This is an action commenced by Union Carbide and Carbon Corporation1 against Stuart Laboratories and William B. Hugle,2 alleging infringement of its U.S. Patent No. 2,488,507 for "Synthetic Star Rubies and Star Sapphires, and Process for Producing Same." Plaintiff is the assignee of inventors Burdick and Glenn. Defendants have appealed from a judgment of the district court which held all claims of the patent valid and infringed.

The making of synthetic sapphires and rubies is almost fifty years old. In 1904 Auguste Verneuil published an article disclosing his method, and in 1911 U. S. Letters Patent No. 988,230 and 1,004,505 were issued to him. The district court found that synthetic sapphires and rubies have been produced on a commercial scale in accordance with the Verneuil process for more than thirty years.

In an occasional rare natural ruby or sapphire, three bands of light intersect to form a star. There are probably fewer than 500 good natural star rubies in the world, and they are valued at about $1500 a carat. For long, the star effect seemed an insurmountable obstacle in man's never-ceasing attempt to reproduce nature. The secret of the star was not to be easily unlocked from nature's bosom. It had been believed prior to this invention that asterism in natural sapphires and rubies was caused by tiny needle-like inclusions of crystallites of titania (rutile) in the host crystal, the star effect being the result of reflection of light from the surfaces of the rutile inclusions. But no one prior to the inventors had discovered how to get the titania in the host crystal in the desired form. In 1946, when inventors Burdick and Glenn were attempting to find a simplified procedure for making plain blue sapphire, they stumbled on their discovery.

The starting point for the Burdick-Glenn patent is the Verneuil process. In the method disclosed by Verneuil, powdered alumna (known to the chemist as aluminum oxide, A1₂O₃, and to the mineralogist as corundum) is placed in a hopper which has a sieve in its base. A hammer intermittently strikes an anvil, sifting the alumina into an oxyhydrogen flame, which raises the alumina to its melting temperature (about 2050° C.). The melted alumina drops on a refractory support and slowly builds up into a long, narrow, approximately cylindrical crystal known as a boule. Ruby and sapphire are both basically alumina, the only difference being the coloring oxides which are added.

The inventors discovered that asterism can be produced in synthetic rubies and sapphires if boules are grown essentially in accordance with the Verneuil process and subsequently reheated. The first crucial requirement of the patent is that the powder used contain between 0.1% and 0.3% of titania. The second is that, after the boules are grown, they be reheated within the temperature range of 1100° C. to 1500° C. until a compound of titanium (probably titanium dioxide) precipitates along the prominent crystallographic planes of the crystal. As examples of suitable heating schedules, the specification prescribes 72 hours at 1100° C., 24 hours at 1300° C., and 2 hours at 1500° C.3 A heavy concentration of the titanium oxide precipitates along the surface of the boule in an area known as the skin while the inner portion of the boule is substantially free from titanium oxide. After reheating, the patent directs that a gemstone be cut en cabochon (flat base and convex crown, like a gum-drop) in such a manner that the base of the gemstone is perpendicular to the C-axis (optical axis) of the crystal, with the C-axis extending symmetrically through the center of the stone and through the center of the convex crown. When so cut, a six-ray star centered in the crown of the stone will appear.

The attack of the defendants on the judgment of the district court is a four-pronged one. First, they contend that in view of the prior art, there is no invention in the process or in the products. Second, defendants assert that the patentees failed to make a full disclosure of the essentials of the process. Third, it is contended that the product claims are defective in that the product is described in functional terms. Finally, we are urged to hold that even if the patent is valid, defendants have not infringed it.4

Defendants assert that both the use of titania and reheating at high temperatures were known in the prior art. It is true that titania was sometimes used in the prior art, but only as a coloring agent. While synthetic gems had been subjected to reheating in the prior art, the primary purpose of such reheating was to improve the quality of the stones. But we think it took invention of a high order to discover the critical amounts of titania to be used, to determine that reheating would result in the precipitation of titania, and to ascertain the critical limits in time and temperature for the reheating. The process in its entirety was new, and the result produced is significant enough to meet the strictest standard of invention.

Principal reliance is placed by defendants on the fact that for many years an effect known as chatoyancy had been produced in synthetic spinel. Chatoyancy in spinel, defendants assert, is closely analogous to asterism in corundum. Defendants' thesis, of course, rests on a double analogy: that spinel is similar to corundum and that chatoyancy in spinel is similar to, if not the same as, asterism in corundum. The record reveals that synthetic spinel is a compound of aluminum oxide and magnesium oxide (MgOal₂O₃), while corundum is just aluminum oxide. Chatoyancy, described by some witnesses as a cloudy, milky appearance and by others as a wavy appearance of light, is caused in spinel by the precipitation of aluminum oxide along the crystal lines or planes of the host crystal. The closeness of the analogy between asterism in corundum and chatoyancy in spinel was, at least in part, a disputed factual question. Defendants' expert testified that chatoyancy and asterism "are really the same thing," while plaintiff's expert termed asterism in corundum "only vaguely similar" to chatoyancy in spinel. In the circumstances of this case, a determination of invention is thus primarily a question of fact, and there is certainly ample evidence to sustain the finding of the trial court that it required patentable invention to produce the star stones described in the patent. See Graver Tank & Mfg. Co. v. Linde Air Products Co., 1949, 336 U.S. 271, 69 S. Ct. 535, 93 L. Ed. 672.

