NIELDS, District Judge.
This suit charges defendants with infringement of two process patents for the production of gasoline by cracking cheap petroleum oil. Patent No. 1,392,629 was granted in 1921 to Carbon Petroleum Dubbs and thereupon assigned to the plaintiff. The other patent, No. 1,537,593, was granted in 1925 to the plaintiff upon an application filed in 1920 by Gustav Egloff. All claims of both patents are in suit except claim 6 of the Dubbs patent.
The defenses are noninfringement and invalidity. Stated more specifically, the defense of invalidity is: The Dubbs patent is invalid for lack of disclosure, the Egloff patent is invalid for lack of utility, and both patents are invalid for lack of patentable novelty over the prior art.
For a number of years plaintiff has been engaged in promoting the use by refiners, in this country and abroad, of the "Dubbs process" for cracking heavy petroleum oils for the production of gasoline. It owns a number of patents, including the two patents in suit, although many of them are not practiced in the Dubbs process. That process is bottomed upon the Dubbs patent in suit. In the decade before 1931 plaintiff received in royalties under licenses of the Dubbs process $33,500,000. Root Refining Company, hereinafter referred to as defendant, is engaged in the business of refining petroleum at Eldorado, Ark. The other defendant was never served with process.
In the early days of refining crude petroleum, gasoline was obtained by distillation. The operation consisted in gradually increasing the temperature of the oil and driving off a part in the form of vapor. The vapors were then condensed and a separation effected by directing the condensate of one volatility into one vessel, and a portion of a lower volatility into another vessel and so on. This operation was essentially a nonchemical one.
The use of combustion engines in motor vehicles created a great demand for gasoline. That demand would have exhausted the available petroleum deposits within the span of a life unless a large additional supply of gasoline could be obtained. Chemists for several generations had studied the production of gasoline by cracking. "Cracking" is the decomposition of petroleum by heat and pressure, with the consequent breaking up of the molecules and the production of both lighter and heavier hydrocarbons. Gasoline is a lighter hydrocarbon produced by this chemical action. Many patents had issued for processes and apparatus to crack petroleum. The first commercially successful process of cracking to produce gasoline was invented by Dr. Burton and used by the Standard Oil Company of Indiana in 1913. Shortly after the advent of the Burton process, it was improved by the invention of Clark.
At the time plaintiff entered the field with the Dubbs process, the most effective process
The Burton-Clark process is a liquidvapor phase process. There is always a very substantial quantity of liquid in the heating zone. In this process, as well as in earlier processes, the accumulation of carbon upon the heating tubes or chambers of the still was rapid and required the shutting down of the still after it had been "on stream" for a very short period. The carbon formed on the inner surfaces of the heating tubes and the heat became localized in the tubes and increased the temperature thereof to a point where their tensile strength was insufficient to withstand the high pressure. The appearance of red spots was a danger signal, and the still was shut down, the contents discharged and the parts thoroughly cleaned before an attempt was made to bring the still again "on stream." Moreover, these stills were confined to the cracking of light oils. The cracking of crude oil would have been out of the question. The normal operating cycle of the Burton-Clark still was 72 hours, of which 36 hours was the "on stream" period and the balance of the cycle was used in cleaning the still and bringing it "on stream." The monthly output was about 4,000 barrels, with a gasoline yield of from 30 to 33 per cent.
In 1919 the Dubbs process embodying the invention of the Dubbs patent in suit was given an experimental demonstration at Independence, Kan., by the plaintiff. Dubbs was in charge of the work. In February of that year the operation was witnessed by Daniel Pyzel in charge of the Royal Dutch Shell group of refineries. The following month Dubbs applied for his patent. In December of that year Pyzel obtained a license for his companies from plaintiff. Pyzel testified that the so-called "clean circulation" of the Dubbs patent was the primary object in obtaining the license. Prior to this demonstration the Dutch Shell group operated under the Trumble patents. After seeing the Dubbs operation, Pyzel testified his companies had been proceeding on the "wrong track," because in the Trumble process a very substantial part of the dirty residuum was returned to the heating coil.
