Air Products, Incorporated, instituted a civil action, in the United States District Court for the District of Maryland, against Boston Metals Company (hereinafter called Boston), alleging infringement by Boston of the Anderson patent No. 2,480,093 relating to the manufacture of oxygen and entitled "method of and apparatus for pumping liquid oxygen." 98 F.Supp. page 720. The opinion of the District Court is reported in 98 F. Supp. 719, 729 and thus concludes: "Summarizing our conclusions, they are as follows: (1) method claims 11, 13 and 19 of the Anderson patent No. 2,480,093 are valid; (2) the method of pumping liquid oxygen employed by the defendant infringes these claims; and (3) the plaintiff has seasonably and adequately purged itself of any violation, and has removed the effects of any violation of the patent or anti-trust laws which it may have committed." Boston has duly appealed.

We append the first six paragraphs of the specifications of the Anderson patent, and claims 11, 13 and 19:

"This invention relates to a method of pumping liquefied gases and to an apparatus adapted to that use.

"An object of the invention is to provide a method and means for withdrawing a liquefied gas from a vessel in which it is stored or is being collected, in such manner as to avoid the possibility of gas locking the pump.

"An object of the invention is to provide a method and means for withdrawing a stream of liquefied gas of any desired constant quantity from a collecting pool in a gas fractionating column.

"An object of the invention is to provide a method and means for withdrawing a stream of gas from a fractionating tower and of reducing the gas to liquid form for delivery under pressure by means adapted to the pumping of liquids.

"An object of the invention is to provide a method and means for pumping liquefied oxygen directly from a pool of commercially pure oxygen in a fractionating tower to cylinders or pipe lines in which gaseous oxygen is transported under high pressure, thereby avoiding the requirement for an oxygen storage tank and a gaseous oxygen compressing system.

"An object of the invention is to provide a method and means for withdrawing oxygen in gaseous form from the pure oxygen vapor space in a fractionating tower, for liquefying the gaseous stream and for delivering the oxygen into pressure cylinders or pipe lines, thereby retaining in the tower any lubricating oil or other combustible substances which may enter the tower with the air stream."

"11. In a method of producing oxygen and conditioning it for delivery to receiving means, in which air after compression and cooling is rectified at a relatively low temperature and reduced pressure thereby producing a cold nitrogen product and a liquid oxygen product having a temperature corresponding to its boiling point at said reduced pressure; the set of steps comprising subjecting fluid from said oxygen product to heat exchange with a colder fluid derived from said rectification and thereby forming a sub-cooled liquid oxygen product; pumping such sub-cooled liquid oxygen product to a desired higher pressure, said sub-cooling reducing the liquid oxygen temperature at least sufficiently to prevent the same from flashing into vapor during such pumping; and converting the liquid oxygen at said higher pressure into a gas by heat exchange with the compressed air to be liquefied."

"13. In a method of producing oxygen and conditioning it for delivery to receiving means, in which air after compression and cooling is rectified at a relatively low temperature and reduced pressure thereby producing a cold nitrogen product and a liquid oxygen product having a temperature corresponding to its boiling point at said reduced pressure; the set of steps comprising subjecting fluid from said oxygen product to heat interchange with a stream of gaseous nitrogen obtained as one of the final products of said rectification and thereby forming a sub-cooled liquid oxygen product; pumping such sub-cooled liquid oxygen product to a desired higher pressure, said sub-cooling reducing the liquid oxygen temperature at least sufficiently to prevent the same from flashing into vapor during said pumping; and converting the liquid oxygen at said higher pressure into a gas by heat exchange with the compressed air to thereby cool the air."

"19. The method of transferring liquefied product of a fractionating operation, in which operation a mixture of component gases having boiling points substantially below atmospheric temperature is compressed and cooled, the compressed and cooled mixture expanded and the effluent of the expansion step subjected to the fractionating operation to produce a liquefied higher boiling point fraction and a gaseous lower boiling point fraction, comprising withdrawing a stream of the liquefied higher boiling point fraction from the fractionating operation, subcooling the stream of liquefied higher boiling point fraction by heat exchange with a relatively colder product from the fractionating operation to reduce the temperature of the higher boiling point fraction to a point below the boiling point of the higher boiling point fraction at the minimum momentary pressure reached in an ensuing pumping step, pumping the sub-cooled higher boiling point fraction in liquid phase to a relatively high pressure and heat exchanging the higher boiling point fraction from the pump with the mixture of gases going to the fractionating operation to cool the mixture and vaporize the higher boiling point fraction."

