Esso Research & Engineering Co. v. Kahn & Company, Inc., 379 F. Supp. 205 (D. Conn. 1974)

*206 Bruce W. Manternach, Hartford, Conn., Robert I. Pearlman, Esso Research & Engineering Co., Linden, N. J., for plaintiff; Richard K. Parsell, Malvin R. Mandelbaum, Richard A. Huettner, New York City, of counsel.

Roger B. McCormick, William M. Pomerantz, Hartford, Conn., for defendants.

BLUMENFELD, District Judge.

Esso Research and Engineering, plaintiff in this action, is a wholly owned and controlled affiliate of the Exxon Corporation (formerly known as Standard Oil of New Jersey, Inc.). Plaintiff is incorporated under the laws of the State of Delaware and has its principal place of business in New Jersey. It is engaged in research and engineering work for Exxon and its parent's other affiliated companies. It does not manufacture or sell any products to third parties. Plaintiff does, however, have a combined patent and licensing department which licenses technology originating or controlled by the company.

Defendant Kahn and Company, Inc. is a corporation located in Wethersfield, Connecticut, where it maintains its main offices and also has manufacturing facilities. Kahn and Company, Inc. is engaged in the manufacture and sale of various commercial and institutional driers and gas separators as well as other test equipment and measuring devices.

Defendant Chandler-Evans, Inc. is incorporated under the laws of the State of Delaware and maintains a regular and established place of business in West Hartford, Connecticut, where it operates a Kahn and Company heaterless drier for the removal of moisture from air.

Plaintiff, as assignee of Charles W. Skarstrom, is the owner of U.S. Letters Patent 2,944,627 for Method And Apparatus For Fractionating Gaseous Mixtures by Adsorption. Plaintiff alleges that defendant Kahn and Company, Inc. has induced infringement of plaintiff's patent by its manufacture and sale of infringing devices to defendant Chandler-Evans, Inc., and has itself infringed and contributorily infringed said patent by its manufacture and sale of heaterless driers within the jurisdiction of this Court.

This action arises under 35 U.S.C. §§ 271 and 281. This Court has jurisdiction under 28 U.S.C. § 1338(a). The Court is a proper venue for this action by virtue of 28 U.S.C. § 1400(b).

Entitled "A Method And Apparatus For Fractionating Gaseous Mixtures by Adsorption," the Skarstrom U.S. Patent No. 2,944,627 describes a process for the separation of one gas from a mixture of gases through selective adsorption by performing a combination of specific designated steps in a particular sequence and manner under defined conditions. Defendants admit that they perform each and every step described and defined in the Skarstrom patent in the *207 same sequence and manner and admit that they obtain the same result as specified in claims, 9, 12, 14, 16, 17, 24 and 34. These are method claims. Both defendants Kahn and Company, Inc. and Chandler-Evans, Inc. have entered into a stipulation stating that if said claims are valid in accordance with the patent laws then the claims are infringed by heaterless driers manufactured and sold by defendants.

Defendants challenge the validity of the claims in suit on the ground that they are invalid because they do not satisfy the requirements of non-obviousness as set forth in 35 U.S.C. § 103, see note 4, infra, in that the subject matter of the Skarstrom invention "and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains."

The invention of the patent in suit was made in September 1956 at plaintiff's laboratories. At the time of the invention various methods for the separation of one gas from another were well known and long in existence.

The specification of the '627 patent commences with a general description of the invention as being one which describes a method and apparatus for removing one gas from a mixture of gases. It then goes on to describe that "one specific adaptation of the invention relates to the drying of a gaseous mixture, such as air by removal of water vapor." Although the patent in suit is in no way limited solely to the removal of water vapor from air the patentee selected this example to describe and explain the operation of his invention.^[1]

Dr. Charles W. Skarstrom, a graduate chemist from Columbia University and the University of Virginia, was employed by the plaintiff for approximately 27 years prior to his retirement in 1970. During that time he had been studying and devising ways to separate gases and measure the separations accomplished thereby. In 1956, immediately prior to the invention in suit, Dr. Skarstrom was working on a water analyzer and also on an analyzer to determine the composition of bottled gas during its manufacture. Both of these projects required dry air, and, consequently, Dr. Skarstrom had conventional, commercially available driers in his laboratory at the Bayway Refinery in Linden, New Jersey.

