This appeal is from the decision of the Patent Office Board of Appeals1 affirming the examiner's rejection of product claims 1 through 17 and process claims 18 through 20 in application serial No. 34,384, filed June 7, 1960, for "Stabilized Polypropylene." No claim has been allowed.

The issue is patentability over the prior art and, in order to develop a clearer picture of what the invention is, its background, and its relation to the art, we shall first present a chronological discussion of that art. The references relied on by the examiner and board are:

  Naunton et al.
    (British)             684,976      Dec. 31,1952
  Imperial Chemical
    (Italian)              495,814      June 25, 1954
  Tholstrup et al.       3,001,969      Sept. 26, 1961
                                          (filed July 8, 1957)
  Hawkins et al.         2,889,306      June 2, 1959
                                          (filed July 15, 1957)

Additional references relied on by appellants are:

  Hardy et al.          2,976,259    Mar. 21, 1961
                                          (filed Sept. 5, 1956)
  Belgian patent           579,636     Published Dec. 1959
  "Modern Plastics," Vol. 37, page 192, Jan. 1960

Naunton discloses that "the degradative effect of bright sunlight on polythene4 may be minimized or obviated by incorporating with the polythene a nickel salt of certain dialkyldithiocarbamic acids" in amounts preferably between 0.5% and 5.0% by weight of polymer. Lesser amounts are said not to appreciably improve stability, and amounts greater than 5.0% "have very little additional effect." The two alkyl groups together contain at least 8 carbon atoms — one of the lowest members of that series, nickel dibutyldithiocarbamate, being specifically disclosed.

The Italian patent is also directed to the reduction or elimination of polyethylene degradation due to direct sunlight by incorporating therein a dialkyldithiocarbamate. The specification is in many respects similar to that of the Naunton British patent to I.C.I., just discussed, one difference being a preferred upper limit of stabilizer of 3 instead of 5% by weight of polymer, another being no express upper carbon atom limitation of the alkyl substituent, still another and more important difference being that a dithiocarbamate of any of 56 elements can be used, viz., those of atomic numbers 21 to 34, 39 to 52, and 57 to 84. Nickel, iron, cobalt, copper, cadmium, lead, bismuth, selenium, and zinc dithiocarbamates are specifically disclosed among some 28 examples.

Tholstrup relates to the stabilization of thermoplastic hydrocarbons such as polyethylene, polypropylene, etc., against heat degradation during extrusion, melt spinning, and other high temperature operations by incorporating a zinc dialkyldithiocarbamate in which the alkyl radicals are lower alkyl containing at least 2 carbon atoms. The polypropylene polymers which can be stabilized include "high density hard solids and crystallizable polymers which can be spun to form useful textile fibers or oriented film."

Hawkins discloses the stabilization of polypropylene, polyethylene, and other specifically listed hydrocarbon polymers against both "thermal oxidation," by which Hawkins means oxidation independent of ultraviolet absorption, and oxidation due to ultraviolet absorption. The rather scholarly discussion points out that whereas both influences are believed to be due primarily to the same oxidation mechanism, different kinds of materials are required to effectively retard each type of deterioration. For example, carbon black is disclosed as a conventional retardant for ultraviolet oxidation while "a general requirement of * * * antioxidants [for thermal oxidation] is that they contain an antioxidant group such as the secondary amino or phenolic group attached to an aromatic ring * * *." It is said that great difficulty had been encountered in attempts to prepare polymeric compositions which are at the same time stabilized against both influences. For example, since certain carbon blacks had been known to have a mild thermal antioxidant effect in addition to shielding the substance against ultraviolet radiation, it was expected that the incorporation of the usual thermal antioxidants into polymeric materials containing such carbon blacks would result in increased stability against thermal oxidation. It was found, however, that not only is the effect not additive, but that the effectiveness of the antioxidant in the presence of carbon black is reduced several fold — in many instances being rendered completely ineffective. Hawkins discovered that a class of materials, thiuram disulfides bearing four N-aliphatic radicals, could be combined with carbon black in his contemplated polymeric materials to obtain a product which "compares favorably" in stablization characteristics with that same material containing the most effective commercially available antioxidants in the absence of carbon black. These thiuram disulfides, says Hawkins, are "for the most part ineffectual in the absence of carbon black * * *."