It seems quite clear that the accused device of Boston infringes the Anderson patent. We, therefore, discuss only the Anderson patent's validity. Since we think the claims in suit of Anderson, Nos. 11, 13 and 19, are invalid for lack of invention, the judgment of the District Court must be reversed.

Of the patents in the prior art, we attribute particular importance to two German patents, Polard No. 495,795, and Messer & Grassman No. 712,480, and to Eichelman No. 1,976,388, Messer No. 2,133,105 and Hansen No. 2,292,375. These patents, in their relation to Anderson, we now discuss.

The Polard patent discloses the essential elements of Anderson, particularly the prevention of vapor-lock in the pumping by sub-cooling with nitrogen, and the pumping of oxygen gas directly into the cylinders, ready for use by the consumer. Polard mentions apparatus for pumping liquid oxygen from a pool in "a container;" Anderson mentions pumping liquid oxygen from a pool in the "column." It is open to question whether Polard is limited to pumping from a particular type of container; also whether Anderson is strictly limited to pumping from the column. In any event, this difference, if it exists, is neither vital nor sufficient to constitute patentable invention.

There are many points of similarity between Eichelman and Anderson, even if (as appellee alleges) Eichelman was a paper patent and would be subject to vapor-lock in the pump; for the problem of vapor-lock had been solved by Polard, Hansen and Messer and Grassman. Unlike Anderson, Eichelman does not sub-cool his liquid oxygen in advance of the pump.

Hansen clearly shows sub-cooling to prevent vapor-lock of the pump. This he does by several methods — one, close to Anderson, by sub-cooling the oxygen by heat exchange with a colder material in advance of the pump.

Unlike Polard and Hansen, but like Anderson, Messer shows the liquid oxygen initially collected, not in a storage container, but in a pool in the column. Messer also shows that the oxygen line runs through a heat exchanger and "is super cooled * * * by the nitrogen flowing from the separating column directly connected with it." Messer and Grassman taught further that suction pumps were preferable to pressure pumps.

Peff, a witness for Boston Metals, testified that he built his apparatus (the accused device here) solely from the teachings of patents prior to Anderson, chiefly Messer and Grassman. The District judge admitted these patents "disclose the first step of Anderson."

The real gist of the decision below, we think, is found in three extracts from the opinion, which we quote:

"It is first necessary to understand specifically of what the Anderson invention consists, that is to say, the steps that are involved in the Anderson method. These steps are three in number as follows: (1) Withdrawing liquid oxygen from the boiling pool in the rectifier column and cooling it on its way to the liquid oxygen pump by heat exchange with the colder gaseous nitrogen withdrawn from the rectifier column; (2) pumping the liquid oxygen from the low pressure of the rectifier column to the high pressure desired in the shipping cylinders; and (3) converting this pumped liquid oxygen to gaseous form in the air heat exchanger by heat interchange with the warm compressed air going to the rectifier column. * * *

"Nor do these prior patents, considered together in any combination, anticipate Anderson because in all of them the oxygen, after its separation from the nitrogen in the fractionating column, enters the heat exchanger as a gas, whereas with Anderson, it enters as a liquid and is converted into gas in the heat exchanger. Also, in the prior art, the gaseous oxygen enters the heat exchanger at low pressure, whereas with Anderson, the liquid oxygen enters the exchanger at high pressure. Anderson pumps the liquid oxygen to the pressure ultimately desired in the cylinders in which it is to be shipped to the consumer; converts the liquid oxygen into gas and delivers it as such, completely dry, without the use of (1) a storage vessel (i. e. the common gas tank or bell water tank); (2) a compressor, or (3) a dryer, all three of which had been used in the prior art for some twenty-five years or more. * * *

"We must not lose sight of the fact that Anderson's invention lies in a method which embraces a combination of three steps which are not found in any prior patent, nor do we find that by piecing together two or more prior patents this threefold combination has been, in fact, disclosed or even suggested by anything in those patents which warrants saying that one skilled in the act, by piecing together what these prior patents teach could, without an inventive act, arrive at Anderson's method and accomplish what he accomplished." [98 F. Supp. 723.]

We think the District Judge gave undue emphasis to the "three steps" of Anderson. It is true that these precise steps, as found in Anderson are not disclosed in any single patent of the prior art. But, as has been indicated, the essential elements of each of the "three steps" of Anderson are disclosed, some in one patent and some in another, in these prior patents. Twomey, plaintiff's expert witness, on whose testimony the District Judge seems to have heavily relied, over-emphasized the minor differences between Anderson and prior patents, and failed to see how clearly these prior patents disclosed practically all that is important in Anderson.

