Appeals from the District Court of the United States for the District of Delaware.
Before BUFFINGTON, DAVIS, and THOMPSON, Circuit Judges.
These suits were brought against the Root Refining Company by the Universal Oil Products Company, hereinafter called plaintiff, for the infringement of all the nine claims, except the sixth, of United States letters patent No. 1,392,629 issued to Carbon P. Dubbs, October 4, 1921, and by him assigned to the plaintiff, and all five claims of United States letters spent No. 1,537,593 issued to Gustav Egloff, May 12, 1925. Each patent is for a process for the manufacture of gasoline from petroleum. The Root Refining Company defended on the ground that both patents were invalid; the Dubbs patent for lack of disclosure and the Egloff patent for lack of utility, and both for lack of novelty over the prior art. All the claims in issue were held to be valid and infringed by the learned District Judge. The two suits were consolidated for trial and will be disposed of in a single opinion here as they are below.
The Root Refining Company, hereinafter called defendant, is a Delaware corporation, and the plaintiff is a corporation of South Dakota.
Both patents relate to improvements in the process of cracking petroleum for making gasoline. Petroleum in its natural state, as it comes from the earth, is made up of many combinations of hydrogen and carbon. The natural constituents of these combinations are classified according to their specific gravity or boiling points. Cracking involves the decomponsition of these complex hydrocarbon molecules of petroleum by the application of heat and pressure. It breaks up the heavier constituent elements and converts them into lighter oils such as incondensable gases, gasoline, kerosene, gas oil, and fuel oil. As the heavy hydrocarbon componds composing the oil to be refined are broken down, new ones are formed. Some of these contain a great number of valuable hydrogen elements while others contain more carbon compounds which have little or no value.
Cracking, as distinguished from distillation, which involves the mere physical separation of the elements of the oil, involves a chemical reaction.This is produced by the application of intense heat, 700 deg. to 850 deg. F. and upward, for a sufficient length of time to break up the molecules. The art of cracking is of great advantage for it is said to save more than 500,000,000 barrels of oil yearly.
The historical development of cracking oil to the state of the art where Trumble left it was admirably set forth by Judge Woolley in the case of Skelly Oil Co. v. Universal Oil Products Co. (C.C.A.) 31 F.2d 427, 429. What Trumble did was to take a typical cycle system, well known in the art, and "disclose a practice and suggest a means continuously to draw off a small portion of the revolving residuum (loaded with the objectionable matter) in each of its cycles through the closed system and replenish the bulk with an equal amount of new oil stock and thereby maintain the bulk constant in volume." This was held to be "something abruptly different from what existed before," and, utility being evident, the patent was held to be inventive and valid.
There are two real questions involved here: (1) The validity of the claims; and (2) their infringement, if valid.
Validity depends upon whether or not Dubbs and Egloff made any inventive advance in, or contribution to, the art over Trumble. If they did, what was it?
The refining industry according to the plaintiff recognizes three types of cracking processes: (1) The liquid phase, all the oil undergoing treatment being in a liquid condition in the zone where heating and carcking take place; (2) the liquid vapor or mixed phase; and (3) the vapor phase, all the oil being converted into vapor before it reaches the heating and cracking zone.
The Dubbs patent practices the mixed phase process, in which all the heating is done in the heating or cracking coils.(Generation occurs in these coils, "but without substantial evaporation.") In the process of refining oil, carbon rapidly accumulates in large quantities upon the heating tubes of the still and causes serious difficulties and great expense. In the development of the art, carbon which thus formed on the inner walls of the heating tubes increased their temperature to such a degree that it became necessary to shut down the plant frequently in order to remove the carbon and to prevent the tubes from bursting. A still under the old Burton-Clark system, the best before Trumble, could not be operated for more than 36 hours before it had to be shut down. It required about the same length of time to clean the tubes, because many of them are in the furnace and are quite long, before another run could be started. After the still had been closed down and cleaned, a considerable length of time was required to heat the tubes again to a temperature sufficient for cracking.
The Trumble patent generally increased the length of the run by removing part of the residue from which the vapor had been separated so that that portion of the residue was never again returned to the cracking coil or tubes. This increased the run from 36 hours to 17 days.
