to Wentorf, Dr. Wentorf does not coat an already preformed diamond with a radioactive metal. The Wentorf patent 'relates to radioactive diamond material and to the preparation thereof. More particularly, this invention is concerned with diamonds containing a surface coating of a radioactive metal * * *.' (Col. 1, lines 12--16). The radioactive diamond is formed by incorporating a radioactive material in the starting carbonaceous material and after subjecting the mixture to elevated temperatures and pressures, a diamond is formed which has radioactivity on the surface only. Of this surface Wentorf says: 'The thickness of the radioactive layer on the outside of the diamond varies to a minor extent with the proportions of ingredients employed in the reaction. However, regardless of the proportions employed, I have found that the thickness of the radioactive metal layer lies within the range of about 0.5 to 5.0 microns.' (Col. 4, lines 55--59).

Noting the words surface, layer, and coating, the District Court inquired as to how those words were used in the patent. Dr. Wentorf replied:

'Well, here it says that our claim 1, in the diamond material 'comprising a diamond having a radioactive adherent metallic surface of from about 0.5 to 5 microns in thickness.'

'That means that somewhere on that diamond is some radioactive adherent metallic material on the surface of that thickness. It doesn't mean that it is continuous. It means that it is here, there and everywhere. It could be 10 percent or 20 percent of the surface area and that would include the layer involving 10 microns deep which could be anywhere in that layer, you see.'¹⁵ 

In contrast, the Lundblad-Lindstrom invention is directed to the coating or cladding of diamonds. The individual (discrete) diamond particles are coated with a substantially continuous layer of metallic material formed on and enveloping the respective particles. The metallic layer conforms substantially to the outer surface configuration of respective particles and has a thickness of between 0.5 to 25 microns throughout the extent of the layer.

Erik Lundblad explained the irrelevance of the Wentorf method for use in coating metal clad diamonds in that Wentorf's radioactive material is 'present inside the surface, not on the outside of the surface layer.'¹⁶  Dr. Seal viewed Wentorf as giving no indication 'that the metal is adherent, that it may be in cracks or fissures in the surface, but it does not form a continuous coating.'¹⁷  Dr. Wentorf himself stated that his process does not produce a metallic coating or cladding of diamonds, but rather that the process produces diamonds just like naked uncoated diamonds, differing from them only in regard to the radioactivity. He further pointed out that since an integral part of the described Wentorf process is the separation of synthesized diamonds by means of fuming red nitric acid, '(o)ne skilled in the art does not expect a continuous metallic layer to be present after a treatment with acid of that kind.'¹⁸ 

The different objectives of the respective inventors accentuates the difference between the Wentorf reference and the claimed invention of Lundblad and Lindstrom. The purpose of Wentorf was to remove the static electricity from the diamonds.¹⁹  Lundblad and Lindstrom were seeking to improve the performance of resin bonded grinding wheels. Dr. Seal was likewise working on the improvement of resin bonded grinding wheels. Both Lundblad and Seal were familiar with the Wentorf patent, and neither felt that the Wentorf disclosure had any relevance to the work he was doing. It was the opinion of both that the incorporation of a radioactive metal in a resin bonded wheel, or any wheel, would make the wheel highly dangerous to the operator, and in a resin bonded wheel, the radioactive material would destroy the resin.

Despite this evidence produced at trial, the Patent Office, in its after-trial brief, argued that it was 'immaterial whether Wentorf actually produced such a surface coating since the patent clearly teaches a metallic surface layer of 0.5--5 microns thickness for diamond grains useful in an abrasive wheel,' and '(t)he reference is available for all it teaches to those of ordinary skill in the art.'²⁰  The District Court agreed, saying: 'This radioactive metallic surface coating around the diamond material lies within a range between 0.5 and 25 microns * * *,'²¹  and '(t)he fact that plaintiff asserts that Wentorf is not a proper reference because the metal used is radioactive and therefore undesirable because dangerous and thus inutile industrially does not in any way change its basic disclosure.'²² 