These particular driers were of the then common, everyday type of commercial drier, known as the heat regenerating drier, operating on twelve-hour cycles. They removed water from the air by passing moisture-laden air through an enclosed bed of desiccant at high pressure where the desiccant absorbed the moisture; the dry product air, still at high pressure, continued on through a pipe leading from the top of the bed. Although this drier had two beds of desiccants, and was referred to as a two-bed drier, it is more accurately described as a pair of driers, each a mirror image of the other, each being used alternately to remove water vapor from separate feeds of moist air, and so connected to each other as to provide reciprocally a portion of the dried product gas to each other. Before the time that one bed became saturated with water, the stream of air to be dried was switched to a feed pipe leading to the other bed. Meanwhile, heat was used to evaporate the moisture out of the saturated *208 desiccant. The most common method employed a heater within the desiccant itself.

Dr. Skarstrom's unit had such internal heaters. During regeneration of a saturated desiccant bed the heaters, in conjunction with a counterflow of some of the dry product gas called "purge gas," would remove the accumulated moisture from the desiccant bed leaving it ready for a subsequent adsorption operation. Thus, while one bed was adsorbing, the other one was being desorbed.

One day in 1956 when Dr. Skarstrom was conducting his experiments on analyzers, the heater in one of the drier beds burned out, thereby incapacitating the unit's normal operation. This incident led Dr. Skarstrom to decide that he would attempt to operate the drier without a heater.

The particular drier in question was a two-bed, heat regeneration drier manufactured by Trinity Corporation and similar to the Skarstrom driers shown in P.Exs. 4A and 4B (see Appendix). The normal cycle time for this Trinity drier was 12 hours, to wit, 6 hours on stream adsorbing, 3 hours desorbing by heat regeneration and counterflow of purge gas, and 3 hours for a resting or a cooling down period. The first thing Dr. Skarstrom did with his broken down drier was to replace the timer element. He used a gear he happened to have in his laboratory to decrease the cycle time from 12 hours to a markedly shorter cycle of 30 minutes on adsorption and 30 minutes on desorption (without heat), and eliminated the cooling period as unnecessary. Dr. Skarstrom found that the drier could in this fashion do some drying even without the use of heat. Dr. Skarstrom next reduced the 30-minute cycle to 20 minutes and found that the product was even drier.

Dr. Skarstrom then directed his attention to the amount of purge gas that was used during the regeneration step. The gas not diverted for use as purge gas continued on under pressure as product; the diverted purge gas passed through a valve to the pipe at the top of the bed to be desorbed, where it was reduced to atmospheric pressure. It then flowed down through the desorbing desiccant bed and out into the atmosphere. The drier as provided to Skarstrom by the Trinity Corporation specified that about 3% of the product gas be used for purge. When in normal use Dr. Skarstrom observed that with the heater in operation the drier actually required only about 2% of the dry product gas to be diverted in order for that relatively small stream of purge gas to effect the complete desorption of water from a desiccant bed in conjunction with the heating of that bed. In modifying his disabled drier, Dr. Skarstrom increased the amount of purge gas diverted from the flow of product gas by substituting a new purge valve system which stepped up the purge flow 25-fold, so that fully half of the product gas was diverted for use as purge gas. By further experimentation, he determined that satisfactory results could be obtained through diversion of only 25% of the product gas as purge gas.