Upon looking back after a discovery is made, it is usually not difficult to find possible analogies in the prior art, for it is axiomatic that progress is always step by step. In applying the standard of patentable invention, our task is to draw the line between the tiny step forward and the large stride. This invention, we feel, is a significant stride forward.

Defendants' second contention is that the inventors failed to make a full disclosure of the essentials of the process, thus breaching their duty under R.S. § 4888 which directs that the inventor describe the invention in "such full, clear, concise, and exact terms as to enable any person skilled in the art or science to which it appertains * * * to make, construct, compound, and use the same * * *." 35 U.S.C.A. § 33. It is on this argument that defendants place the greatest stress in this appeal.

It is acknowledged that plaintiff, in its commercial production, engages in certain practices which are not mentioned in the patent. In growing its boules, plaintiff employs orientation control. Orientation, as the term is used in the art, is the angle between the growth axis and the C-axis or optical axis. Orientation is controlled by the use of a seed, which is merely a tiny corundum crystal. When no seed is used in growing a boule, any orientation ranging from 0° to 90° may result. When a seed is used, however, the boule will always take the orientation of the seed, thus allowing one to control the orientation of the boule by choice of a seed of pre-determined orientation. In its early commercial production, plaintiff used 90° orientation almost exclusively.

Another technique practiced by plaintiff is annealing which, as the term is used in this case, is the reheating of a boule to about 1900° C., just below its melting temperature. Annealing was known in the art prior to the invention here in question, and was used to relieve internal strains and stresses so as to prevent shattering of the boule upon cutting.

Defendants contend that 90° orientation (orientation necessarily includes seeding) and annealing are essential to the production of star stones of gem quality and that inventors Burdick and Glenn knew this at the time the patent was applied for. Proof that these steps were essential to the success of the patented process rests primarily on the testimony of defendant Hugle and upon that of Dr. Sossman, professor of ceramics at Rutgers University. Dr. Sossman prepared a number of lucite models designed to demonstrate that even though asteriated boules could be produced by the patented process, it was impossible to cut gem quality stones from such boules unless 90° orientation and annealing were also utilized.

The basis of Dr. Sossman's thesis is that since titania is substantially concentrated in a relatively thin skin, a centered six-legged star stone can be produced only if a stone is cut entirely from the skin, or if it is so cut that substantially the entire crown is taken from the skin of the boule. A stone cut entirely from skin will, in the expert's opinion, be too small to be of gem quality. On the other hand, he attempts to prove that annealing and 90° orientation are both necessary if a larger stone is to be cut with the crown composed of skin. His reasoning to support this proposition is as follows: An unannealed boule will almost invariably split longitudinally along a plane defined by the growth axis and the C-axis. Hence, when a gem is cut en cabochon with its base perpendicular to the C-axis, even though there be 90° orientation, only one-half the crown will be composed of skin, the other half being taken from the interior portion of the boule. Such a stone, in the opinion of Dr. Sossman, will show an incomplete star. As the orientation angle decreases, a stone cut en cabochon with its base normal to the C-axis will have a crown containing less and less skin. Dr. Sossman's testimony was not entirely theoretical. He also testified that tests were made at the Stuart laboratory under his supervision in an endeavor to determine whether star sapphires and star rubies of gem quality could be made by following the patented process. The results were almost a complete failure.

Defendant Hugle, who had been employed by plaintiff as plant chemist, testified that he knew the patented process, but was unable to produce successful results without annealing and 90° orientation. Defendants also point to several idea records of inventor Burdick and an intra-company patent memorandum wherein the desirability of annealing and 90° orientation is mentioned.

Experts testifying for plaintiff sharply contradict the above evidence. They are unanimously of the opinion that annealing and 90° orientation are not essential and that star stones of gem quality can be produced by one skilled in the art if the patented process is followed. Further, there is substantial evidence that star sapphires and star rubies of gem quality had actually been produced by following the patented process. Plaintiff's experts assert that the thesis of Dr. Sossman is erroneous because it is based on the assumption that there is titania only in the skin, whereas there is in fact some titania in the interior of the boule, although in lesser concentrations. There is no discontinuity in the distribution of titania, we are told, but a mere lessening in concentration as the interior of the boule is approached.

While the evidence produced by defendants may be persuasive, that of plaintiff is to the contrary, and the trial court has accepted plaintiff's version. We can hardly say that its findings on these hotly disputed questions of fact are clearly erroneous. See Rule 52, Federal Rules of Civil Procedure, 28 U.S.C.A. The finding that the description of the process in the patent is sufficient to enable a person skilled in the art to grow star rubies and sapphires successfully is supported by adequate evidence and should be sustained. See Graver Tank & Mfg. Co. v. Linde Air Products Co., supra; Bank v. Rauland Corp., 7 Cir., 1944, 146 F.2d 19.