The Patent Office analysis of this art and conclusions as to what would be suggested thereby to one of ordinary skill in the art can be summarized thus: Polyethylene is "closely related" to polypropylene. One skilled in the art, accordingly, would expect an ultraviolet light stabilizer for polyethylene to be effective as an ultraviolet stabilizer in polypropylene. There can be little doubt about this similarity in behavior of stabilizers in polyethylene and polypropylene for both Tholstrup and Hawkins teach the stabilization of hydrocarbon polymers as a class, which class includes both polyethylene and polypropylene. Heat stabilization of the polymers as a class is taught by Tholstrup, and light stabilization of the polymers as a class using carbon black is taught by Hawkins. Furthermore, since Naunton discloses nickel dialkyldithiocarbamates and the Italian patent teaches metal (including nickel) dialkyldithiocarbamates broadly to be ultraviolet light stabilizers for polyethylene, use of these compounds in polypropylene as ultraviolet light stabilizers would be suggested. In brief, it would be entirely obvious, say the examiner and the board, for a skilled chemist at least "to try to stabilize polypropylene with a known stabilizer for polyethylene." We will return to these points of the Patent Office argument later.

We now turn to what appellants claim as their invention. Claims 1 and 18 are illustrative:

1. A light-stable composition comprising solid, isotactic, substantially crystalline polypropylene and a stabilizing quantity of a dithiocarbamate having the general formula:

NOTE: OPINION CONTAINING TABLE OR OTHER DATA THAT IS NOT VIEWABLE

wherein each of R, R₁, R₂, and R₃ is a hydrocarbon radical containing 1 to about 18 carbon atoms and M is selected from the group consisting of nickel and cobalt.

18. A process of inhibiting degradation of polypropylene caused by exposure to light which comprises admixing solid, isotactic, substantially crystalline polypropylene and a stabilizing quantity of dithiocarbamate having the general formula:

NOTE: OPINION CONTAINING TABLE OR OTHER DATA THAT IS NOT VIEWABLE

wherein each of R, R₁, R₂, and R₃ is a hydrocarbon radical containing 1 to about 18 carbon atoms and M is selected from the group consisting of nickel and cobalt.

Claims 2 through 17 are dependent on claim 1 and more specifically define M, the various R groups, and "stabilizing quantity," the largest recited range for the latter being "from about 0.05% to about 5% by weight of said composition." Claim 19 and 20 are similarly dependent on claim 18, differing therefrom only in the amount of stabilizer.

Appellants' class of dithiocarbamates, while quite extensive, is so only by reason of the broad definition of the R substituents; it is quite limited in the definition of M. There is no dispute about breadth or operability. To the contrary, the attention of both parties has been centered on only so much of appellants' class of dithiocarbamates as is disclosed in the above discussed prior art to be useful as light stabilizers for polyethylene, viz. those compounds wherein the R groups are alkyl, particularly lower alkyl. We will treat the case accordingly.

The issue therefore is: would nickel and cobalt dialkyldithiocarbamates be obvious effective light stabilizers for polypropylene in view of the array of known stabilizers for polyethylene?

The parties pursue two distinct, but related, analyses of the case. One reason for this is that each provides, and of course stresses, its chosen body of prior art. Another derives from what appear to be divergent views about the nature of appellants' invention. The Patent Office looks strictly to what is claimed; appellants stress the surrounding circumstances and background of their invention as well, pointing out what they did as well as what is ultimately claimed. We agree with appellants that claims should not be considered in the abstract, particularly when the invention is the result of a selection or screening process. While the ultimate question, of course, is whether the invention defined by the claims is patentable over the art, frequently evidence on background and the circumstances surrounding the making of the invention is helpful in deciding that question.

