Sextant Avionique, S.A. appeals from the decision of the United States District Court for the Northern District of California granting summary judgment that Analog Devices, Inc. did not literally infringe any claim of Sextant's patents and granting judgment as a matter of law that Analog did not infringe under the doctrine of equivalents.1 We affirm.

A. Sextant's "Marcillat" and "Boura" Patents

Sextant's patents pertain generally to accelerometers, which are small devices capable of detecting acceleration and which find utility in avionic and automotive applications. The first of Sextant's patents at issue in this appeal, U.S. Patent 4,663,972, was issued to inventor Gerard Marcillat on March 6, 1985. One embodiment disclosed in the Marcillat patent shows a "test body" 6 connected to and supported by two "flexible blades" 4 and 5. The flexible blades are anchored at their "lower ends" to a "fixed part." The test body further comprises "lateral edges" 16 and 17 having metallized edges 18 and 19. The fixed part comprises two "wings" 20 and 21 having metallized edges 22 and 23. This configuration thus produces two capacitors, the first formed by plates 19 and 22, the second by plates 18 and 23, with air gaps in between the capacitor plates. This embodiment is shown in Figure 2 of Marcillat:

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

The accelerometer operates as follows: when the device is subject to an accelerative force in the "sensitive axis" (i.e., from left to right in Fig. 2), the test body sways like a pendulum within the fixed part and thus varies the distance between the capacitor plates. This variation corresponds electrically to a change in the capacitance of the capacitors, which can be detected by circuitry residing elsewhere on the fixed part. Specifically, the specification discloses that signals can travel from the metallized edges 18 and 19 of the test body to the detection circuitry by metallizing the edges of the flexible blades 4 and 5.

The specification also discloses that the accelerometer can be manufactured using semiconductor processing techniques. For example, the specification discloses that the claimed "pendular structure" can be "formed i[n] a single piece by micro-machining a crystalline silicon or quartz wafer which may further serve as [a] substrate for an integrated electronic circuit." Marcillat patent, col. 3, ll. 10-13.

Claim 2 of Marcillat, which is representative of the claims at issue, reads as follows:

a flat pendular structure made from one and the same crystalline wafer, said structure having in a same plane[ ] a flat fixed part, two parallel blades flexible in the said plane and delimiting therebetween a space, each of said blades having a first end portion fixedly connected to said fixed part, and a second end portion, said structure further comprising a flat test body connected to the second end portions of said blades so as to be suspended from the fixed part and to be able to move in translation in the said plane along a sensitive axis, said flat test body extending at least partially into said space,

wherein electrical connections between the test body and the fixed part of the pendular structure are formed by metallizations formed on the thin faces of said flexible blades.

Marcillat patent, col. 8, ll. 33-47 (paragraphing added).

The second of Sextant's patents, U.S. Patent 4,711,128, was issued to inventor Andre Boura on April 14, 1986. The Boura patent is not formally related to the Marcillat patent; it is not, for example, derived from any application related to Marcillat by continuation or division. However, Boura's written description makes clear that his invention constituted an improvement over Marcillat's invention. In fact, the "Field of the Invention" is described by Boura as "an accelerometer using an accelerometric sensor with flat pendular structure of the type described in U.S. Pat. 4,663,927 [i.e., Marcillat]." Boura patent, col. 1, ll. 8-10. This statement is followed by Boura's "Description of the Prior Art," which focuses on Marcillat's disclosure of a magnetic circuit capable of returning the test body to a proper rest position. The "aim" of Boura's invention is summarized as "to provide ... an accelerometric sensor of a type similar to the one described above [i.e., Marcillat's sensor]," id. at ll. 49-50, but which has an improved return mechanism. Specifically, Boura's mechanism involved applying certain optimal mathematically-derived voltages to the various capacitor plates of the sensor structure.

Claim 1, the only independent claim in the Boura patent, reads as follows:

a flat pendular structure made from one and the same crystalline wafer, said structure having in a same plane[ ] a flat fixed part, at least two parallel blades flexible in the same plane and delimiting therebetween a space, each of said blades having a first end portion fixedly connected to said fixed part, and a second end portion, said structure further comprising a flat test body connected to the second end portions of said blades so as to be suspended from the flat fixed part and to be able to move in translation in the same plane along a sensitive axis under the effect of an acceleration with a position which varies in relation with said acceleration, said flat test body extending at least partially into said space,

wherein said flat test body comprises at least a first edge which carries a first metallization having first and second opposite faces and said flat fixed part comprises at least a second and third edge carrying respectively a second and third metallizations, said second and third metallizations respectively facing said first and second opposite faces, so as to form capacitors whose capacities vary depending on the position of said test body, said first metallization being brought to a first voltage V sub0 , whereas the second and the third metallizations are respectively brought to a second and third voltages V sub1 and V sub2 which are capable of generating an electrostatic return force on the flat test body.

Id., col. 5, ll. 15-42 (paragraphing added).

