Springs Cotton Mills, Inc. v. Hall Laboratories, Inc.

SOPER, Circuit Judge. This is a suit for infringement of United States patent No. 2,337,856 which was granted to Owen Rice and George B. Hatch on December 28, 1943 upon an application filed October 27, 1942. The plaintiff is Hall Laboratories, Inc., of Pennsylvania, assignee of the patent, and the defendant is The Springs Cotton Mills, Inc., of South Carolina for which Rumford Chemical Works of Rhode Island, supplier of the chemical used in infringement of the patent, has assumed the defense. The only defense is invalidity of the patent based on the ground that, although the patentees made a valuable discovery, they did not contrive a new procedure to put it into effect. The District Judge rejected this defense. He held that the patent was valid and infringed, and the defendant appealed. 112 F.Supp. 29. The patent relates to a process of retarding the corrosion of metal by water which is especially notable in the corrosion of iron pipes in water distribution systems. Iron rust in the pipes contaminates the water or forms extensive growths called tubercles which decrease the carrying capacity of the lines. The problem is encountered in many parts of the country where the water is corrosive by nature. In 1913, Baylis, a water chemist, who was studying the corrosion problem, found that it was helpful to add an alkali to the water so as to supersaturate it with calcium carbonate and thereby cause the formation of a protective deposit or scale on the interior surface of the pipes. This procedure, however, had disadvantages in reducing the carrying capacity of the pipes, notably in hot water pipes and heaters, and in increasing the costs in industries where lime, the cheapest alkali, cannot be used. Further scientific research resulted in the adoption of a mathematical formula, devised by Langelier in 1936, relating to the saturation level of calcium carbonate in the water; and waters have since been classified as Langelier positive, negative or zero, so as to indicate that they *501 are supersaturated and thereby scale forming, or undersaturated and therefore tend to dissolve scale, or merely saturated with no tendency in either direction. Natural waters, which have a positive Langelier, are scale forming and if they are also corrosive, that is, if they contain dissolved oxygen, as most natural waters do, the scale protects the pipes. In certain situations, however, scale builds up too rapidly and reduces the efficacy of the distribution system of the appliances to which the water is supplied. It was found that the addition of small amounts of a metaphosphate, such as sodium metaphosphate, would slow down the formation of scale; but it was also recognized that in the case of corrosive water the maintenance of a layer of scale sufficient to protect the pipes was necessary. The process of the Rice and Hatch patent in suit was devised to meet this need. For its operation it is necessary that the water be corrosive, that it possess a calcium content so limited that it has a negative Langelier Index, and that the metaphosphate be added in an amount which bears a definite relation to the aalcium content of the water, that is two formula weights 1 of the metaphosphate for each formula weight of calcium. When these requirements are followed a submicroscopic iridescent film is formed upon the inner surface of the pipes which forms a protective covering and prevents corrosion. The metaphosphate constituent of the film gradually dissolves in the water but the film may be maintained by the continued addition of small amounts of the chemical. The advantages of the patented process are that the film does not diminish the capacity of the pipes and that inexpensive alkalies may be used in the operation. Accordingly the new method has had great commercial success. It was exploited commercially by the plaintiff as soon as its efficacy was proved and the records show that it is in use in 598 municipal plants and 1300 industrial plants. It has been notably successful in the textile industry in which it was formerly necessary to use the relatively expensive soda ash rather than lime in the practice of the Baylis system. The defense is based on the contention that the precise step of the patent was old in the prior art, and that the only novelty was the new effect which was not patentable although it gave a new and unsuspected value. Three prior water treatments are relied on which were disclosed by (1) the Hall and Jackson patent No. 1,903,041 of 1933, (2) the Hall reissue patent No. 19,719 of 1935, and (3) the so-called threshold treatment shown in the Rosenstein patent No. 2,038,316 of 1936 and the Fink and Richardson patent No. 2,358,222 of 1944 which was granted upon an application prior to the application of the patent in suit. The Hall and Jackson patent relates to the treatment of steam boiler water by chemicals to prevent the formation of adherent boiler scale. The treatment involves the continuous addition of meta-phosphate to natural or pretreated water fed into the boiler, the amount of the addition being gauged by the amount of calcium carbonate therein. The significance of the disclosure is said to reside in an example given in the specifications of the Hall and Jackson patent in which the amount of the metaphosphate added falls within the range of the process described in the Rice and Hatch patent in suit. It is evident, however, that the Hall