It appears to us that the Patent Office and the District Court misconstrued the terms 'surface coating' and 'surface layer' as used in Wentorf. It is our view that the language used in a patent must be measured and interpreted in relation to the disclosure of the patent as a whole. If there was any question during the proceedings before the Patent Office, it became clear at the trial that Wentorf does not disclose the coating or cladding of diamonds, and that one skilled in the art would not construe Wentorf as describing a diamond material coated with a radioactive material in the sense clearly contemplated by Lundblad and Lindstrom. The health hazard and the deleterious effect of radiation from radioactive material on organic resins aside, the record shows that diamonds produced according to the Wentorf disclosure would not improve in any way the grinding performance of the wheel. It follows that the intermittent thickness of Wentorf's interior surface layer of 0.5--5 microns does not reveal that an exterior metallic coating of between 0.5 and 25 microns would improve the performance of resin bonded grinding wheels.

The French patent of Soulard gives us more difficulty. For the purpose of improving abrasive tools such as grinding wheels, Soulard teaches the use of a metal coating on the abrasive grains to achieve uniform distribution of the grains and to improve the adhesion of the abrasive grains.²³  Similarly, appellant's purpose is to improve the performance of resin bonded grinding wheels by embedding metal coated diamond grains in a synthetic resin material. Thus the Patent Office contends that the only difference is in the thickness of the coating, and it is an obvious expedient to determine the optimum range of thickness which gives this desired result.

We agree that the most significant difference lies in the thickness of the coating. We do not agree that 'it would be mere routinization to thus arrive at an optimum range of thickness which would produce the claimed result.'²⁴ 

Soulard surrounds the abrasive grain with a spherical cover in order to obtain the most uniform possible distribution of the grains in the grinding tool--that is, equal distance of the grains from one another. The spherical metal cover is obtained by the diamond grains passing through a screen of 100 to 120 mesh per square inch, the grains being galvanically metal coated. This results in the grains receiving a diameter of 200 to 2000 microns. Thus, the cover thickness of each grain is different, varying even within one grain, and in places it may be far above 500 microns. The wide diameter variations, between 200 and 2000 microns, of the metal spheres obtained from grains with a diameter of 150 microns, indicate that Soulard was interested exclusively in obtaining a spherical cover over each grain--not a metal deposit purposely limited in its thickness. This would explain why in the Soulard patent there is no reference to the thickness of the covering.

The Patent Office calculates by simple arithmetic the diameter of the coating of the largest grain as per Soulard to be around 25.5 microns, submitting that the difference is after all insignificant from the 25 microns cited in some of appellant's claims. We find this argument to be untenable. Soulard is talking about enlargement in the thickness range from the minimum of 25.5 microns up to 937.5 microns, whereas the patent claims here are 0.5 to 25 microns. Judging, as we must, the invention of Soulard in its entirety, it would be inconsistent to base a comparison on extrapolated random values, equating the top of one range that has a significant criticality with the bottom of a distinctly different range.

The drawing in Soulard shows a thick cover. This is supported by the description which states that a spherical cover is to surround the small abrasive grains. There is a stark contrast between a spherical cover and the thin metal coating applied to notoriously sharp edged, multiangular diamond grains with a jagged surface. Further, it is essential to appellant's invention that the diamond grain retains its jagged form; hence, the surface must be metallized at equal cover thickness. Soulard makes a sphere of the jagged grain, changing its form by covering it with a metal cover differing considerably in thickness in places on the grain. In some cases thicknesses comparable with those of the subject of the application are obtained. However, the purpose attained differs completely from the one attained by the metallizing of the grain as shown in the application of Lundblad and Lindstrom.

Soulard directs the expert toward providing the diamond grains with the thickest possible metal coating for the purpose of uniform distribution of the grains, whereby such grains assume a ball shape. However, diamond grains cannot assume ball shape if the metal cover is very thin--0.5 to 25 microns. When such thin coats are applied there is no surface leveling of the unevenesses of the ragged synthetic diamond surfaces. It is specifically the maintaining of this ragged surface structure which accounts for the increased inherence of the metal covered diamond grains and better anchoring within the resin bond in the application herein.