As we see it, appellant's invention is the discovery of what stabilizers for other materials, known in the art, will, and which will not, stabilize polypropylene against degradation by light. The solicitor asserts that one skilled in the art "would expect an[y] ultraviolet light stabilizer for polyethylene to be effective as an ultraviolet stabilizer in polypropylene." If this be true, then appellants have clearly discovered, uncovered, or invented the unobvious for the record shows that many which they tested were found not to be effective. The examiner, with whom the board expressed "total agreement," did not go that far, saying "it would be obvious for a skilled chemist to try to stabilize polypropylene with a known stabilizer for polyethylene," and that it would be "routine experimentation for a skilled chemist to attempt to stabilize polypropylene against the deteriorative effect of light by first trying the known stabilizers for polyethylene such as the nickel and cobalt dialkyldithiocarbamates," citing In re Moreton, 288 F.2d 940, 48 CCPA 928, for the proposition that obviousness does not require absolute predictability. Our reply to this view is simply that it begs the question, which is obviousness under section 103 of compositions and methods, not of the direction to be taken in making efforts or attempts. Slight reflection suggests, we think, that there is usually an element of "obviousness to try" in any research endeavor, that it is not undertaken with complete blindness but rather with some semblance of a chance of success, and that patentability determinations based on that as the test would not only be contrary to statute but result in a marked deterioration of the entire patent system as an incentive to invest in those efforts and attempts which go by the name of "research."

Appellants contend that the combined effect of all the art of record, including that cited by them, shows the invention to be unobvious. Certainly they are entitled to cite art for this purpose, which must then be considered. See In re McKenna, 203 F.2d 717,40 CCPA 937. Principal reliance is placed on "Modern Plastics," which says:

Of all the problems that polypropylene producers have faced, stabilization has been perhaps the most difficult. Polypropylene, when it is unstabilized, deteriorates rapidly upon exposure to heat or ultra-violet light. The difficulty with proper stabilization of polypropylene is that, in general, the vast numbers of stabilizers developed for other polyolefins, vinyl chlorides, and the like proved ineffective for polypropylene. [Emphasis ours.]

It was necessary to develop unique systems for specific end uses.

Hardy and the Belgian patent, also relied on by appellants, disclose dihydroxy alkoxybenzophenones and nickel phenolate of a monosulfide of bis-(p-alkyl phenol) respectively, other classes of known stabilizers, certain members of which appellants' specification shows to be very poor stabilizers for polypropylene, notwithstanding the Belgian abstract is directed to polypropylene, without mentioning the type of stabilization involved, and Hardy is directed to ultraviolet stabilization of a "tremendous array of suitable synthetic resin carrier materials * * * [including] the polyolefins, as for instance, the polymers of ethylene, propylene, isobutylene, etc. * * *."

Example 18 of Hardy shows a 0.02 inch thick sheet of polypropylene containing 0.1% by weight of 2,2'-dihydroxy-4-methoxybenzophenone to be superior in resistance to ultraviolet deterioration after 200 hours exposure than a control sheet containing no benzophenone. Appellants' specification shows that 125 to 150 denier polypropylene monofilaments, containing from 0.5% to 2.0% by weight of this same compound, break after 20 to 40 hours exposure, and upon those results appellants base part of their argument of unobviousness. The solicitor makes a point of the differences between Hardy's and appellants' test conditions, contending that since the claims are not limited to monofilaments appellants' results would not be indicative of nonobviousness.

Appellants' specification contains some 59 examples which allegedly demonstrate the unexpected nature of their invention. The first 17 show the substantial effectiveness of nickel and cobalt dithiocarbamates. Examples 18 to 44 show the ineffectiveness, under conditions used in examples 1 to 17, of the following "stabilizers" disclosed in the art of record: certain zinc, lead, selenium, bismuth, and tellurium dialkyldithiocarbamates; various tetra (lower alkyl) thiuram disulfides; nickel stearate; certain hydroxy methoxybenzophenones; and octyl phenyl salicylate. Examples 45 to 59 show the "hours to break" and "% tenacity retained" (after a given number of hours) of monofilaments of polypropylene containing small percentages of the following: cobalt dibutyldithiocarbamate, nickel dibutyl- and di-sec. butyldithiocarbamate, ferric dibutyldithiocarbamate, and a nickel phenolate. With one exception, the claimed compositions were found to be greatly superior to those not claimed. That exception is that the results of nickel phenolate compositions compared favorably with those of nickel di-sec. butyldithiocarbamate compositions. For comparison purposes non-stabilized polypropylene monofilaments were similarly tested and found to be quite inferior.

Thus, the crux of the Patent Office position is that polyethylene and polypropylene are so closely related that one of ordinary skill in the art, seeking an ultraviolet stabilizer for polypropylene, would expect those known in the art as useful stabilizers for polyethylene to be useful stabilizers for polypropylene. If that theorem falls, its entire case falls.