As we have indicated, the District Judge admitted that prior patents "disclose the first step of Anderson." His second step, also the first, were shown in the Polard patent in substantially the same way as they are shown in Anderson. In other words, Polard used the cold nitrogen first to cool the oxygen as it entered the pump and also to cool the pump itself. It is significant that these two steps of Polard performed the same function and achieved the same result in Anderson as they had already done in Polard. It is, of course, essential for a patentee who uses a combination of old steps to show that in the new combination they perform a new function.

The third step upon which the plaintiff relies, that is to say, conducting the condensed oxygen back into the original heat exchanger, was not discovered by Anderson. In addition there is nothing new in the mere fact of transferring the completed product into the small sized cylinders rather than into a large storage tank. This is shown in Eichelman, amongst others.

A great deal was made by the patentee of the fact that the Polard patent shows a large storage tank from which the oxygen was conducted into the pump, and that after going through the pump, then the oxygen went into the individual cylinders. It hardly required invention to eliminate the large storage tank from the process. Anderson accumulated the gas to be condensed in the double rectifying column and from that conducted the oxygen to the pump. If the manufacturer did not wish to store the oxygen in a large container, but preferred instead to conduct the gas immediately into individual cylinders, it did not require invention to make this change.

Clearly, the essential features of the Anderson apparatus were all old; there was nothing new in the fractionating column, the heat exchanger, valves, pipes and cooled pump. Old, too, were the uses and functions of these parts, and solutions to the problems set out in Anderson's specifications. Anderson's only achievement was a non-patentable aggregation of parts of prior patents, with some clever, but not important, changes and adaptations which would readily suggest themselves to those skilled in the art. Those changes and adaptations do not amount to patentable invention.

Quite germane here is the opinion of Mr. Justice Jackson in the very important case of Great Atlantic and Pacific Tea Company v. Supermarket Equipment Corporation, 340 U.S. 147, 71 S. Ct. 127, 95 L. Ed. 162. See, also, the following cases, all decided by our Court. Ingersoll-Rand Co. v. Black & Decker Manufacturing Co., 4 Cir., 192 F.2d 270, and opinion in the District Court, 94 F. Supp. 938; Novocol Chemical Manufacturing Co. v. Powers & Anderson Dental Co., 4 Cir., 128 F.2d 904, 907-908; McElrath v. Industrial Rayon Corporation, 4 Cir., 123 F.2d 627, 630-631; Doughnut Machine Corporation v. Joe-Lowe Corporation, 4 Cir., 67 F.2d 135, 137; Victor Cooler Door Co. v. Jamison Cold Storage Door Co., 4 Cir., 44 F.2d 288. And there are many controlling differences between the instant case and the case of Eibel Process Co. v. Minnesota & Ontario Paper Co., 261 U.S. 45, 43 S. Ct. 322, 67 L. Ed. 523, upon which Air Products so strongly relies.

In arriving at our conclusion, we have not overlooked the presumption of validity that attaches upon the issuance of a patent by the Patent Office. Nor have we been unmindful of the fact that in very close cases commercial success of a patent may turn the scale in favor of validity. We have, too, given weight to the findings of the District Judge, who saw and heard the witnesses, and his decision that the Anderson patent is valid.

To sum up, we think from the standpoint of the instant case most favorable to the patent in suit, it appears that Anderson was the first person to produce high pressure oxygen gas and deliver it to receptacles in one process without the use of an intermediate receptacle between the fractionating column and the ultimate containers; that this was accomplished by a method never in the precise form employed by Anderson theretofore used, viz., pumping liquid oxygen which had been withdrawn from the fractionating column to a high pressure and avoiding vapor lock in the pump by sub-cooling the oxygen and the pump by the use of nitrogen gas from the top of the fractionating column; and that this process entered into immediate use and achieved commercial success. Yet, though embodied in no single previous patent, the essential elements of Anderson were all found in the prior art. What Anderson did was merely to aggregate these old elements, with changes involving the mere skill of the art, which did not attain the dignity of patentable invention.

The judgment of the District Court is reversed and the case is remanded to that court with instructions to enter judgment invalidating the claims in suit of the Anderson patent and dismissing the plaintiff's civil action.

Reversed.