Dr. Skarstrom further refined his apparatus by changing the purge piping arrangement to provide an even quantity of purge gas at a steady flow rate. He continued to reduce his cycle time, to 11 minutes and then down to 3 minutes, until eventually, in his words, "the product became quite dry and I was very pleased." He termed the Trinity heater-type drier on which he had experimented "very much like" the model heaterless drier of the patented apparatus used to display the patented method, "only it had heaters in it."

In brief, the method described in the patent in general terms contemplates the removal of water vapor from air^[2] in a drier containing a pair of desiccant *209 beds. Each of these are used in the same way to adsorb water vapor, but only one at a time is being fed moist air under pressure. In operation, the moist feed air is introduced at high pressure through a pipe to the bottom of one bed, (bed on the right side in Ex. 4B shown in the Appendix). As it flows through that bed it is dried by the desiccant's adsorption of its water vapor, and it is then passed along as dry product air through another conduit leading from the top of the bed, still at high pressure. A valve in that conduit diverts a portion of the dry product air to the pipe leading from the top of the drier on the left side where it is relieved of pressure, flows down through the pipe at the top of the desiccant bed on the left side, through the bed and then on through the conduit below that bed and out to the open air. As it flows down through that bed it desorbs previously collected water vapor from the desiccant in that bed, again becoming moist air, and is discharged.

In continuing operation of the apparatus, the left and right beds are alternated on adsorption and desorption at the selected cycle time, cycle after cycle. The cycle time selection is quite flexible and will vary with conditions such as the size of the beds, the desiccant being used, the key component (such as water vapor) being adsorbed, the gaseous mixture (such as air) from which it is being separated, and the total amount or degree of separation desired.

Dr. Skarstrom stressed that it is important to select a cycle time "at less than that required to secure complete saturation" of the adsorbing bed. If complete saturation of the adsorbing bed occurs, it would no longer adsorb the water vapor and the process would fail. Since some heat is given off into the bed on adsorption the cycle time should be sufficiently short to permit the process to function without the addition of heat, as from an imbedded electrical heater, to aid the desorption. Thus, in order to make the process work heaterlessly, the adsorbing bed must be switched to desorption before it loses the heat of adsorption, so that the heat of adsorption may be employed to assist the desorption. In other words, the cycle time should be sufficiently brief so that there is no significant change in temperature in either of the beds when adsorbing and being desorbed. The patent states that it can be on the order of seconds or minutes.

This concept of maintaining a substantially constant temperature during adsorption and desorption dictates that during each cycle the same amount of water vapor must be purged or desorbed from a bed as was just adsorbed by the bed. This will insure that the same amount of heat will be given up by a bed during its desorption cycle as was generated during its immediately preceding adsorption cycle. If this equilibrium is not maintained cycle after cycle, heaterless operation will eventually fail because an incremental temperature change in one direction or the other will ultimately make a bed either too hot for successful adsorption or too cold for successful desorption.

The plaintiff's brief has translated into more easily understandable language the specialized jargon used by the patent attorney in stating what in the above described method is claimed as being patentable.

The Skarstrom invention is:

"A completely self-contained, self-regenerating gaseous fractionating process utilizing a selective absorbent in which the absorbent bed is the sole agent for fractionation and heat exchange which comprises:

(1) a repetitive adsorption/desorption process utilizing differences in pressure;

(2) wherein gaseous feed is introduced at one end of the adsorbent bed and product effluent recovered at the other end;

(3) a part of the product effluent is used to desorb the bed (or essentially *210 the same product effluent from a paired companion bed on a complimentary cycle);

(4) by counter-current flow at lower pressure than the adsorption pressure;

(5) with cycle times sufficiently short so that heats of adsorption/desorption are substantially retained and balanced within the adsorbent bed;

(6) the adsorption/desorption being effected so as to create an oscillating concentration front which remains in the adsorbent bed with only a fraction of the adsorbent bed being utilized for most of the adsorption/desorption;

(7) said process operating so as to eliminate the need for any supplemental heat exchanges; and

(8) the need for any external purge gas supply."