and Jackson patent, which was well known to the industry, did not disclose the corrosion preventive process contained in the patent of Rice and Hatch. Hall himself, when first informed of their discovery, did not credit it and the Patent Office did not find it anticipatory. Hall and Jackson were not dealing with the problem of corrosion. On the contrary they expressly taught that the content of dissolved oxygen in the water, *502 that is the corrosive producing element, should be made as small as possible in their method, and they based their discussion and illustrations on the assumption that provision had been made for satisfactory deaeration of the water. Moreover, the uneontradicted evidence in the present case shows that Hall and Jackson contemplated the use of water with a positive Langelier Index in which the Rice and Hatch process is inoperable. The Hall reissue patent of 1919 relates to the softening of hard waters containing calcium by adding an alkali meta-phosphate, such as sodium metaphos-phate. It teaches that when sodium metaphosphate is used in the ratio of two formula weights of meta to one formula weight of calcium, calcium meta-phosphate is formed; but at this point the softening process is not complete and an objectionable precipitate which clouds the water is produced and it is necessary to add at least one additional formula weight of sodium metaphosphate. When this is done a soluble complex results without the formation of the undesirable precipitate and without increasing the alkalinity in such a way as to attack the skin of the user’s hands or the fibers of fabrics that are being washed. The defendant contends that in these statements Hall makes the incidental disclosure of the addition of metaphosphate in amounts up to two formula weights for one formula weight of calcium, that is, the very amounts covered by the Rice and Hatch process. Hence it is said that the two processes are the same, and that Rice and Hatch merely discovered that when the ratio of 2 to 1 is maintained, a protective film is formed on the surface of the metal. The variation between the teaching of the two patents both as to purposes and as to results is however obvious. Hall does not teach that the operator, in making the experiments which are necessary to determine the amount of meta to be added to the water for softening purposes, should use corrosive water and metal containers. The Rice and Hatch effect is not obtained in non-corrosive water and their protective film which forms on metal will not form on glass. Hall teaches that a 2 to 1 mixture will produce a cloudy precipitate while Rice and Hatch show that under these conditions an iridescent film is formed on the surface of the metal. This film is the final result of the Rice and Hatch process-while Hall goes on to create a soluble-complex which remains in the water. The evidence furnishes nothing more than a conjecture as to whether Hall at any time actually performed the Rice and Hatch process. The term “threshold treatment” has been given to another use of metaphos-phate in the treatment of water later than that described in the above mentioned patents but prior to the Rice and Hatch invention. The defendant regards this practice as its major defense in the present suit. Workers in the art observed that if meta is added to water in much smaller quantities than those necessary to produce the soluble complex of Hall, the precipitation of calcium carbonate and the formation of scale was prevented. Use of the term “threshold” was made because the small amounts required were on the threshold of those used by Hall in his water softening process. This method of scale control was extensively practiced by the plaintiff’s organization and by others in the industry, especially where the scale forming tendency was accelerated by heating. The practice was suggested in the Rosenstein patent of 1936 No. 2,038,316 which was concerned with the addition of large amounts of ammonia to irrigation water as a means of getting nitrogen into the soil. He found that large deposits of scale resulted from the addition of the ammonia. If 0.5 to 1 part per million of metaphosphate were added, the formation of scale was prevented. He said that the treating reagent may be added to the water at the same time as the ammonia but if metaphosphate is used it is preferable to introduce it at a point upstream from the point of ammonia injection. *503 Fink and Richardson disclosed the same method in their patent No. 2,358,-222 of 1944 which was applied for prior to the Rice and Hatch application. They dealt with industrial water systems where the water becomes heated at various points and causes scale deposits. They also prescribed for the prevention of scale formation the addition of a few parts per million of metaphosphate to “raw” or “untreated water” and they said that the metaphosphate may be injected into the supply line or into the tank or reservoir. It will be observed that the threshold treatment was designed to deal with the prevention of scale formation and not corrosion. Here again the position of the defendant is that the practice of Rosenstein and of Fink and Richardson was identical with that of Rice and Hatch insofar as the amount of metaphosphate is concerned. The defendant concedes that there is an important difference between the threshold treatment and the method of Rice and Hatch in that the metaphosphate does not do its work of preventing scale