Defendants argue, alternatively, that even if annealing and 90° orientation are not vitally essential to the production of gem quality stones, they are nevertheless highly important steps and result in greatly increased productive efficiency. Defendants urge that full disclosure under R.S. § 4888 means not merely the revealing of bare essentials, but the disclosure by the patentee of the best methods then known to him. See Stelos Co. v. Hosiery Motor-Mend Corp., D.C.N.Y.1932, 60 F.2d 1009, 1011.5 But see Sewall v. Jones, 1875, 91 U.S. 171, 185-186, 23 L. Ed. 275.

R.S. § 4920, 35 U.S.C.A. § 69, allows a defendant in an infringement action to plead as a defense that the patentee, for the purpose of deceiving the public, filed a description and specification which contained less than the whole truth relative to the invention. Whether R.S. § 4888 or R.S. § 4920 is relied on is probably not of practical importance, for where the evidence is clear that the patentee omitted elements which he knew were important, fraud will be implied by the fact-finder. Electric Boot and Shoe Finishing Co. v. Little, C.C.Mass., 1896, 75 F. 276, affirmed on other grounds, C.A. 1, 1905, 138 F. 732. While there is evidence that annealing and 90° orientation were considered desirable, inventor Burdick testified that at the time of the patent application, the wisdom of both annealing and 90° orientation was seriously questioned. On the basis of this evidence, the trial court made a specific finding that at the time of the patent application, it was not determined that annealing was a preferable procedure. It also found generally that there was no fraudulent withholding of information. This finding, we believe, was made in the light of R.S. § 4920 and thus constitutes a finding that the patentees disclosed "the whole truth relative to [their] invention".

The third point raised by defendants is that the product claims are inexact and indefinite. Their principal argument here is that the term "gem quality" used in six of the product claims is a functional description. Claim 11 is typical of the product claims under attack: "An asteriated massive nongranular single crystal of synthetic corundum of gem quality containing a precipitate of oxide of titanium, and characterized by having curved growth lines internally thereof."

Defendants place considerable emphasis on Holland Furniture Co. v. Perkins Glue Co., 1928, 277 U.S. 245, 48 S. Ct. 474, 72 L. Ed. 868, wherein the Supreme Court condemned "the attempt to broaden product claims by describing the product exclusively in terms of its use or function". 277 U.S. at page 257, 48 S.Ct. at page 479 (emphasis supplied.) That case is clearly inapposite here. Not only is the term gem quality a definite term, but its use in the product claims operates not to broaden the claims, as in the Holland case, but actually to narrow them. Gem quality is defined in the specification as a crystal "sufficiently perfect to warrant cutting, polishing, and offering for sale as an ornament." The testimony of an expert in the jewelry trade is that this term has a definite meaning in the trade. Further, we are convinced that the effect of the term is to restrict the claims of the invention to the field of ornamentation rather than to industrial uses. It is highly unlikely that star sapphires would possess any additional utility to industrial users; hence, the term gem quality merely limits the product claims to the area of usefulness.6 Cf. Dennis v. Pitner, 7 Cir., 1939, 106 F.2d 142, 146 certiorari denied 308 U.S. 606, 60 S. Ct. 143, 84 L. Ed. 507; Smith, Kline & French Laboratories v. Clark & Clark, 3 Cir., 1946, 157 F.2d 725, 729 certiorari denied 329 U.S. 796, 67 S. Ct. 482, 91 L. Ed. 681.

The significance of the language in the Holland Furniture case was brought into sharper focus in General Electric Co. v. Wabash Appliance Co., 1938, 304 U.S. 364, at pages 371-372, 58 S. Ct. 899, at page 903, 82 L. Ed. 1402, where the court declared: "A limited use of terms of effect or result, which accurately define the essential qualities of a product to one skilled in the art, may in some instances be permissible and even desirable, but a characteristic essential to novelty may not be distinguished from the old art solely by its tendency to remedy the problems in the art met by the patent."

The above quotation is particularly applicable to the product claims now before us in view of the expert testimony that the term gem quality is considered by those skilled in the art to be an accurate description of one important quality of the product. The use of gem quality is, at most, "a limited use of terms of effect or result;" for, unlike the claims involved in the Holland Furniture and the Wabash cases, functional language is not employed at the point of novelty.

Finally, defendants contend that they have not infringed the patented process because they reheat their boules only incidentally as a part of an annealing cycle. The record reveals that defendants first anneal their boules by heating them to about 1900° C. and then lower the temperature slowly. During such lowering the boules are in the range of 1500° C. to 1100° C. long enough for the titania to be precipitated. We believe there is sufficient evidence to support the trial court's finding of infringement. The law is clear that an addition to a patented process does not avert infringement, even where the addition is an improvement. 3 Walker on Patents p. 1737 (Deller's ed. 1937). Defendants infringed the patented process when they used the critical amounts of titania and operated within the temperature range of 1500° to 1100° C. long enough to precipitate the titania.

The judgment of the district court will be affirmed.