These allegedly unique steps in the Skarstrom process upon analysis can be distilled to four:

1. Repetitive adsorption and desorption at short cycle time at substantially the same temperature, thus eliminating the need for heaters.

2. Adsorption being at high pressure, and desorption at low or atmospheric pressure.

3. A portion of the dry product being used in reverse flow as the purge gas.

4. The adsorption and desorption creating an oscillating front within each bed.

The explanation of the results obtained under the Skarstrom method is founded largely on previously known scientific principles.

Use of a heater not necessary. The plaintiff's expert testified that it has long been known that the adsorbing bed becomes heated as a result of adsorbing, and in desorbing it must give up heat. For it to retain sufficient heat for the desorption stage, the adsorbing bed must be desorbed before the heat created by adsorption is lost by dissipation. Thus the cycle time should be sufficiently brief so that no significant amount of the heat created during adsorption is lost by conduction through the walls of the bed before the switch from adsorption to desorption.

Adsorption at high pressure and desorption at low pressure. It is obvious that if a drier is to yield a net amount of product gas, all of the product gas dried during the adsorption cycle cannot be diverted back through the desorption cycle for use as purge gas. Both parties presented evidence that two of the fundamental laws of the behavior of gases account for the ability of the Skarstrom drier to function successfully using only one quarter of its product gas as purge gas. Boyle's Law dictates that at any given temperature the volume of a gas varies inversely with its pressure. Dalton's Law provides that the total pressure of a mixture of gases is the sum of the partial pressures of the individual gases in the mixture, and hence that a change in the partial pressure of one gas in a mixture, with a consequent change in the total pressure of the mixture of gases, has no effect on the partial pressures of the other gases in the mixture. Boyle's Law means that only one quarter of the product gas is needed to flow back through the desiccant as purge gas, if the 60 pounds per square inch pressure of the product gas is reduced to 15 pounds per square inch for its use as purge gas. And Dalton's Law means that the partial pressure of water vapor which the purge gas can carry with it without becoming saturated is the same whether the purge gas is at 15 or at 60 pounds per square inch pressure. Thus the counterflow of purge gas at 15 pounds per square inch of pressure back through the desiccant during the desorption cycle can carry as much moisture with it back into the atmosphere as was imparted to the desiccant by the moist feed gas which flowed through the desiccant at 60 pounds per square inch pressure during the adsorption *211 cycle.^[3] In their explanation through their respective experts of the Skarstrom drier's dependence on Boyle's and Dalton's Laws, neither party suggested that the Skarstrom drier represented a novel application of these fundamental principles or achieved theretofore unpredictable results. It should also be noted that the commercial Trinity air drier on which Dr. Skarstrom experimented was equipped by its manufacturer with valves and conduits for withdrawing a portion of the dry product gas as it was being produced, and the function of these parts was to permit a portion of the dry product gas to be withdrawn and allowed to expand in volume for use as a washing gas during desorption. Plaintiff itself recognized in its brief: "During regeneration the heaters [in the Trinity drier] in conjunction with a counterflow of washing gas would drive out the accumulated moisture and prepare the desiccant bed for a subsequent adsorption operation."

In determining whether the patent in issue is invalid by application of § 103^[4] in that "the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains," several aspects must be separately considered. As outlined by the Supreme Court in Graham v. John Deere Co., 383 U.S. 1, 17, 86 S. Ct. 684, 694, 15 L. Ed. 2d 545 (1966), the formula by which non-obviousness is to be tested is as follows:

"[1] [T]he scope and content of the prior art are to be determined;

[2] differences between the prior art and the claims at issue are to be ascertained; and

[3] the level of ordinary skill in the pertinent art [is to be] resolved.

[4] Against this background, the obviousness or nonobviousness of the subject matter is determined."