The crux of appellants' position is that "Modern Plastics," combined with those examples in the application showing the ineffectiveness of various prior art stabilizers, is evidence showing appellants' invention to be unobvious — that those skilled in the art would not expect the polyethylene stabilizers to be useful stabilizers for polypropylene — particularly when considered in view of Hawkins, Hardy, Tholstrup, the Italian patent, and the abstract of the Belgian patent. That is, to the extent art cited and relied upon by the Patent Office makes out a "prima facie case of obviousness," appellants allegedly have presented sufficient evidence to rebut that case.

After extensive deliberation, we are disposed to agree with appellants and reverse the rejection of all claims. However, for reasons which will be discussed later, we must exclude from the reversal claims 1, 2, 4, 5, 7, 8, 10-13, and 18-20 — those composition and process claims reading on nickel dithiocarbamates as the stabilizing ingredient.

Considering the evidence as a whole, we think it inescapable that we are here dealing with an art that is quite empirical. Nearly every reference of record speaks of the unexpectedness of the behavior of "related" materials, or of problems such as discoloration encountered in attempting to use many then known stabilizers in new materials. Exemplary is the following passage from Tholstrup:

The close relationship between zinc and cadmium apparent from the periodic table indicates that the cadmium salts which correspond to the zinc salts of this invention would function somewhat analogously, and it may be that they do; however, even here there is no known basis for predictability of results since investigations in the art in general reveal a wide variation in effects between various metal salts without any apparent basis for prognostication.

In addition, we consider quite significant the above passage from "Modern Plastics," notwithstanding that disclosure is general and might even be said, by one desiring to find obviousness of the invention, to imply that some prior art stabilizers do work with polypropylene. We think such reasoning is farfetched, however, and that it fails to answer the further, more critical question: which ones? There is not the slightest hint of even the class wherein to search. As to the point about generality, appellants' specification provides numerous examples of the unworkability in polypropylene of specific stabilizers, disclosed in the examiner's cited art as effective in polyethylene.

Further, it is of particular importance, we think, that no reference of record expressly discloses stabilizing polypropylene against the degradative effects caused by light, with the exception of example 18 in Hardy, to which we are not inclined to give much weight since the stabilizer there used, 2,2'-dihydroxy-4-methoxybenzophenone, is quite unlike appellants' metal dialkyl-dithiocarbamates, and also since a polypropylene filament containing that compound was found by appellants to break after 20-40 hours exposure to ultraviolet light. We think this offsets what evidence of obviousness is to be found in that example.

There remains for consideration the legal effect to be given the following disclosure in Tholstrup:

In addition to the excellent degree of heat stability provided by this invention, it is quite significant that the zinc di-lower alkyl-dithiocarbamates do not produce discoloration in the stabilized thermoplastic hydrocarbons. Other salts, such as the nickel salt, suffer serious disadvantage from the discoloration standpoint which militates against their use in producing the colorless products which are often much desired.

The second paragraph of Tholstrup's specification says the invention relates to stabilizing "high polymers of mono-a-olefins containing 3 to 8 carbon atoms * * *." The specification as a whole leaves no doubt that polypropylene is of particular concern. We therefore conclude that polypropylene compositions containing nickel di-lower alkyl dithio-carbamates are unpatentable over Tholstrup for the reason stated in the examiner's answer, that novelty is required for patentability and novelty is lacking as to the claimed compositions stabilized with nickel salts.

In finding anticipation of some claims by Tholstrup, however, we stress three points: (1) that polypropylene and nickel salts have each been clearly singled out by Tholstrup from classes of materials, (2) that the negative aspects of Tholstrup's disclosure go not to the making of the composition here claimed, but rather to its use for Tholstrup's purpose, and (3) that appellants have made no statement tending to establish the incorrectness of that disclosure. As to the claim expression "light-stable composition," we think it immaterial that Tholstrup does not in haec verba make that disclosure. The claims as a whole must be analyzed to see if the composition defined thereby is distinguished in terms from the prior art. In re Neugebauer, 330 F.2d 353, 51 CCPA 1138. Since we think those composition claims reading on nickel dithiocarbamates are not so distinguishable, the rejection of those claims is affirmed.

