PER CURIAM.
This appeal is from the July 8, 1992, decision of the United States Patent and Trademark Office (PTO) Board of Patent Appeals and Interferences (Board) affirming the examiner’s rejection of claims 6 through 10 of Michihiko Ochiai
et alls
(collectively “Ochiai”) application serial no. 07/462,492, claiming priority from parent application serial no. 642,356, filed December 19,1975, now U.S. Patent No. 4,098,888 (methods for the manufacture of cephems).
Ex parte Ochiai,
24 USPQ2d 1265 (Bd.Pat.App. & Int.1992). The real party in interest is Takeda Chemical Industries, Ltd., the assignee of any patent issuing from the application.
The rejection of the above claims was predicated on an asserted view of the law of obviousness, per 35 U.S.C. § 103, in view of the combined teaching of six references.
Because, under the legally correct method for determining obviousness, the claimed process is not obvious in view of the cited prior art references, we reverse.
The Invention
Ochiai’s application is directed to a process for using an acyl side chain from a particular type of organic acid having a 2-aminothiazo-lyl group, and a particular type of amine to make a particular cephem compound having antibiotic properties. Claim 6, the principal claim on appeal,
is as follows:
6. A process for preparing a cephem compound of the formula:
[[Image here]]
wherein R3 is hydrogen or methoxy, R4 is hydrogen or a residue of a nucleophilic compound, R5 is hydroxyl or a protected hydroxyl, and R8 is hydrogen or a halogen, or a pharmaceutically acceptable salt or ester thereof, which comprises introducing an acyl group of the formula:
[[Image here]]
wherein R5 and R8 are as defined above into the amino group of the molecule of the formula:
[[Image here]]
wherein R3 and R4 are as defined above or a salt or ester thereof.
Id.
at 1266.
Ochiai’s U.S. Patent No. 4,298,606 covers the cephem compound resulting from the process of claim 6, and Ochiai’s U.S. Patent No. 4,208,899 covers the organic acid used in the process of claim 6.
Id.
at 1267. In other words, viewed as of the time the claimed process was invented, claim 6 recites a process of using a new, nonobvious acid to make a new, nonobvious cephem. The ’606 and ’899 patents, like the application at bar, claim priority from the December 1975 parent application.
The Rejection
The examiner rejected claims 6 through 10 as obvious in light of the combined teaching of the six references noted above. All six references, as Ochiai acknowledges, teach the use of a type of acid to make a type of cephem by a standard acylation reaction with the very same amine recited in claim 6. The examiner explained the rejections thusly in his answer to Oehiai’s appeal to the Board:
It must again be stressed that the citation of six references is to demonstrate convincingly that a
standard, conventional
process of preparing cephalosporin com
pounds is being claimed. The
only
difference between what is being claimed and the prior art is the selection of a
slightly
different acylation agent
[ie.,
acid] to result in a slightly different final product. The
closest
prior art of the six references is represented by the Cook et al. 4,024,138 and, Gregson et al. patent 4,024,134. These two references use [sic, are] quite similar in their disclosure, Cook being the
most
[sic, more] relevant. Both of these references
genetically
disclose the “2-ami-no-thiazolyl” group which appellants seek to introduce....
The examiner recognizes that the
specific “2
amino thiazolyl” moiety has
not
been
specifically
named in [the] Cook et al[J patent. However, Cook et al. when viewed from the standpoint of one skilled in the art would recognize the use of “2-amino-thiazolyl” if the final products sought were to contain this moiety. This merely states the obvious....
The facts presented here are
identical
to those that occurred in the Durden decision
(In re Durden
[763 F.2d 1406] 226 USPQ 359). The
acylating
agent herein being used has been patented by appellants, see Ochia et al. 4,203,899. The final products have also been patented by appellants which appellants acknowledge, brief page 5 footnote 4. The
only
difference between the facts in Durden those Durden [sic] and the instant situation is that appellants have not
admitted
on the record that the process is obvious. Appellants seek to distinguish the Durden decision based on this difference. However, the Durden decision is believed to be controlling because of the
reasoning
used therein and not an admission or lack of admission of the obviousness of the process. The references discussed above abundantly demonstrate the
routineness
of the claimed process. Thus, the Court rejected the argument that a conventional manipulation or reaction was
unobvious
“notwithstanding the specific starting material or resulting product or both, is not to be found in the prior art”.
(Emphasis in original). Importantly, the examiner conceded the total absence from the prior art of both the acid used and the ce-phem made in the process recited in claim 6. In addition, the examiner discussed no references containing any suggestion or motivation either (a) to reject known acids and select instead the particular one used in claim 6, or (b) to obtain the particular ce-phem made according to the process of claim 6.