See also Monaplastics, Inc. v. Caldor, Inc., 264 F. Supp. 57, 59 (D.Conn.1966), aff'd, 378 F.2d 20 (2d Cir. 1967).

". . . [T]he proper way to apply the [§] 103 obviousness test to a case like this is to first picture the inventor as working in his shop with the prior art references which he is presumed to know hanging on the walls around him." Application of Winslow, 365 F.2d 1017, 1020, 53 CCPA 1574 (1966).^[5] The complete knowledge of prior learning in the art of fractionating gaseous mixtures which an inventor "having ordinary skill in the art" must be regarded as having would include knowledge of Boyle's Law, Dalton's Law, and the thermodynamic characteristics of fractionation. While these laws were cited to explain scientifically how use of the patented process desorbs the desiccant beds which dry the air, the scientific explanation of what takes place cannot be patented. DeForest Radio Co. v. Gen'l Electric Co., 283 U.S. 664, 684-685, *212 51 S. Ct. 563, 75 L. Ed. 1339 (1931); Aetna Steel Products Corp. v. Southwest Products Co., 282 F.2d 323, 334 (9th Cir. 1960), cert. denied, 365 U.S. 845, 81 S. Ct. 804, 5 L. Ed. 2d 810 (1961). Most pertinently, as part of the prior art the inventor had at hand an air drying device which was already rigged with a pair of desiccant beds and a timing device which regulated operation in a repetitive alternate cycle and valves which diverted a portion of the dried air under pressure into a pipe large enough to permit the gas to expand to atmospheric pressure, thus increasing its volume by a factor of at least four, whereupon as "washing gas" or "purge gas" it flowed through a wet desiccant bed functioning as a desorption agent to dry out the bed and make it ready for adsorption again.

In addition to the demonstrative teaching in the art of drying air furnished by the commercial drier made by and purchased from the Trinity Corporation, there was other learning which would have to be attributed to the hypothetical person "having ordinary skill in the art." A substantial amount of prior art was developed by Kahle, an earlier German inventor, and disclosed by him in publications. The most significant of these are German Patent 970,223, German Patent 871,886, and the Chemie Ingenieur Publication of 1953. None of these were cited by the Patent Office against the Skarstrom patent. These deal specifically with methods of fractionating gas. The fact that there was prior art not considered by the Patent Office is sufficient to dissipate the presumption of validity which otherwise would be attributed to a patent. See Monaplastics, Inc. v. Caldor, Inc., supra, 264 F. Supp. at 62-63.

Circumstances not usually present in litigated patent cases pointedly establish which claims the patentee contends are not disclosed in prior art. During the prosecution of Dr. Skarstrom's application, Esso's United States patent department was also prosecuting an application for a German patent in the Skarstrom invention through Esso's German patent agent. Under the German procedure, a patent is tentatively issued and published, so as to permit other parties to oppose its final grant.^[6] During that opposition proceeding the applicant may amend his claims to avoid the opposition that is offered by the other parties. That was done by Esso in those German proceedings, in July 1972, when it submitted the following claim to the German Patent Office:

"[Process for separation of one or more components from gaseous mixtures by selective adsorption and subsequent desorption at substantially the same temperature wherein one of several adsorbing beds are alternatingly charged during an adsorption phase by introduction of the feed gas and withdrawal of the non-adsorbed product gas, and wherein a desorption phase at a lower pressure desorbs the adsorbing bed by feeding in counter-current direction a part of the fractionation product as purge gas,] characterized in that one uses as purge gas a part of the product gas directly recovered from the adsorber and one adjusts the duration of the respective adsorption and desorption phase to a time period of not more than 2-3 minutes each, [for adiabatic use of the sorption heat as well as for formation with the phase alternation of an oscillating loading front, known per se, in a middle section of each bed.]"