In the summer of 1919 the operation of the Dubbs still at Independence was witnessed by 21 refinery representatives. These observers were familiar with the Burton and Burton-Clark type of still. The "on stream" period of the Dubbs still in this demonstration was 800 per cent. longer than it had been possible to obtain in any other gasoline cracking process. The explanation of the length of the run was the employment of the "clean circulation" feature of the Dubbs process.
The Standard Oil Company of California also recognized the superiority of this process. It had used the Burton and Burton-Clark types of still for cracking the heavy California oils, but found them unsatisfactory. That company maintained a development department of 200 men at a yearly cost of a million dollars in a vain effort to develop a successful cracking process. It sent one of its engineers to a plant of the Shell Company to observe the operation of a Dubbs unit. Upon receiving the report of the engineer, Hanna, vice president of the California Company, wrote to his superior officers, inter alia:
"In studying over these reports, it would seem to me that we have picked up the following:
"First: Cleanliness is next to godliness. * * *
"Sixth: That if the maximum degree of cleanliness is maintained, higher furnace efficiency can be attained."
In 1925, the California Company obtained a license from the plaintiff and installed 15 Dubbs plants in California. For the oil processed to June 30, 1930, it paid to the plaintiff royalties in excess of $2,300,000.
The apparatus disclosed in the Dubbs patent comprises five important parts: (1) Tubes called "cracking tubes" in a furnace through which a mixture of fresh charging stock and reflux condensate is continuously flowing as a stream with enough liquid oil to wash the tubes and prevent the deposit of carbon. Here the oil is raised to the maximum temperature in the process and is
The process disclosed in the Dubbs patent is characterized by the following features: (1) The passage of oil in a stream through cracking tubes where its temperature is raised to the maximum and cracking occurs without substantial separation of the generated vapors from the liquid. The vapors generated in the cracking tubes are homogeneously disbursed in the liquid oil and pass with the liquid as a stream through the tubes whereby carbon deposits are prevented. (2) The discharge of the heated stream from the cracking tubes into the separating chamber where cracking is continued and the vapors free themselves from the liquid and pass up and out of the top. (3) Subjecting the vapors from the separating chamber to condensation in a dephlegmator. The lighter vapors pass on to another condenser and are finally drawn off as gasoline. The heavier fractions are returned as a clean condensate to the cracking tubes. (4) Continuously withdrawing this condensate, called "reflux," and mixing it with a fresh charge of oil. (5) Withdrawing permanently from the system the residue accumulating in the separating chamber. (6) Imposing the pressure of self-generated vapors on the oil undergoing treatment.
A comparison of the Dubbs process with the Burton-Clark process shows: (1) In the Burton-Clark process widely used before Dubbs, the residue accumulating in the separating chamber was recirculated through the cracking tubes. In Dubbs this recirculation does not occur. (2) In the Burton-Clark process the reflux condensate was returned to the cracking tubes through the separating chamber and therefore became contaminated with the residue. (3) The Burton-Clark process is a semibatch process, whereas the Dubbs process is a continuous one. With these differences there are striking consequences. Burton-Clark required a special cracking stock, whereas Dubbs handles any petroleum oil. Burton-Clark had an "on stream" period of approximately 36 hours, whereas Dubbs has an "on stream" period many hundred per cent. longer. The gasoline yield of Burton-Clark was from 30 to 33 per cent., whereas the Dubbs yield is approximately 50 per cent.
The Egloff process was designed to further mitigate the carbon evil in connection with the cracking of heavy oils. This process utilizes the salient features of the Dubbs patent. It provides a preliminary screening action for removing from the system the free carbon and heavy polymers which rapidly snowball to coke. The apparatus includes a separating chamber with a residue draw-off followed by a partial condenser, final condenser, and receiver. With these features two heating coils are associated in separate furnace chambers so that different heat and pressure conditions may be obtained. The fresh feed is pumped through the low temperature coil and its temperature is raised to effect a very mild cracking. The oil discharges into the vaporizing chamber and the vapors pass to the dephlegmator. The condensate in the dephlegmator is pumped through the high temperature coil and mixed with the reflux condensate from the other coil. There is an intermingling of two oil streams on their way to the vaporizing chamber. In this process there are two charging stocks: one, the fresh oil which is pumped into the low temperature coil, and the other, a charge formed by the partial condensing of the vapors from the first charge.