In Dubbs' process the oil to be refined enters the still through a pump J in Fig. 2 and is forced into the "cracking tubes B" where it is heated to the proper temperature, the maximum which it ever attains in the system. It is then discharged into "vaporizing tubes C" which are only partially filled with oil, and as the oil is passing through these tubes the cracking reaction is completed and the vapors formed therein as well as the vapors generated in the B tubes separate themselves from the liquid oil and pass up through tubes D and E to the partial condenser, called a "header" G, thence to the aerial condenser G1, and through the pipe G2 to the final or water condenser G3, which it is discharged through the pressure-regulating valve H. The condensate from the partial condenser G and G1, large in amount and uncontaminated by the residue left in the vaporizing tubes, while still at a heated temperature, is discharged through pipes F F1 into the fresh feed line and this mixture composed of reflux and fresh feed under pressure of the high-pressure pump J is forced into the entrance of the cracking tubes B. The distillate from the final condenser G3 is the desired product of the process. All the residue left in the tubes C from which the vapor has been separated is drawn off through the line K and the pressure regulating value H1, and is never again returned to the system.
Two things distinguish Dubbs from Trumble: (1) Drubbs returns a large amount of reflux uncontaminated, from the partial condensers G and G1 to the entrance end of the cracking tubes, admixed with a charge of fresh oil. (2) He removes all of the residuum left in the vaporizing tubes so that none of it ever again enters into the system. He thus secures what the patent calls a continuous "clean circulation," free from carbon-producing residuum. Because Trumble removes only some of the residuum he does not secure a "clean circulation." This is admitted by counsel for appellant as a distinguishing feature between Trumble and Dubbs. He says that "a comparison between the Dubbs and Trumble operations shows that whereas Trumble withdraws a portion of the residuum, Dubbs withdraws all of it." Trumble thus circulates the residuum through the heating tubes over and over again, but Dubbs prevented the return to the cracking coils of any of the carbon which had been freed and also any of the "asphaltines" which are present in the oil which has once been subjected to cracking temperature, but not vaporized, and these, when again subjected to cracking temperature, rapidly turn into carbon and settle on the cracking tubes. "Trumble returned dirty and Dubbs clean reflux for cracking. There is the difference," said Judge Nields in his comprehensive opinion ([D.C.] 6 F. Supp. 763, 769).
This forward step by Dubbs, though simple as we look back upon it, produced astounding results.
The maximum length of run under the Trumble process, using gas oil as a charging stock, was 17 days.The average run under the Dubbs process using fuel oil, a poorer grade of oil, was 25 to 40 days. The gasoline yield using the Trumble process was 28 per cent. to 30 per cent.; under the Dubbs process the yield was 47 per cent. to 52 per cent.
These advantages were so great and unmistakable that the whole industry immediately recognized them. The Royal Dutch Shell interests which owned the Trumble patent immediately took out a license under Dubbs and paid in royalties over $12,000,000, and later, in 1930, with the Standard Oil Company of California, which spent $5,000,000 in experimental work trying to solve the problem, purchasing all of the stock of the plaintiff company, which owns the Dubbs patent for $25,000,000. The licensed users of the process expended over $50,000,000 in apparatus for practicing it and paid appellees in royalties in a period of seven years a sum in excess of $33,500,000. It is improbable that shrewd and experienced businessmen, doubtless guided by able and competent counsel, would have paid such a price for an invalid patent.
Trumble was a distinct advance over the Burton-Clark process. We held that the Skelly Company infringed the Trumble process because it returned a dirty reflux and drew off only a part of the residuum which had once gone through the cracking tubes without being vaporized, but it did not infringe the Dubbs process because it did not return a clean reflux and did not draw off all the residuum. The Skelly Company could not have been sued at that time under the Dubbs patent, for it was not infringed. The Skelly Company followed the teaching of the Trumble patent and not that of Dubbs. Consequently, that suit was confined to the Trumble patent. The advance of Trumble over Burton-Clark and the prior art was pointed out and emphasized in the Skelly Oil suit. It was also brought ...