In German patent law, as Esso was advised by its German patent agents, Esso was required to acknowledge the state of the prior art in its German claims by first stating at the beginning of any claim those steps known in the prior art and thereafter asserting in a "characterizing" clause what it claimed *213 was novel. Thus everything within the first set of brackets in the quoted portion was admitted by Esso to be relevant prior art. Secondly, if anything mentioned in the "characterizing" clause was old, this must be disclosed as "known per se." Since everything within the second set of brackets was accordingly admitted by Esso to be not new or novel, this matter could not satisfy the conditions of patentability imposed by our law in 35 U.S.C. §§ 101 and 102. See Graham v. John Deere Co., supra, 383 U.S. at 12, 86 S. Ct. 684.

It should be emphasized here that, in considering what Esso submitted to the patent authorities in Germany as its claim, this Court is primarily concerned first with Esso's own admissions of the scope and content of the prior art and second with its analysis of the differences between the prior art and its claims for a patent, not with the decision the German Patent Office made with respect to patentability. For although the procedure in Germany for testing the validity of a patent is different from ours, the factors considered in applying the test for non-obviousness of § 103 are in substance the same. The Supreme Court has directed in Graham v. John Deere Co., supra, 383 U.S. at 17, 86 S. Ct. 684, see p. 211, supra, that the scope and content of the prior art are to be determined and the differences between the prior art and the claims at issue are to be ascertained. What Esso submitted in Germany sheds considerable light on these two aspects of the inquiry.

Having thus carefully restricted its contention that there is no prior relevant art to only those unbracketed elements of the method, which call for (1) the use of a part of the product gas directly recovered from the adsorber as "purge gas" and (2) the adjustment of the adsorption and desorption cycle to a time period of not more than 2-3 minutes each, the question still remains whether there was prior art relevant to these two elements of the claim. While the claim submitted in Germany by Esso can be regarded as an admission against its interest, it was still self-serving. Firstly, insofar as the use of a part of the product gas as "purge" gas is concerned, the Trinity two-bed drier upon which Skarstrom conducted his experiments did just that, although the proportion of product gas diverted as purge gas was relatively small. See pp. 207, 209, 211, supra. Furthermore, the article by Kahle in the Chemie Ingenieur Publication of 1953 (#22) disclosed at page 148, left col., ¶ 2, lines 7-10: "One can also use as a purge gas, a partial flow of the clean gas branched off downstream from the adsorbent."

Secondly, it is not surprising that a cycle time of 2-3 minutes is not specified in prior art references, in view of Skarstrom's own statement that "The particular times utilized depend upon various factors such as the particular adsorbent utilized, the height of the bed, the nature of the key component, and other variables." Skarstrom patent, col. 7, lines 12-15. Nonetheless, the British patent 633,137 does disclose a 3-minute cycle time. The subject matter of all of the prior art referred to was the fractionation of gases, so there can be no doubt about its relevance.

The hypothetical person described by § 103 as having ordinary skill in the art is assumed to possess a detailed knowledge of all there is to know in his own field, for the concept underlying § 103 is that a patent shall not be issued for ways of doing things which are not really new or which do not add to what was known before. One of "ordinary skill" in the art is chargeable with comprehensive knowledge of the art, Continental Can Co. v. Crown Cork & Seal Co., 415 F.2d 601, 603 (3d Cir. 1969), cert. denied, 397 U.S. 914, 90 S. Ct. 916, 25 L. Ed. 2d 94 (1970). Thus it does not matter whether Dr. Skarstrom had acquired this knowledge so long as it was published, cf. Hazeltine Research, *214 Inc. v. Brenner, 382 U.S. 252, 86 S. Ct. 335, 15 L. Ed. 2d 304 (1965), and this is without regard to whether the relevant references were of foreign or domestic origin. Application of Baum, 374 F.2d 1004, 1009, 54 CCPA 1430 (1967). Moreover, the hypothetical "person having ordinary skiill in the art" does not deserve that ancient tag about facts passing from the notebook of the teacher to the notebook of the student without passing through the mind of either, for he must be held accountable not only for specific teachings of references, but also inferences which those skilled in the art may reasonably be expected to draw. Application of Hoeschele, 406 F.2d 1403, 1407, 56 CCPA 1001 (1969).