unless the water is Langelier positive, that is over saturated with a calcium and therefore capable of producing scale, whereas the Rice and Hatch process for preventing corrosion will not work unless the water is Lange-lier negative. The defendant says that The Rosenstein and Fink and Richardson -patents incidentally suggest that the meta may be introduced into the water .as it is supplied in its natural state and before it has become over saturated through heating or through the addition •of alkali; and since water in its natural state is ordinarily corrosive and Lange-lier negative there were present in this earlier process all the elements of the Rice and Hatch invention, and conse- . quently a protective ñlm must have been formed upon the surface of the interior of the pipes and corrosion must have been prevented, whether the operators knew it or not. The argument in short is that Rice and Hatch did not discover .a new method but only a new result ■which is not patentable. The fact remains, nevertheless, that prior to Rice and Hatch none of the users of the threshold treatment showed that the addition of small amounts of meta to water will prevent the corrosion of the metal pipes of a distribution system provided the water is both corrosive and Langelier negative. Nor did the earlier users of meta whether patentees or not instruct the public by practice or disclosure as to the essential features of the Rice and Hatch process. They did not make it clear that waters which are both corrosive and negative must be used, and even though these factors were conjoined in some part of their operations, it was done as a preliminary part of their plan without clear recognition of the nature of the operation or its utility in the prevention of corrosion. Bearing in mind that Richardson and Fink were directing attention to the treatment of overly saturated waters so as to prevent the formation of scale, it cannot be said that their preliminary incidental use of natural waters, which in most instances but not always are corrosive and Langelier negative, would impart to persons skilled in the art the knowledge that waters of this kind must be used in the process or the knowledge that the preliminary step was a complete process in itself. The advance in the art which constitutes the substance of the present invention was not in our opinion derived from prior practices or prior patents. It is true that Rice and Hatch were familiar with the Baylis system and with the practice of using metaphosphate to control scale formation. Starting with this background they made a series of experiments in their study of the corrosion problem using water treated with lime-soda in glass beakers with and without the addition of the metaphosphate, but the results were inconclusive. Finally they set up an apparatus simulating an actual distributing system and subjected certain metal pipes to contact with tap water that was treated with lime and soda to produce the deposit of a protective scale and other pipes in which they used ordinary tap water without the ad *504 dition of the chemical. In both sets of pipes they used metaphosphate and found to their surprise that there was more corrosion in the pipes in which the water had been treated with the protective alkali, than in the pipes in which no such protection was used. They also found in the latter pipes an iridescence on the inside surface which was finally identified as the protective film formed by the reaction between the metaphosphate and the calcium in the water. The terms and conditions of the process covered by the patent in suit were derived from these experiments and not from prior practices or patents. The principles of law applicable to these facts are well established. It is settled, as Judge Parker said in Allen v. Coe, 77 U.S.App.D.C. 324, 135 F.2d 11, 12, that a patent may not be granted for scientific explanations or discoveries of new uses for or unexpected merit in old substances or processes. See also O’Reilly v. Morse, 15 How. 61, 14 L.Ed. 601; General Electric Co. v. Jewel Incandescent Lamp Co., 326 U.S. 242, 247, 66 S.Ct. 81, 90 L.Ed. 43; Davison Chemical Corp. v. Joliet Chemicals, Inc., 7 Cir., 179 F.2d 793. The doctrine is graphically illustrated in the early ease of Morton v. N. Y. Eye Infirmary, 17 Fed.Cas.No. 9,865, p. 879 upon which the defendant, especially relies. There the patent covered the administration of ether by inhalation to create insensitivity to pain during a surgical operation; but the-patent was held invalid because the process of inhalation of ether to induce-stupefaction was old and hence the restatement of the process to cover the-new result was beyond the scope of the-patent law, notwithstanding the inestimable value of the discovery to mankind. That case, however, does not govern the present controversy; for there the prior-practice was clearly established, while-here the information available to Rice and Hatch, either from the practices of the industry or from prior patents, lacked-, the certainty and precision which would, enable the process to be performed and would serve as an anticipation of the patent. See Skelly Oil Co. v. Universal Oil Products Co., 3 Cir., 31 F.2d 427; Badische Anilin & Soda Fabrik v. Kalle & Co., 2 Cir., 104 F. 802, 808. Tilghman v. Proctor, 102 U.S. 707, 726, 729, 26 L.Ed. 279. Affirmed. Footnotes [1] The formula weight of a compound is the sum of the atomic weights of the component atoms.