Added to the aforesaid differences between Soulard and appellant's invention is appellant's showing that the 0.5--25 micron range is critical. It is an established principle of law that 'a limitation merely with respect to proportions in a composition of matter or process will not support patentability unless such limitation is 'critical."²⁵  "The term 'criticality' as used in the patent sense is not synonymous with invention, but rather, more precisely, is a device for determining the presence or absence of invention."²⁶ 

The Commissioner has steadfastly maintained that appellant cannot base patentability upon a showing of criticality. Pointing to the application-in-part which discloses a coating thickness range of 0.5--155 microns with a specific example giving a coating thickness of 100 microns, he asserts that there is no disclosure in the application that the claimed range of 0.5--25 microns is critical, and such a disclosure is required to support patentability of the claims on that basis. Actually, in the original application filed by Lundblad and Lindstrom on October 19, 1965, the thickness range disclosed was 'essentially 0.5--25 microns.'²⁷  Shortly thereafter, based on studies and theoretical considerations, but without further testing, Lundblad thought the invention might have a wider scope, and he filed a continuation-in-part, which added to the disclosure a thickness 'of between 0.5 microns and 155 microns, suitably between 0.5 microns and 25 microns and preferably between 5 microns and 20 microns.'²⁸  After additional testing, however, it became evident that the desirable range lay between 0.5 microns and 25 microns. On May 12, 1967, the claims were amended to conform to the original application.²⁹ 

In In re Saunders and Gemeinhardt³⁰  the Patent Office took substantially the same position which it has taken here. Citing prior decisions of the Court of Customs and Patent Appeals, the Solicitor stated that "it is well established law that claimed features, proportions or values which are not disclosed as critical or critically significant and/or are described in the application only as being merely preferred 'cannot be considered as critical', i.e., patentability cannot be predicated thereon."³¹  To this, Judge Rich, speaking for the Court, replied:

'(R)egardless of what this court may have said in the past, it is now our view that, in those cases where the applicants have attempted to rebut a showing of the prima facie obviousness of the subject matter claimed by the introduction of objective evidence of non-obviousness, both we are the tribunals of the Patent Office must give full consideration to the applicant's evidence and reach a decision on the question of non-obviousness on the basis of the relative strength of the applicant's showing and the prima facie case made by the Patent Office. (Citations omitted.) Appellants' evidence may not be disregarded simply because of the manner in which the now claimed processes were denominated in the original application. To rule otherwise would let form triumph over substance, substantially eliminating the right of an applicant to retreat to an otherwise patentable species merely because he erroneously thought he was first with the genus when he filed. (Citations omitted.) Since the patent law provides for the amendment during prosecution of claims, as well as the specification supporting claims, 35 U.S.C. § 132, it is clear that the reference to 'particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention' in the second paragraph of 35 U.S.C. § 112 does not prohibit the applicant from changing what he 'regards as his invention' (i.e., the subject matter on which he seeks patent protection) during the pendency of his application.'³² 

The application of this principle to the case herein is readily apparent.

The record established at the trial contains test evidence which clearly shows a coating thickness of 0.5 to 25 microns to be a critical range for improvement of grinding efficiency, and that there was no increase in efficiency above 25 microns. Fifteen charts were introduced into evidence depicting results in both wet and dry grinding, using diamonds coated with nickel, cobalt, nickel-tin, and cadmium (a soft metal), with a range of coating thicknesses from 0 (no coating) through 41.1 microns. The tests all show a maximum in the particular range of 0.5 to 25 microns; i.e., it increases to the maximum and then falls off. The maximum occurs between 5 and 20 microns and between 10 and 15 microns. In this maximum range, the grinding efficiency (G ratio) increased up to 200% to 300%. Of the 15 tests, in all cases but one, there was no 'glazing' until the thickness coating past 25 microns.³³  A wheel tested with diamonds having a coating thickness of 41.1 microns obtained no results for usage for grinding operations because the wheel 'glazed.' Erik Lundblad explained the term 'glazing' to mean that 'the metal which came from the coat was smeared out around the hub of the wheel, the periphery of the wheel and formed a continuous metallic film around the surface of the wheel.' This 'narrowed the fissures' to the extent the 'diamonds wouldn't protrude through that film' and not only would the wheel not operate properly, it would not cut at all.³⁴  Thus, the tests using a coating of over 25 microns up to 41.1 microns indicate that the Soulard range simply does not work in a resin bonded grinding wheel.