Taking a kind of a grandstand view of what Dr. Skarstrom did to arrive at the patented method, it appears that, when his routine research to determine the composition of bottled gas was interrupted by the failure of the heaters in the Trinity drier which he was using as a tool, he turned his attention to that drier to see if he could not make it do its job without them. Then by an empirical approach he shortened the cycle time, and solely using part of the product gas as purge gas he achieved desorption without the addition of any heat. And as noted above, these are the specific contributions to the method which Esso claims are new. If these were not directly taught by the prior art, the differences are certainly not outstanding.

In determining whether Dr. Skarstrom's adjustments to the process employed by the Trinity drier merit the grant of a patent, the Court has been instructed on the criteria to be used in testing for non-obviousness and must exercise strict observance in applying them. Graham v. John Deere Co., supra, 383 U.S. at 18-19, 86 S. Ct. 684; Anderson's-Black Rock, Inc. v. Pavement Salvage Co., Inc., 396 U.S. 57, 62, 90 S. Ct. 305, 24 L. Ed. 2d 258 (1969). Since to be patentable, "[a] discovery or devising must add something of significance, though not necessarily very much, to scientific knowledge if it is to take on the quality of invention under a statutory scheme which gives the inventor, for a number of years, a legally protected monopoly in a legal order which generally abhors monopoly," Packwood v. Briggs & Stratton Corp., 195 F.2d 971, 973 (3d Cir.), cert. denied, 344 U.S. 844, 73 S. Ct. 61, 97 L. Ed. 657 (1952), Dr. Skarstrom's fractionation process falls short of the mark. Compare American Potato Dryers v. Peters, 184 F.2d 165, 170 (4th Cir. 1950), cert. denied, 340 U.S. 930, 71 S. Ct. 491, 95 L. Ed. 671 (1951). All of the steps he took led in the directions which the prior art would lead one to expect. There was no sudden break-through which opened up exciting possibilities for the future. Qualitatively, his method produced no result different from that of the Trinity drier. And, since quantitatively the Skarstrom method used ten times as much of the product gas for purging (25% as against 2%), the use of heaters in the Trinity drier was not entirely a negative step.

It may be that Dr. Skarstrom was pleased when he discovered that he could use the Trinity drier without the aid of heaters, but this does not mean that he must be rewarded with a patent, for neither his, nor the court's subjective reaction has any place in the test which must be made. Obviousness is a legal conclusion which the court is required to draw from the facts appearing of record. Application of Lettvin, 339 F.2d 249, 251, 52 CCPA 903 (1964). In deciding the judgmental issue against the background of what is set out above, it is clear that what the patentee did was simply a matter of following the prior art in a logical way, applying those insights which would suggest themselves to one skilled in the art. The prior art pointed to the use of purge gas for desorption, and if no additional heat was to be supplied, the use of more purge gas was certainly not an unobvious step to take. Since it was well known that heat was generated in the bed on adsorption, it was not non-obvious to speed *215 up the cycle in order to desorb before the bed, heated by adsorption, had cooled off.^[7] In Kahle's British Patent 677,150, counterpart of German 970,223, it is specified that "owing to the short adsorption desorption time, practically the whole heat of the adsorption process remains in the adsorber and is consequently usefully employed for the subsequent desorption." What Dr. Skarstrom did showed technical competence, but was far short of scientific innovation. From the standpoint of science his method does not add more knowledge or shape what was known before into greater order or harmony. And, unless no attention is paid to what had already been disclosed by Kahle, his perception of how technologically to apply that learning was not something non-obvious to a person having ordinary skill in the art of fractionating gaseous mixtures. It follows that the Skarstrom method did not rise to the level of patentability. Had the prior art which was before this Court been before the patent examiner who processed the application, it is most unlikely that he would have granted the patent.