The charge of infringement involves a consideration of the claims of the Dubbs patent of which claim 1 may be taken as typical. This claim reads as follows: "1. A process of converting relatively heavy into lighter hydrocarbons, consisting in passing a stream of oil through a zone where the oil is heated to a cracking temperature but without substantial vaporization, passed thence to and through a zone where vaporization takes place, removing the vapors, subjecting them to reflux condensation, returning reflux condensate for further cracking, removing the residue, after said vaporization, from the system without permitting intermingling of any substantial portion thereof with the oil in the heating zone and maintaining a pressure upon the material during distillation." The language of the claim is plain and simple. There can be no controversy about the subject-matter of the grant of the patent and of the limits thereof with the possible exception
The whole defense of noninfringement hangs upon the construction and interpretation of the words "without substantial vaporization" in the cracking tubes. Defendant states this "is a matter of vital and, so far as infringement goes, controlling importance." Defendant holds that "vaporization" in the claims of the patent is synonymous with cracking. It stakes its whole defense of noninfringement upon this position. It contends the Dubbs patent excludes substantial cracking in the cracking tubes, whereas defendant's process involves substantial cracking in the cracking tubes.
"Vaporization," in the phrase "without substantial vaporization" in claim 1 of the Dubbs patent and in the equivalent phrases of the other claims, does not mean cracking. This conclusion follows from a consideration of the patent and of the pertinent evidence. In the specification of his patent, Dubbs says: "B are the cracking tubes. C are the vaporizing tubes." From these terms one would naturally conclude that Dubbs designed a process with substantial cracking in the cracking tubes. The primary function of the C tubes is to separate the vapor from the liquid. One would also naturally conclude that Dubbs had this separation primarily in mind when he called C tubes "vaporizing tubes." A patent, like any grant, should be interpreted as a whole and, as near as may be, a consistent whole. On page 1 of the specification, lines 75 to 83, inclusive, Dubbs states that the reflux and fresh charge are discharged "through tubes B and during the time they are passing through said tubes, they are subjected to sufficient heat to cause the desired amount of cracking. Said oil is then passed into the tubes C which are only partially filled with the oil and as the oil passes through these tubes, there is a liberation of vapors from same and which vapors pass up," etc. Moreover in the illustrative run given in the specification of the patent on page 2, lines 38 to 43, inclusive, Dubbs states: "Thence to the 4 inch coils B which are subjected to a temperature of 750° to 860° F. The heated oil then passes to the 10 inch coils which are maintained about half full of oil and wherein vaporization takes place." This range of heat in the cracking tubes is cracking temperature and at the stated pressure would inevitably result in cracking. Again, in claim 5, the second step of the Dubbs process is defined as "affording a vaporization space above the stream during the second stage of the travel thereof." Vaporization here cannot mean cracking as a cracking space above the stream would be meaningless. A "vaporization space" means a space for liberation of vapor. The words in the claim immediately preceding the words quoted are "while preventing substantial vaporization." "Vaporization" used in successive lines undoubtedly connotes the same meaning.
Dr. Lewis testified in detail concerning the action in the cracking tubes B in the practice of the process of the Dubbs patent. He is professor of chemical engineering in the Massachusetts Institute of Technology and a consulting engineer who, for the last twelve years, has co-operated with the technical staffs of various refineries in studying their problems. Dr. Lewis testified:
"Q. 113. As to the cracking which takes place, where does that occur in this set-up? A. It occurs both in the cracking tubes B, where the heat is supplied, and in the vaporizing tubes C, where no heat is supplied, but cracking continues because of the high temperature obtaining in that zone.
"Q. 114. Will you set forth in more detail the action which occurs in the cracking tubes B in the practice of this process of the Dubbs patent in suit? A. Remembering that the cracking tubes B are the first part of the apparatus which I described, as disclosed in the patent and drawings for the carrying on the process, the process provides that these tubes shall be fed with a mixed liquid feed, consisting in part of fresh feed to the process, and in part of reflux condensate produced in the dephlegmator under pressure and returned to the cracking tubes.