On appeal, the Board affirmed the examiner’s rejection. After reviewing the examiner’s reliance on
In re Durden,
763 F.2d 1406, 226 USPQ 359 (Fed.Cir.1985), and the “standard” nature of the acylation reaction disclosed in the rejected claims, the Board acknowledged Ochiai’s contention that the fact that “neither the final product nor the method of introducing the particular [acid] component were known, obvious or even remotely suggested in the prior art ... should be dispositive of the obviousness of the invention” recited in claim 6.
Ochiai,
24 USPQ2d at 1267. The Board did not, however, find Ochiai’s contention persuasive. According to the Board,
[ w]e are not here concerned with the pat-entability of the starting materials, the final compounds or other processes of making the [cephem] compounds. We are concerned only with the claimed process and the patentability thereof. Cases such as
In re Larsen,
292 F.2d 531, 49 CCPA 711, 130 USPQ 209 (CCPA 1961);
In re Albertson,
332 F.2d 379, 51 CCPA 1377, 141 USPQ 730 (CCPA 1964) and, particularly,
In re Durden, supra,
all of which were directed to processes of making chemical compounds, are controlling herein.... In each case, a material A, either known or novel, was subjected to a standard process of reacting with a standard reactant, B, in order to produce the result expected from the reaction of A with B. Indeed in
Albertson
as in the instant case, the only manipulative step of the process is that which is embodied in the word “reacting.”
Id.
The Board also rejected Ochiai’s assertion that cases such as
In re Pleuddemann,
910 F.2d 823, 15 USPQ2d 1738 (Fed.Cir.1990),
In re Moncy,
499 F.2d 1289, 182 USPQ 303 (CCPA 1974), and
In re Kuehl,
475 F.2d 658, 177 USPQ 250 (CCPA 1973), are in tension with
Durden
and
Albertson
and counsel allowance of the rejected claims. Distinguishing
Pleuddemonn, Money,
and
Kuehl
as “method of using” rather than “method of making” cases, the Board summarized its decision as follows:
In the case before us, appellants have admitted the claims are directed to a process of making a desired AB product. The process steps, “introducing” A into AB or “reacting” A with B are standard processes used by practitioners in the prior art for reacting similar A moieties with the same B moiety. We are in agreement with the examiner that there is nothing unobvious in the particular
process
chosen and claimed by the appellants.
Ochiai
24 USPQ2d at 1270 (emphasis in original).
Ochiai appeals, contending that both the examiner and the Board failed to apply the proper test for obviousness established by
Graham v. John Deere Co.,
383 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545, 148 USPQ 459 (1966), and its progeny. Specifically, according to Ochiai, both the examiner and the Board, on the assumption that our decision in
Durden
controlled the outcome of the instant ease, failed to weigh the specific differences between the claimed invention — with
all
its limitations — and the prior art references, the so-called “second
Graham
factor.”
See id.
at 17, 86 S.Ct. at 693-94 (“Under § 103 ... differences between the prior art and the claims at issue are to be ascertained]].]”). In addition, Ochiai contends that the decisions in
Money
and
Kuehl
which, like all Court of Customs and Patent Appeals decisions, were in bane, limit the decision in
Albertson
to its facts.
The Solicitor, while defending the correctness of the Board’s conclusion and, unlike the Board itself, doing so in the familiar terms of
Graham,
also asserts that a supposed irreconcilable conflict in our cases — between
Al-bertson
and
Durden,
on the one hand, and
Pleuddemonn,
on the other — “makes it very difficult for patent attorneys to give cogent advice to clients or for patent examiners to render consistent decisions on the patentability (under § 103) of processes involving the use of new and unobvious starting materials.” Unlike Ochiai, however, the Solicitor asks us to take the opportunity to reaffirm the vitality of
Albertson
and
Durden
in the course of deciding this appeal.
The Issue
The issue before this court is whether the Board erred in upholding the examiner’s rejection of claim 6 as obvious under 35 U.S.C. § 103 in view of
Larsen, Albertson,
and
Dur-den
as interpreted by the PTO when neither the particular acid used nor the particular cephem produced is either taught or suggested by the art that predates the parent application.
The Analysis
The test of obviousness
vel non
is statutory. It requires that one compare the claim’s “subject matter as a whole” with the prior art “to which said subject matter pertains.” 35 U.S.C. § 103. The inquiry is thus highly fact-specific by design. This is so “whether the invention be a process for making or a process of using, or some other process.”
Kuehl
475 F.2d at 665, 177 USPQ at 255. When the references cited by the examiner fail to establish a
prima facie
ease of obviousness, the rejection is improper and will be overturned.
