Appellants have appealed from a decision of the Board of Appeals which affirmed the examiner's rejection of claims 1 through 7, inclusive, of appellants' application for patent1 for "Stainless Steel and Article." No claims were allowed.

Claims 2 and 6 are illustrative and are as follows:

"2. Stainless steel possessing precipitation-hardening and good welding characteristics consisting essentially of chromium 16.00% to 18.00%, nickel 6.50% to 7.75%, aluminum .75% to 1.50%, carbon .09% maximum, at least one of the group consisting of titanium about .05% to .20%, zirconium about .10% to .40% and uranium about .30% to 1.00%, and the remainder substantially all iron.

"6. A welded article comprising precipitation-hardened stainless steel, and a fusion weld, said stainless steel consisting essentially of chromium 16.00% to 18.00%, nickel 6.50% to 7.75%, aluminum .75% to 1.50%, carbon .09% maximum, at least one of the group consisting of titanium about .05% to .20%, zirconium about .10% to .40%, and uranium about .30% to 1.00%, and the remainder substantially all iron."

The following patents were relied upon as prior art references in support of the rejection:

  Goller  (1)         2,506,558      May 2, 1950
  Goller  (2)         2,505,762      May 2, 1950
  Nisbet              2,564,498      Aug. 14, 1951
  Pakkala             2,738,267      Mar. 13, 1956
  British Patent        407,052      Mar. 9, 1934

The invention relates to a particular class of stainless steel alloys containing chromium, nickel and aluminum. As pointed out in appellants' specification, stainless steel alloys containing chromium, nickel and aluminum constituents in the proportions recited in appellants' claims are well-known precipitation-hardenable austenitic steels which are particularly useful in the aircraft industry because of their high ratio of strength to weight. The specific object of appellants' invention is stated in the specification as follows:

"In welding the steel, particularly when using the inert-gas tungsten arc technique at substantial welding speeds, i. e., on the order of 25 inches per minute, substantial porosity is found within the region where weld and parent metal meet, that is the sides of the weld bead. This porosity appears as a multiplicity of minute pockets and is inclined to develop areas of weakness. Particularly, the points of porosity established are inclined to cause local stress concentration and ultimate deterioration.

"Now the object of our invention is to overcome the difficulties noted above and give a precipitation-hardenable austenitic chromium-nickelaluminum stainless steel which is weldable by the inert-gas tungsten arc technique, as well as other techniques, in rapid, reliable, efficient manner to give a sound weld free of porosity."

Appellants also state in the specification that:

"* * * we have discovered that in the welding of the precipitation-hardenable chronium-nickel-aluminum stainless steels, especially by way of the inert-gas tungsten arc technique, there are encountered a substantial number of non-metallic inclusions within the region where the weld bead meets the base metal being welded. Actually, these non-metallic inclusions are initially present in the base metal. It is our thought that under the intense heat of the welding arc and the high temperatures reached in the weld metal, certain of these non-metallic inclusions at the sides of the weld are decomposed, yielding gas, a portion of which is there entrapped as the bead solidifies. Gas pockets are formed, giving the objectionable points of porosity. * * *

"Our investigations lead us to feel that the objectionable non-metallic inclusions are aluminum nitrides or perhaps aluminum carbonitrides. We find that stainless steel whether melted in the induction furnace or melted in the electric arc furnace picks up nitrogen on the order of 0.02% to 0.06%, this as a result of the melting operation. And it combines with the aluminum, where present. It is this nitrogen gas which appears with the decomposition of the aluminum nitrides or the carbonitrides and gives rise to the difficulty."

After pointing out the problem existing in the industry and their recognition of the cause of the problem, appellants state their solution to this problem as follows:

"We have discovered that by adding a certain critical amount of titanium to the composition of the precipitation-hardenable austenitic chromium-nickel-aluminum stainless steel, the objectionable porosity in welding is markedly and sharply reduced."

Appellants further point out that small amounts of zirconium or uranium have a similar effect in reducing weld porosity. Appellants have included in their specification tables showing the results of a series of welding tests made on stainless steels containing varying amounts of titanium, zirconium or uranium. These tests show a reduction in weld porosity due to the presence of the above substituents. The validity of these tests has not been questioned by the examiner or board.

"(a) The problem facing the industry was due to objectionable porosity in the weld bead of chromium-nickel-aluminum stainless steels.

"(b) Appellants surmize that the cause of this porosity was the inclusion of aluminum nitrides or aluminum carbonitrides, which, when heated during welding, decomposes, giving off gaseous nitrogen.

"(c) Appellants' solution to the problem which consists in the addition of specified amounts of either titanium, zirconium or uranium to the chromium-nickel-aluminum steel to reduce the objectionable porosity in the weld bead."