We likewise must affirm the rejection of process claims 18-20, notwithstanding appellants' urgings that those claims "stand in a different position regarding obviousness" than the composition claims by setting forth "a use and the means whereby such new use is obtained," thus coming "within the definition of a process set forth in 35 U.S.C. 100(b)" and that this "new use of the nickel and cobalt dithiocarbamates as an ultraviolet stabilizer for polypropylene constitutes one of the fundamental aspects of appellants' invention." As indicated, Tholstrup shows the combination of polypropylene with nickel dithiocarbamates to be old. While the method of combining is not clear, we thing it conventional, and therefore obvious, to do so by "admixing," as claimed. We therefore find process claims 18-20 unpatentable by reason of their reading on an obvious process, the admixing of polypropylene and a nickel dithiocarbamate, an old mixture.

As to the introductory language, "A process of inhibiting degradation of polypropylene caused by exposure to light," again we do not think these words can serve to patentably distinguish the claimed process from the prior art. That language, in effect, states the result of admixing the two materials. While the references do not show a specific recognition of that result, its discovery by appellants is tantamount only to finding a property in the old composition, not in the nickel compound for which, it is argued, a new use has been found.

We need not comment in this regard about cobalt dithiocarbamates since no process claim is limited thereto.

Reference to the Tholstrup disclosure on which we feel compelled to affirm the rejection of claims reading on compositions stabilized with the nickel salts will show, in view of its negative character which says, in effect, that the patentee does not favor the use of nickel salts for any purpose, the rejection to be a highly technical one for lack of novelty and not based on any real teaching of the effect of nickel salts on stabilization against ultraviolet light. For this reason we do not believe this rejection should carry with it the claims limited to the use of cobalt salts on a theory that the nickel salts and cobalt salts are closely related in chemical properties. In view of the empirical nature of this invention, as above analyzed, and in view of the lack of any teaching in the prior art about the light-stabilizing effect of the nickel salts in polypropylene, we think that portion of appellants' invention concerned with cobalt salts remains patentable notwithstanding our partial affirmance of the rejection.

The decision of the board is affirmed as to claims 1, 2, 4, 5, 7, 8, 10-13, and 18-20, and is reversed as to claims 3, 6, 9, and 14-17.

Modified.

SMITH, Judge (concurring in part and dissenting in part, with whom MARTIN, J., joins).

The majority affirms the rejection of claims 1, 2, 4, 5, 7, 8, 10-13, and 18-20 as unpatenable over the prior art use of nickel dithiocarbamates in stabilizing polypropylene. I agree with this portion of the majority opinion. However, I dissent from the majority opinion in which it reverses the board as to the rejection of claims 3, 6, 9, and 14-17. While I agree with the majority that the art with which we are here dealing is an empirical art and there is no direct teaching that polypropylene can be stabilized by the use of the cobalt salts as claimed in these claims, it seems to me that these claims do not define a non-obvious invention when measured against the requirements of 35 U.S.C. 103.

If we compare the claims which we agree should be rejected with those whose rejection the majority would reverse, we find that composition claims 3, 6, 9, and 14-17 are dependent claims which are based on the rejected claims. The only difference in these claims is that instead of the nickel salt specified in the rejected claims, the corresponding cobalt salts are specified therein.

I note the majority's reference to the reading of the rejected claims on the prior art which suggests to me that the rejection is treated by the majority essentially as an anticipation rejection under 35 U.S.C. § 102(a). I think there is no question, despite the generality of some of the terms in which the rejection was phrased by the examiner and affirmed by the board, that the real statutory basis with which we should be here concerned is obviousness of the invention under 35 U.S.C. § 103. The board stated:

* * * we find ourselves in agreement with the Examiner that claims 1 through 20 are "unpatentable over Naunton et al. or Imperial Chemical alone or in combination with Tholstrup et al. or Hawkins et al. under 35 U.S.C. 103." It is the Examiner's stated position that where the prior art disclosed stabilization of polyethylene against the effects of light by means of nickel or cobalt dialkyldithiocarbamate (Naunton et al. and Imperial Chemical), it would be obvious for a person skilled in the art to try these salts to stabilize polypropylene, particularly in view of the suggestion in Hawkins et al. and Tholstrup et al. to the effect that certain stabilizers can be used effectively for both polyethylene and polypropylene. * * *