"In the second place, it provides that this reflux shall be substantially free from carbon. In other words, it provides that the feed to these cracking tubes shall be a clean stock.
"In the third place, it provides that throughout these tubes an adequate amount of liquid shall be maintained, an amount adequate to prevent any substantial carbon formation or deposition in these tubes.
"In the fourth place, it provides that there shall be no liberation of vapors in these tubes; that any vapors generated in these tubes shall travel through these tubes, commingled with the liquid in them in the form of a froth or a spray or intimate mixture without any separation or segregation or liberation
"Q. 145. Doctor, is it proper to say that there is no substantial vaporization in the tubes B of the Dubbs patent in suit in practicing the process of that patent? A. Yes, in my opinion.
"Q. 146. Would there in those tubes be vapor generation? A. Yes, a very considerable amount.
"Q. 147. Would there in those tubes be both vapor generation and vapor liberation? A. No, there would be vapor generation only."
The phrase "without substantial vaporization" was manifestly intended to exclude the familiar vapor phase cracking. In this phase, vapors travel through the cracking tube at a speed appropriate to vapors and independent of any liquid. This phase would be baneful rather than beneficial. The Dubbs process contemplates a stream of liquid filled with generated vapors washing through B tubes, thereby preventing the deposit of carbon. Surely if Dubbs or his solicitor intended vaporization to mean cracking, why did they not say without substantial cracking in the cracking tubes. The word "cracking" appears perhaps 100 times in the patent, and if they meant "cracking" they would have said "cracking."
Claim 5 of the Egloff patent is typical and reads as follows: "5. A process of oil conversion, consisting in maintaining a body of heated hydrocarbons in an enlarged zone where substantial vaporization occurs, in subjecting the vapors to reflux condensation to condense the heavier vapors, in passing the reflux condensate in an advancing stream through a heating zone where it is subjected to cracking conditions of temperature and pressure, in delivering heated condensate to said enlarged zone, and simultaneously heating an independent stream of charging oil for delivery to said enlarged zone to a temperature sufficient to cause a substantial vaporization thereof in said zone, in introducing said heated charging oil to the enlarged zone, and in withdrawing unvaporized oil from said enlarged zone without permitting the same to again enter either of said oil streams."
The claims of the Egloff patent are concerned with two heating coils and a separating chamber for both. The initial fresh charge is heated at an appropriate temperature and pressure. There is then a separation of the vapors from the liquid oil. The vapors are subjected to partial condensation and the condensate is used as a charge for the other heating coil operating under higher temperature and pressure. Cracking in the liquid-vapor phase is accomplished. Vapor from the second coil is separated in the separating chamber. There is partial and final condensation of the separated vapors. The residue in the separating chamber from the oil from the high temperature coil and from the low temperature coil is discharged from the system. It is apparent that the salient features of the Dubbs process are employed by Egloff.
Defendant is charged in this suit with using at its plant at Eldorado, Ark., a process for cracking petroleum oil to produce gasoline in a cracking unit furnished by the Winkler-Koch Engineering Company. The unit was installed in 1928 and began operations in 1929. Koch of the Winkler-Koch Engineering Company testified on behalf of defendant. Winkler did not testify. He had been employed by plaintiff from May, 1921, until April, 1925, in a position enabling him to become thoroughly familiar with the Dubbs process. In 1927 Perdew also of the Engineering Company wrote to the president of plaintiff that "all three of us" are "post graduates of Universal." Afterwards a second Dubbs unit was installed by defendant and put into operation. Defendant entered into a license with plaintiff for the use of the Dubbs process providing for the payment of royalties at the rate of 15 cents per barrel of oil processed. The royalties paid by defendant to plaintiff until November, 1930, totaled approximately $350,000.