The examiner's rejection, affirmed by the board, is that the claimed invention is obvious [a] over the Goller patents, (1) and (2), in view of Nisbet or Pakkala or the British patent or [b] over Goller (2) alone. The disclosures of these references are discussed below.

Goller (1), which is cited in appellants' specification as exemplary of the chromium-nickel-aluminum steels known to the industry, states that stainless steels of chromium-nickel varieties frequently include various elements such as columbium or titanium "for special purposes." The patent then states that:

"* * * Some few of the chromium-nickel stainless steels have been known to respond to hardening by heat treatment, this by virtue of the addition of titanium or columbium in well studied proportionment with other elements present coupled with a critical form of heat treatment to effect precipitation hardening. Columbium and titanium, however, are relatively expensive materials. The chromium-nickel-titanium or chromium-nickel-columbium stainless steels, moreover, usually contain stress laden ferrite as an essential to hardenability from the annealed condition. * * *

* * * * * *

"Accordingly, one of the objects of my invention is the provision of a chromium-nickel-aluminum stainless steel and method, which steel is readily fabricated into a host of different products hardenable at temperatures sufficiently low to avoid substantial scaling and distortion due to heat effect, and yet which are sufficiently stable in the prehardened and prefabricated condition that premature hardening is not encountered in cold weather as in shipping for example."

From the above quoted excerpts, it is clear that while Goller (1) did disclose that titanium could be used in chromium-nickel steels "for special purposes," it is equally clear that he did not disclose a chromium-nickel-aluminum steel with titanium. This latter point was conceded by the solicitor at oral argument. Goller (1) makes no mention of welding his steel or any prior art steel.

Goller (2), also cited in appellants' specification, discloses the prior art's use of titanium with chromium-nickel steels "for special purposes" and further states:

"* * * The titanium also has a thirst for nitrogen and is subject to become tied up in the form of nitrides with consequent impairment of the hardening effect." [Emphasis added.]

It is clear from the remainder of the specification that the patentee does not disclose a chromium-nickel-aluminum steel with titanium and no mention is made of welding. As was true in Goller (1), the object in Goller (2) is to provide a precipitation-hardenable stainless steel by properly proportioning aluminum as the precipitation hardening agent in place of more expensive agents such as titanium.

The patent to Nisbet concerns the preparation of "alloys characterized by exceptionally high strength, ductility and corrosion resistance at high temperatures." Nisbet states that:

"The process involves a combination of steps designed to remove the volatile impurities and reduce the oxygen and nitrogen contents of the alloy. No one of the steps comprising the process is adequate to accomplish the high degree of purity desired although each step contributes to the purifying process."

One of the steps in Nisbet's process is the introduction of an element "having a high affinity for oxygen and nitrogen," which the patentee aptly describes as a "getter." Among the "getters" described and claimed are titanium and zirconium. The patentee makes no mention of welding his alloys.

The patent to Pakkala discloses and claims a stainless steel having chromium-nickel-aluminum in amounts within those ranges claimed in the appealed claims and containing "about .25%" or "below .25%" titanium. Pakkala does not mention welding, but states that reducing the titanium and proportioning the amount of columbium in the steel will reduce the "formation of segregations of a non-metallic nature" which are said to be "an agglomeration of titanium nitrides and titanium cyanonitrides."

The British patent states that the problem of inter-crystalline weakness caused by the formation of free carbides when welding certain stainless steels is alleviated by using a welding rod containing various amounts of titanium. The patentee states that:

"* * * The effect of the titanium is definitely to locate and fix inter-crystalline precipitations in the form of what may be termed titanium cyanonitride, which is made up of carbides and nitrides of titanium. The titanium cyanonitride appears to be very stable and to be unaffected by subsequent heat treatment, and also it appears to have the effect of preventing excessive grain growth, which may be another source of weakness."

The patent, while concerning with the welding of chromium-nickel steel, does not mention chromium-nickel-aluminum steel nor does it mention the problem of porosity in the weld bead.

The Board of Appeals summarized its affirmance of the examiner's rejection on the above described references as follows:

"* * * As a result of our study of the art relied upon by the Examiner, we have come to the conclusion that there is an ample teaching therein of the desirable effect of a small amount of titanium in stainless steel alloys, including the type presently claimed (see Goller (2) and Pakkala). Accordingly we believe that appellants here have merely followed the teachings of the prior art in compounding their alloy. The explanation on page 4 of the specification of the origin of the weld weakness as being nitrogen gas seems to us to lead to the use of titanium to bind this gas because that action or function of the titanium is taught by the references. For these reasons we will sustain the rejection."

Appellants argue that the board erred because none of the references except the British patent mentions the problem of improving the welding characteristics of chromium-nickel-aluminum stainless steels and that the British patent is not concerned with weld porosity. Therefore, since none of the references recognizes the cause of the problem faced by appellants, they do not suggest the solution.