All of the elements used are old. Metal bonded and resin bonded grinding wheels were known to the art. Both natural and synthetic diamond was being used in each type of wheel. Interestingly enough, 'a thin coating of metallic copper was electrically deposited about a diamond and the diamond thus coated, was embedded in a metal matrix for attachment to a metal abrading or cutting tool as early as 1884, some 81 years prior to the appellant's date'³⁵  and 72 years prior to Soulard. But Lundblad and Lindstrom discovered that in a resin bonded grinding wheel, diamonds had to be coated within the 0.5--25 micron range to achieve the new result, the unexpected result, the far more efficient result, and the more economical result--the grinding efficiency of the wheel improved up to 200% to 300%. Further, metal clad diamonds within the critical range are not damaged during or after their embedment in the matrix wheel, and the grits remain fixed when subjected to high loads or temperatures.

Although the Patent Office did not regard the disclosure in the application of Lundblad and Lindstrom as sufficient, it has been held that 'the critical nature of a difference or of a limitation need not necessarily be pointed out in the specification, and it need not be expressly stated to be critical. It is sufficient if the evidence at the trial shows that the difference is critical and vitally significant.'³⁶  As is often the case, the record before us has the benefit of evidence introduced at the trial de novo which was not available to the Patent Office. It contains pertinent affidavits and test results which came after the decision affirming final rejection by the Board of Appeals. Although the District Court did not rule on the issue of criticality, we think appellant established beyond peradventure the criticality of the 0.5 to 25 micron range in the claims.

The record leaves no doubt that the invention satisfied a long felt need and its commercial success borders on dazzling. There was uncontradicted direct evidence at the trial that skilled scientists tried without success for a number of years to improve the grinding performance of resin bonded grinding wheels, and that such scientists reacted with skepticism and disbelief when informed of the Lindstrom-Lundblad discovery, and the fact that their discovery improved the grinding performance of resin bonded grinding wheels by as much as 200% to 300%.³⁷ 

The record is replete with evidence that the invention was hailed by the trade, including disinterested third parties and even competitors as being 'of major importance,' a 'milestone,' a 'major advance,' and 'a dramatic discovery and development.' Prior to 1965, metal clad diamonds were not commercially available. Yet the figures furnished by the Market Division of the De Beers Company show the 1971 market for resin bonded grinding wheels with unclad diamonds had recreased in value from approximately $36 million U.S. dollars to approximately $20 million, and the market for resin bonded grinding wheels incorporating metal clad diamonds had increased from 0 to $80 to $90 million.³⁸  Thus, at a market valued at $100 million, resin bonded grinding wheels incorporating metal clad diamonds pursuant to the invention of the application in suit account for at least 80% of such total market, and have a market value of approximately $80 million. On the basis of equal wheel cost, minimum total grinding costs were obtained through the use of metal coated diamonds.³⁹ 

It is our view that because the o.5 to 25 micron range is a critical range for coating abrasives used in a resin bonded grinding tool, thus achieving the new and unexpected result of an increase of the grinding efficiency up to 200% to 300%, appellant's invention is not clearly obvious, precluding consideration of the evidence of commercial success and a long standing need. Considering those factors along with the other evidence, we conclude that appellant's invention would not have been obvious to one of ordinary skill in the art within 35 U.S.C. § 103.

We have a substantial record and have carefully analyzed it in the light of the well presented arguments of counsel. Uncontroverted by any affirmative evidence from the Patent Office, the evidence in the record demonstrates the uniqueness and value of the claimed invention, and we are 'left with the definite and firm conviction that a mistake has been committed.'⁴⁰ 

Accordingly, we reverse and remand with the instruction that judgment will be entered for the appellant as to all claims in issue. The Commissioner is authorized to issue a patent.

So ordered.