October 5, 1926, defendant wrote plaintiff, inter alia: "During the last ten days of September our thru-put was at the rate of 1200 barrels daily or 18,000 per unit per month. Our yield of pressure distillate in accordance with figures submitted me was 67% of the charging stock and while all of this pressure distillate has not been re-run, our average has been 78% gasoline from pressure distillate, which indicates a total percentage of gasoline based on charging stock of in excess of 52%. This figure is after all treating, re-run and evaporation losses have been considered."
The Winkler-Koch apparatus used by defendant comprised a tall vaporization chamber, 28'×8½'; with a residue draw-off
In defendant's process crude petroleum is forced through the coils of the low temperature furnace. It is then discharged into a vaporizing chamber. Here the generated vapors separate themselves from the liquid and pass to the bubble tower, where the vapors not condensed pass on to a final condenser. The condensate is pumped from the bubble tower to the second heating chamber where it is cracked. The material from the first and second units then pass into the vaporizing chamber. In the vaporizer there is a body of oil predominantly in liquid phase in state of ebullition approximately 10' deep. The vapors free themselves and pass upwardly through the space of approximately 18' into the bubble tower. There is a constant draw-off of the residue from the bottom of the vaporizing chamber which never returns to the system. The reflux condensate is returned to the high temperature high pressure unit uncontaminated by residue. In the traveling of the oil through the soaking section considerable generation of vapor occurs but there is no liberation of the vapor from the oil until it reaches the vaporizing chamber.
After fully discussing defendant's procedure Dr. Lewis testified: "It seems to me that the process of defendant is certainly a substantial equivalent of the Dubbs process. The differences that do exist are modifications that do not affect the principle and the process as Dubbs disclosed them, in my opinion."
Also Dr. Lewis testified: "The process as practiced by defendant, certainly, it seems to me in the light of all this information with regard to the prior art, is in every essential respect identical with the process as described by Egloff."
There is no evidence in this case of any commercial prior art except the Burton process and a few variants thereof and the Burton-Clark process already considered. The prior art relied upon by the defendant as being in anticipation of or in limitation of the patents in suit are the patents of Hall, Green-street, Alexander, Trumble, and Pielsticker, of which the first three are paper patents. Cracking processes are divided into (1) vapor phase, (2) liquid phase, and (3) vapor-liquid phase. It may be stated broadly that the vapor phase process was a failure because of the deposit of carbon in the cracking zone.
The Hall patent describes a vapor phase process. The oil is cracked in vapor form and discharged directly into a condenser without any vaporizing chamber. The reflux does not return directly to the cracking tubes. It returns indirectly in contaminated form.
The Greenstreet patent also discloses a vapor phase cracking process. The material subjected to a cracking temperature in one of the cracking coils includes the combined residue of a series of cracking operations to which the vapors are subjected. The cracking digesters load the reflux with carbon and prevent clean circulation.
The Alexander patent discloses another vapor phase process. It failed although an effort was made to make it work under favorable conditions. Considerable carbon produced in the cracking operation is carried back with the reflux to the cracking tubes.
The Pielsticker patent belongs to the last century. Although it had always been available, no one used it. There is uncertainty as to what it discloses. For obtaining light oils by cracking, Pielsticker discloses a coil without a vaporizing chamber or reflux condenser. He also discloses a heating coil and shell with a plurality of baffles and a dome. Defendant asserts that cracking occurs in the first coil, vapor separation in the shell, the dome is a reflux condenser, and the reflux may be returned to the coil for recracking. This part of the system, however, was intended for distillation. The dome is an entrainment knockout common in distillation. If the shell was intended for a vapor separator it should be heated to a temperature like the coil and have substantial pressure. No one used it because it was useless.
The Trumble patent discloses the return of a large part of the residue of the vapor separating step to the heating zone and the return of the reflux through the dephlegmator with the resulting contamination. Trumble returned dirty and Dubbs clean reflux for cracking. There is the difference.
These patents do not seriously challenge the validity of the Dubbs patent. A consideration
What are the applicable rules of law?