In re Fine,
837 F.2d 1071, 1074, 5 USPQ2d 1596, 1598 (Fed.Cir.1988).
Applying this statutory test to the art of record, we conclude that Ochiai’s process invention as claimed is not
prima facie
obvious. The process invention Ochiai recites in claim 6 specifically requires use of none other than its new, nonobvious acid as one of the starting materials. One having no knowledge of this acid could hardly find it obvious to make any cephem using this acid as an acylating agent, much less the particular eephem recited in claim 6. In other words, it would not have been obvious to those of ordinary skill in the art to choose the particular acid of claim 6 as an acylating agent for the known amine for the simple reason that
the particular acid was unknown but for Ochiai’s disclosure in the ’429 application. As one of our predecessor courts had occasion to observe, in a case involving a highly analogous set of facts, “one cannot choose from the unknown.”
Mancy,
499 F.2d at 1293, 182 USPQ at 306.
In addition, although the prior art references the examiner discussed do indeed teach the use of various acids to make various cephems, they do not define a class of acids the knowledge of which would render obvious the use of Ochiai’s specifically claimed acid.
The Board noted that Ochiai’s specifically claimed acid is “similar” to the acids used in the prior art. Likewise, the examiner asserted that the claimed acid was
“slightly
different” from those taught in the cited references. Neither characterization, however, can establish the obviousness of the use of a starting material that is new and nonobvious, both in general and in the claimed process. The mere chemical possibility that one of those prior art acids could be modified such that its use would lead to the particular cephem recited in claim 6 does not make the process recited in claim 6 obvious “unless the prior art suggested the desirability of [such a] modification.”
In re Gordon,
733 F.2d 900, 902, 221 USPQ 1125, 1127 (Fed.Cir.1984). As we noted above, the examiner discussed no references containing any suggestion or motivation either (a) to modify known acids to obtain the particular one recited in claim 6, or (b) to obtain the particular new and nonobvious cephem produced by the process of claim 6. In short, the prior art contains nothing at all to support the conclusion that the particular process recited in claim 6 is obvious.
In light of the above, the examiner’s errors are evident. First, the examiner concluded that one of ordinary skill in the art would “recognize the use of ‘2-aminothiazolyr if the final products sought were to contain this moiety.” The prior art, however, contains nothing at all to suggest that one seek this concededly nonobvious final product. The examiner erred by indulging in an essentially hindsight comparison of the functioning of the new acid in claim 6 as a precursor to the claimed cephem with that of other acids in the prior art processes that produced other cephems. Such a comparison uses Ochiai’s specification as though it were prior art in order to make the claim to a method that uses the nonobvious acid to make the nonob-vious cephem appear to be obvious. Second, the examiner incorrectly drew
from Durden,
a case turning on specific facts, a general obviousness rule: namely, that a process claim is obvious if the prior art references disclose the same general process using “similar” starting materials.
No such
per se
rule exists. Mere citation
of Durden, Albertson,
or any other case as a basis for rejecting process claims that differ from the prior art by their use of different starting materials is improper, as it sidesteps the fact-intensive inquiry mandated by section 103. In other words, there are not
“Durden
obviousness rejections” or
“Albertson
obviousness rejections,” but rather only section 103 obviousness rejections.
The Board essentially repeated the examiner’s error of sidestepping the particularized inquiry required by section 103 by grounding
the rejection on the supposedly “controlling” effect of “[cjases such as
In re Larsen, In re Albertson,
and, particularly,
In re Durden,
all of which were directed to processes of making chemical compounds.”
Ochiai,
24 USPQ2d at 1267 (citations omitted). After categorizing the process recited in claim 6 as a “process of making” rather than as a “process for using,” the Board reached its conclusion according to the following syllogism: (a) “process of making” claims have led to rejections, as in
Larsen, Albertson,
and
Durden,
whereas “process for using” claims have led to allowances, as in
Kuehl, Mancy,
and
Ple-uddemann;
(b) Ochiai’s claim is directed to a “process of making”; (c) therefore, the rejection should be affirmed.
Id.
at 1268-70. This method of analysis is founded on legal error because it substitutes supposed
per se
rules for the particularized inquiry required by section 103. It necessarily produces erroneous results. Moreover, the Board indulged a non sequitur when it grounded its conclusion of obviousness on the assertion that the starting materials recited in claim 6 are “similar” to those of the prior art. The recited acid is nonobvious, having itself been patented based on the parent application. Nor did the Board justify its characterization of “similarity]” in any other manner. Similarity is, as we noted above, not necessarily obviousness.
The Alleged Conflict in Our Case Law
Both the Solicitor and Ochiai devote substantial portions of their briefs to purported demonstrations that our precedents on the obviousness
vel non
of chemical processes are, if not in conflict, at least in severe tension with one another and thus create unnecessary confusion. Both parties identify the same two sets of three cases as presenting the conflict:
Larsen, Albertson,
and
Dur-den,
upholding rejections on appeal, are said to be inconsistent with
Kuehl, Mancy,
and
Pleuddemann,
reversing rejections on appeal. While we agree that
some
generalized commentary found within several of these decisions may present minor tensions, both Ochiai and the Solicitor draw far too bleak a picture of the state of our case law. Other language in these cases, like their actual holdings, obviates any real inconsistency.