The majority's disagreement with this proposition pivots about the phrase "obvious * * * to try," and appears to support its ultimate conclusion by finding that there was no prior art which taught the use of claimed cobalt salts to stabilize polypropylene compositions. I fully agree that in here considering the invention as a whole, the selection of the particular cobalt salts is a factor to be weighed before reaching the ultimate legal conclusion of obviousness or non-obviousness required by 35 U.S.C. § 103. However, I do not agree with what seems to me to be the majority's rationale that such selection per se supports a finding of unobviousness under the facts here. I do not find that we are here concerned with a problem of the selection of particular materials from a universe which presents many choices. The art here is so developed that the majority finds the use of nickel salts to stabilize polypropylene to be unpatentable. In the composition claims 1, 4, 7, as well as in the process claims 18-20, the rejection of which the majority affirms, the stabilizing salt is selected from the group consisting of nickel and cobalt.

Thus, the issue here turns on the specific selection of the claimed cobalt salts for stabilizing polypropylene in view of the prior art and in the light of the agreed unpatentability of the invention where the corresponding nickel salts are claimed for the same purpose. The prior art teachings showed stabilization of polyethylene from degradation due to sunlight in Naunton, Italian and Hawkins. Naunton and Italian disclose the use of nickel dibutyldithiocarbamate for this purpose. Italian further discloses a group of metal salts including cobalt dibutyldithiocarbamate for stabilizing of polyethylene. Prior art recognized similarities between polyethylene and polypropylene as regards stabilization are shown in Hawkins and Tholstrup both of which teach that both polyethylene and polypropylene may be stabilized by the addition of the same compounds and Hawkins deals with degradation by sunlight. Thus, it seems to me one of ordinary skill in this art would be led by the Italian reference directly to the choice of cobalt dibutyldithiocarbamate as a specific salt which is there shown to be effective in stabilizing polyethylene against degradation from sunlight. Thus the only difference between appellant's claimed invention and the prior art is that they followed Tholstrup and Hawkins' teachings and treated polypropylene with what the art had used to stabilize polyethylene. These circumstances suggest that the factor of selection stressed by appellants should be given relatively litthe weight in reaching the legal conclusion of obviousness under section 103.

The selection argument advanced by appellants and relied upon by the majority is not supported by the facts of record. The majority states, "There is not the slightest hint of even the class [of stabilizers] wherein to search." It seems to me that this statement does not consider the teachings as a whole of the Italian patent. It is true that Italian begins by stating a large number of metal dithiocarbamates may be used. However, five specific examples employing metal dithiocarbamates are set forth. Example 1 shows tests with iron, cobalt and nickel; example 2 employs copper, cadmium, lead, bismuth and selenium; examples 3 and 4 show nickel and zinc; and example 5 shows nickel. The qualitative results shows the preferred group of metals by a substantial margin is iron, cobalt and nickel with nickel being the best metal by a very small margin. Four of the five tests set forth employ nickel showing the effectiveness of it as a stabilizer.

Appellants set forth certain metal stabilizers which are allegedly ineffective, i. e., zinc, lead, selenium and bismuth. Italian specifically discloses the ineffectiveness of these metals as stabilizers as compared to the preferred group of iron, cobalt and nickel.

Appellants, in citing art to demonstrate the non-obviousness of their invention, place principal reliance on the statement quoted by the majority from Modern Plastics. I find that this statement is only general in nature and encompasses stabilizers developed for "polyolefins, vinyl chlorides and the like." It seems to me that the specific teachings in Italian, Tholstrup, etc., relied upon by the Patent Office, are the "teachings which a person of ordinary skill in the art would consider the most pertinent and accord the most weight" as they are "closely concerned with the claimed invention." In re Lunsford, 357 F.2d 385, 53 CCPA ____.1

On the present record, it seems to me that either the substitution of cobalt salts for nickel salts or the substitution of polypropylene for polyethylene in the art of record would have been obvious to one of ordinary skill in this art under the conditions specified in 35 U.S.C. § 103. I would, therefore, affirm the decision of the board as to claims 3, 6, 9 and 14-17.