The Dubbs patent, having substantially advanced the industry to which it relates, is entitled to be construed liberally. In Eibel Process Co. v. Minnesota & Ontario Paper Co., 261 U.S. 45, 63, 43 S.Ct. 322, 328, 67 L. Ed. 523, the invention consisted only in a change in an old paper machine. When operated at a speed of 500 feet per minute the machine gave a defective product. By slightly changing the pitch of one of the parts of the machine, Eibel was enabled to produce a satisfactory product with the wire traveling at a speed of approximately 600 feet per minute. "In the case before us, for the reasons we have already reviewed, we think," said Chief Justice Taft, "that Eibel made a very useful discovery, which has substantially advanced the art. His was not a pioneer patent, creating a new art; but a patent which is only an improvement on an old machine may be very meritorious, and entitled to liberal treatment." The above language was adopted by this court in Westco-Chippewa Pump Co. v. Delaware Electric & S. Co., 57 F.2d 550.
In process patents the claims are not to be limited by the apparatus which may be disclosed for carrying out the process. In Moore Filter Co. v. Tonopah-Belmont Development Co., 201 F. 532, 541, the Court of Appeals for this circuit said: "In considering the question of the infringement of a process patent, it must be borne in mind that the monopoly secured by the claims is, generally speaking, a monopoly of the process, and the test of infringement is whether such process is utilized by the infringer. As the apparatus shown in a process patent is only to show that the process may be practically applied, it follows that such illustrative apparatus does not limit the process patentee to that type of machine alone. If that were the case, a process patent would be of little value. * * * It will therefore be evident that the test of process infringement is not the similarity of apparatus, but rather whether the apparatus, no matter what its form, utilizes the process." In this suit there are differences between defendant's apparatus and the apparatus disclosed in the patents but that is entirely immaterial since the apparatus of defendant carries out the principle of the two patents in suit.
In considering the claims of a patent, that construction should be adopted which will utilize and sustain rather than one which will destroy it. To adopt defendant's contention as to the phrase "without substantial vaporization" would result in a finding that the claim is invalid. Vaporization can connote the mere generation of vapors in a body of liquid. The adoption of this construction would result in a finding that the claims in suit are invalid because they involve a departure from the specification of the patent. On the other hand, vaporization may connote the generation of vapors in a body of liquid accompanied by the liberation or separation of the vapors from the liquid or may connote the mere liberation or separation of the vapors from the body of the liquid. The adoption of either of the latter constructions will not result in destruction of the claims. In Consolidated Rubber Tire Co. v. Firestone T. & R. Co. (C. C. A.) 151 F. 237, 238, Judge Coxe said: "It should be and is the desire of the court in approaching the consideration of a patent for a structure which has thus won a position of unchallenged supremacy in the commercial world, to endeavor to sustain rather than defeat the claims."
Patentable novelty cannot be negatived by showing that all features of the combination may be found by selecting features from a number of patents in the prior art. In McMichael & Wildman Mfg. Co. v. Ruth, 128 F. 706, 708, the Court of Appeals for this circuit said: "Attentive examination of the testimony and exhibits has fully satisfied us that although it is perhaps possible for an expert, having the patent in suit before him, to build up the structure covered by these claims, by selecting and deftly adapting appliances theretofore known, `yet it would still be true that neither the same combination in its entirety nor the same mode of operation' had previously been described or in any manner exemplified. Parks v. Booth, 102 U.S. 96, 26 L. Ed. 54; Bates v. Coe, 98 U.S. 31, 25 L. Ed. 68."
The prior art relied upon as anticipating the patents in suit or as negativing the novelty
In Boyer v. Keller Tool Co., 127 F. 130, 138, the Circuit Court of Appeals of this circuit said: "Convinced, as we are, that the plaintiff has supplied features that have brought success, where others who had preceded him failed, we are not inclined to scan narrowly the means by which it has been obtained. The mechanical elements combined are no doubt old, and so, to a certain extent, may be the result accomplished. But nowhere do we find the same combination employed to produce it, and the efficiency attained is so much in advance of that which had gone before as of itself to suggest, if it does not prove, the exercise of inventive skill."
This opinion contains a statement of the essential facts and of the law applicable thereto in conformity with Equity Rule 70½ (28 USCA § 723).
I conclude that the patents in suit are valid, that the charge of infringement has been sustained, and that the plaintiff is entitled to a decree, with injunction and accounting.