In
Albertson,
the court “reiterate[d] that all of the evidence must be considered on the ‘subject matter as a whole,’ from the viewpoint of one skilled in the art, in the determination of obviousness, and not simply the patentability of one of the starting reactants in a process.”
Albertson,
332 F.2d at 382, 141 USPQ at 732. Thus, the Board in this case looked to the general result in
Albertson
while ignoring the
Albertson
court’s explicit methodology. Every subsequent case that the parties discuss has been grounded on the same analytic principle: namely, that section 103 requires a fact-intensive comparison of the claimed process with the prior art rather than the mechanical application of one or another
per se
rule.
See Pleuddemann,
910 F.2d at 827, 15 USPQ2d at 1741 (“We repeat that the controlling law is in § 103 of the statute, which must be applied to the facts of this ease.”);
Durden,
763 F.2d at 1411, 226 USPQ at 362 (“Our function is to apply, in each ease, § 103 as written to the facts of disputed issues, not to generalize or make rules for other cases which are unforeseeable.”);
Mancy,
499 F.2d at 1292, 182 USPQ at 305 (“[T]he statutory standard of § 103 for determining obviousness of an invention is whether in view of the prior art the invention as a whole would have been obvious at the time it was made.”);
Kuehl,
475 F.2d at 665, 177 USPQ at 255 (“The test of unobviousness is a statutory test and requires comparison of the invention with the prior art in each case.... ”). As a consequence, these cases do not — indeed,
cannot
— present or create conflicting legal rules. They present, instead, applications of a unitary legal regime to different claims and fields of art to yield particularized results. It is thus surprising that the Board relies on
Durden
for a general rule when the
Durden
court expressly cautioned the bar “not to generalize or make rules for other cases.”
Because the regime of section 103, much like the Fourth Amendment proscriptions against “unreasonable” searches and warrants issued upon less than “probable cause,” mandates that legal outcomes turn on the close analysis of facts, reasonable persons may well disagree about the outcome of a given obviousness determination. These dis
agreements over the application of a legal rule can, however, be transformed into perceived “irreconcilable conflicts” between legal rules only when, as occurred here, examiners, members of the Board, and patent lawyers purport to find competing
per se
rules in our precedents and argue for rejection or allowance of a particular claim accordingly. We acknowledge that some generalized commentary found in these cases reviewing rejections of claims directed to chemical processes may, if viewed in isolation, have inadvertently provided encouragement to those who desire
per se
rules in this area. For example, one case includes an extensive discussion of the conceptual link between the obviousness
vel non
of a chemical composition and the obviousness
vel non
of a process for making the composition.
Such discussion, while entirely accurate, may have contributed to the erroneous view that one may determine the obviousness of a chemical process merely by determining whether it is a process for making a composition. As the cases noted above make clear, however, this is not and has never been the law of section 103. Indeed,
Durden,
the very case relied on by the examiner and the Board for a purported
per se
rule, clearly states that there are no such
per se
rules.
The use of
per se
rules, while undoubtedly less laborious than a searching comparison of the claimed invention — including all its limitations — with the teachings of the prior art, .flouts section 103 and the fundamental case law applying it.
Per se
rules that eliminate the need for fact-specific analysis of claims and prior art may be administratively convenient for PTO examiners and the Board. Indeed, they have been sanctioned by the Board as well. But reliance on
per se
rules of obviousness is legally incorrect and must cease. Any such administrative convenience is simply inconsistent with section 103, which, according to
Graham
and its progeny, entitles an applicant to issuance of an otherwise proper patent unless the PTO establishes that the invention
as claimed
in the application is obvious over cited prior art, based on the specific comparison of that prior art with claim limitations. We once again hold today that our precedents do not establish any
per se
rules of obviousness, just as those precedents themselves expressly declined to create such rules. Any conflicts as may be perceived to exist derive from an impermissible effort to extract
per se
rules from decisions that disavow precisely such extraction.
In sum, as we clearly indicated in
In re Dillon,
a recent in banc decision, “[w]hen any applicant properly presents and argues suitable method claims, they should be examined in light of all ... relevant factors, free from any presumed controlling effect of
Durden ”
or any other precedent. 919 F.2d 688, 695, 16 USPQ2d 1897, 1903 (Fed.Cir.1990) (in banc),
cert. denied,
500 U.S. 904, 111 S.Ct. 1682, 114 L.Ed.2d 77 (1991). Having compared Ochiai’s claims, limited as they are to the use of a particular nonobvious starting material for making a particular nonobvious end product, to the prior art of record, we reverse the rejection of claims 6 through 10 as an incorrect conclusion reached by incorrect methodology.
Reversed.