Our decision as to whether the board was correct in finding the claimed invention to be obvious over the cited references requires at the outset a careful analysis of appellants' invention as a whole. As summarized earlier, appellants' specification sets forth that appellants, (a) knowing porosity to be a problem when welding chromium-nickel-aluminum steels, (b) surmized that this undesirable porosity was caused by gaseous nitrogen given off when aluminum nitrides or carbonitrides were heated during the welding, and (c) solved the problem by introducing a specified amount of titanium, zirconium or uranium to reduce the nitrogen available to form the aluminum nitrides or carbonitrides which caused the porosity.

35 U.S.C. § 103 requires us to compare the prior art and the subject matter sought to be patented to see if 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." [Emphasis added.] Our initial consideration must, therefore, be to determine from appellants' disclosure and claims just what constitutes "the subject matter as a whole."

Appellants contend that their invention, considered as a whole, includes both the recognition of the cause of the weld bead porosity and its elimination and that nothing in the cited references in any way suggests a prior art recognition of the problem or its cause and hence could not and did not suggest any solution of this problem of weld porosity. Viewing the invention as a whole, we agree with appellants that the references would not, at the time of appellants' invention, lead a person of ordinary skill in the art to recognize the cause of weld porosity in chromium-nickel-aluminum stainless steels. While we agree with the board that the cited prior art, particularly the Nisbet patent, clearly suggests that titanium or zirconium2 may be used as "getters" for nitrogen, this solution to the problem is only a part of the invention which 35 U.S.C. § 103 specifies must be viewed "as a whole."

In In re Antonson, 47 CCPA 740, 741, 272 F.2d 948, the problem facing the art was the failure of airplane tires when landing at high speeds with heavy loads. Appellants, recognizing that these failures were caused by concentrated stresses in the central portion of the tire tread due to centrifugal force and high air pressure, solved the problem by reducing the stress concentrations in the tread by including a plurality of plies of reinforcing cord. In reversing the rejection, the court stated:

"* * * The obvious problem was simply the failure of tires on airplanes landing at high speeds with heavy loads. Much of what we have quoted above is, in fact, appellant's own analysis of the causes of the failure which led him to the means for its solution. In cases of this kind it must not be lost sight of, as pointed out by the Supreme Court in Eibel Process Co. v. Minnesota & Ontario Paper Co., 261 U.S. 45, 67, 43 S. Ct. 322, 67 L. Ed. 523 that [1] the inventive act which entitles an applicant to a patent resides as well in the discovery of the source of trouble as in the application of the remedy."

See also In re Pennington, 241 F.2d 750, 44 CCPA 789, In re Shaffer, 229 F.2d 476, 43 CCPA 758, and In re Tanczyn, 202 F.2d 785, 40 CCPA 886.

The solicitor, in his brief, argues that the claims were properly rejected and that the above cases cannot be relied upon because:

"* * * While the appellant in each cited case solved a problem, the physical means by which the problem was solved, was recited in the claims which the Court found to have been improperly rejected by the Patent Office. Here, on the other hand, porosity is neither recited in, nor is a necessary factor to be considered in the claims, since the claim language does not require that appellants' steel be put to a use involving welding, or that it be welded under conditions conducive to the production of porosity."

We do not understand the solicitor's position since it seems to us that the "physical means by which the problem was solved" is the inclusion of the small amounts of titanium, zirconium or uranium, as recited in the claims.3 The above statement of the solicitor seems to equate the "means" by which the problem was solved with "porosity." This statement therefore equates the problem with its solution, which, as pointed out, supra, evidences a misunderstanding of the invention "as a whole." Regardless of what is meant by the solicitor's statement, he and the board seem to have ignored what we think is the essence of appellants' invention, i. e., the determination of the cause of weld porosity which, we think, is an inseparable part of the "invention as a whole," and makes the invention here claimed unobvious over the suggestions of either the combination of prior art references or over the Goller (2) references alone.

Viewed in the light of appellants' specification, the solution to the problem of weld porosity may seem obvious over the combined teachings of the references. Such a hindsight analysis, however, is not allowed by 35 U.S.C. § 103 which requires a comparison of the prior art and the invention as a whole at the time the invention was made. In re Rothermel and Waddell, Jr., 276 F.2d 393, 47 CCPA 866.

"It is easy now to attribute to this prior art the knowledge which was first made available by appellants and then to assume that it would have been obvious to one having the ordinary skill of the art to make these suggested reconstructions. While such a reconstruction of the art may be an alluring way to rationalize a rejection of claims, it is not the type of rejection which the statute authorizes. 35 U.S.C. § 103 is very specific in requiring that a rejection on the grounds the invention "would have been obvious" must be based on a comparison between the prior art and the subject matter as a whole at the time the invention was made.

For the foregoing reasons, the decision of the Board of Appeals is reversed.

Reversed.