As Mr. Justice Douglas noted in A & P Tea Co. v. Supermarket Corp., 340 U.S. 147, 154-155, 71 S. Ct. 127, 131, 95 L. Ed. 162 (1950) (concurring opinion):

"Every patent is the grant of a privilege of exacting tolls from the public. The Framers plainly did not want those monopolies freely granted. The invention, to justify a patent, had to serve the ends of science to push back the frontiers of chemistry, physics, and the like; to make a distinctive contribution to scientific knowledge. That is why through the years the opinions of the Court commonly have taken `inventive genius' as the test. It is not enough that an article is new and useful. The Constitution never sanctioned the patenting of gadgets. Patents serve a higher end the advancement of science. An invention need not be as startling as an atomic bomb to be patentable. But it has to be of such quality and distinction that masters of the scientific field in which it falls will recognize it as an advance."

I conclude that the Skarstrom patent in issue does not meet the non-obvious test for patentability set forth in 35 U.S.C. § 103. It follows that the defendant cannot be held liable for infringement of an invalid patent. And since this is dispositive of the case, there is no reason to reach the other issues raised by the defendant. Accordingly, the defendant is entitled to a judgment dismissing the complaint and declaring that U. S. Letters Patent No. 2,944,627 is invalid, and it is

So ordered.

[1] Analysis of the method as applied to a specific gas, water vapor, imparts a sharper focus to a discussion of the issues. The removal of water vapor from air in effect dries the air. Under normal atmospheric conditions where the temperature is about 70° and the relative humidity is 50%, the surrounding air will contain about 10,000 parts per million of water vapor. Stated in another way, the surrounding air would contain somewhere around 1% by volume water vapor.

[2] As previously indicated, this is the example chosen by the parties to illustrate the removal of one of the key components from a gaseous mixture.

[3] Atmospheric pressure is 14.7 pounds per square inch. Although the evidence did not touch on this point, presumably the purge gas must be under a pressure somewhat higher than this in order to insure that the purge gas does indeed flow out through the desiccant to the lesser pressure of the atmosphere outside.

[4] 35 U.S.C. § 103 provides:

"A patent may not be obtained though the invention is not identically disclosed or described as set forth in section 102 of this title, if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. Patentability shall not be negatived by the manner in which the invention was made."

[5] But see Application of Antle, 444 F.2d 1168, 1171, 58 CCPA 1382 (1971), limiting the applicability of this smile to cases, such as the instant case, where the prior art is in the inventor's field of endeavor.

[6] In this country it is only after litigation that the validity of issued patents can be conclusively determined.

[7] Although Dr. Skarstrom may have been unaware of the fact that whatever changes he made had been previously pointed out by Kahle to those interested in the process of fractionating gases, that cannot be said of Esso's patent counsel who were prosecuting his application for a patent. When a tabulated comparison between what Kahle had done and what Esso proposed to claim in Germany was made by Esso's U. S. patent department, the similarities were made clear, and differences sought by Esso were negated by Esso's German patent in dialectic interaction. See Def.'s Ex. V(5). This comparison revealed an awareness that Kahle's 1953 article disclosed a heaterless drier. Furthermore, in criticism of the tabulated comparison, Esso's German patent agent wrote to Esso:

"I agree to the tabulated comparison of the features of the invention and the Kahle references with two exceptions. Firstly, in my opinion there is no doubt that German patent 970,223 discloses the method of cyclical heat conservation. Moreover, both German patents 871,886 and 970,223 as well as the Kahle article (1953) clearly disclose that adsorption and desorption are effected at the same temperature. Secondly, cyclical heat conservation is a result rather than a teaching of the operating conditions. For both reasons the cyclical heat conservation itself will hardly provide a distinguishing feature."