B. The Prosecution Histories of Marcillat and Boura

The salient portions of the prosecution history of the Marcillat patent are easily summarized. Original application claim 7 generically claimed "capacitor plates." This claim was rejected under 35 U.S.C. § 103 over a prior art patent issued to Rudolf in light of other references. Rudolf discloses an accelerometer formed by semiconductor processing techniques in which the test body (a pivotable flap) is formed in monocrystalline silicon that has been doped with impurities to make it conductive. Similar to the Marcillat disclosure, the test body of Rudolf constitutes one of the capacitor plates of the sensor. In distinguishing over this art, Marcillat canceled claim 7 and substituted claim 26 in its place. Claim 26 was more explicit as to the nature of the capacitor plates, claiming the plates as "metallizations." Marcillat's amendment was accompanied by the following statement: "In Rudolf, the flap and attachment means comprise monocrystalline silicon heavily doped with boron and form a capacitor plate without metallizations...." Joint App. at A2224. Application claim 26 eventually issued as claim 14 in the Marcillat patent. The term "metallization" also appears in other claims of Marcillat to describe the composition of structures other than capacitor plates, such as the electrical connections claimed in Marcillat's claim 2, quoted above. In addition to the "metallization" limitation, the parties do not dispute that the claim limitations "made from one and the same crystalline wafer" and "flat fixed part" were added to the claims of Marcillat to distinguish them over the prior art.

As originally filed, the Boura application contained one independent claim, claim 1, which read as follows:

In an accelerometer using at least one accelerometric sensor with a flat pendular structure formed by micromachining a substrate formed of a fine monocrystal wafer and comprising a flat mobile mass suspended from the rest of the structure by means of two thin parallel strips situated on each said of said mass, said mobile mass comprises at least one mobile capacitor plate and the fixed part of the pendular element comprises at least two fixed capacitor plates disposed on each side of the mobile plate and the mobile plate is brought to a first voltage V sub0 , whereas the fixed plates are respectively brought to [ ] second and third voltages V sub1 and V sub2 which are capable of generating an electrostatic recall force F subR on the mobile mass.

Id. at 2281. This claim was rejected under 35 U.S.C. § 112, p 2 for indefiniteness and under 35 U.S.C. § 103 in light of a prior art patent issued to Deval (U.S. Patent 4,345,474). The statement accompanying the examiner's rejection read: "Although Deval provides a common ground potential between fixed plates ..., it would have been an obvious design modification to provide a common ground potential between [the] mobile capacitor plates ... and to vary the voltage to the fixed plates. Forming [the wafer in Deval] by micromachining a monocrystal would have been an obvious manufacturing expedient." Id. at A2302.

After this rejection was made in Boura, the examiner of the Marcillat application allowed the Marcillat claims. Accordingly, Sextant's attorney, the same attorney who was prosecuting the Marcillat application, canceled claim 1 in Boura and substituted claim 10 in its place, which eventually issued as claim 1 in the Boura patent, quoted above. Boura stated the following about claim 10:

[Deval] does not present a flat pendular structure but a structure comprising a flat test body hinged for rotation....

In contrast [ ] with the Applicant's sensor, [Deval's] capacity electrodes of the test body are formed by a thin layer of metal which covers the two sides of this test body and does not consist in [sic] metallizations formed on edge portions of this test body. In addition, [Deval's] capacity electrodes of the fixed structure cover the faces of the two side members which are facing the two sides of the test body. This is quite different from the Applicant's sensor in which these capacity electrodes are formed on the edge portions of the fixed flat part.

An important advantage of the Applicant's structure consists in that, due to the parallel flexible blades, ... the capacity electrodes remain substantially parallel [when the test body moves]. Therefore, a high precision can be obtained.

[T]he width of the facing metallizations which form the capacity electrodes [of Deval] is substantially equal to the thickness of the substrate which consists of a fine wafer. The surfaces of these facing metallizations are very small and it was not foreseeable that it would be possible to obtain a sufficient return force F subR (which is proportional to the area of the facing metallizations).

Id. at A2310-11. Thus, Boura argued that claim 10 was distinguishable from Deval because (1) Deval disclosed a flat rotational structure as opposed to the "flat pendular structure" of the claim, (2) Deval disclosed forming capacitors on the sides of the test body and flat fixed part instead of their "edges," (3) Deval did not disclose parallel flexible blades permitting the capacitor plates of the sensor to remain parallel, and (4) Deval did not enable the use of a return force. Other statements made by Boura concerning the "good results" that are obtained by his structure have no relation to distinguishing the prior art from claim 10 and are therefore not noted here. See id. at A2311-12. However, new claim 10 contained the three limitations that Marcillat had added to his claims in order to overcome the Rudolf reference.

Boura claim 1 is similar in many respects to Marcillat claim 2. In fact, comparison of Marcillat claim 2 and Boura claim 1 shows that, through amendment, the language of these claims became substantially identical except for the "wherein" clauses. Sextant does not dispute that use of this format was an attempt to secure a quick allowance of the Boura claims, even though the Marcillat claims had been allowed by a different examiner. Thus, through this strategy, the "metallization," "made from one and the same crystalline wafer," and "flat fixed part" limitations became part of the claims of Boura.

Sextant sued Analog for infringement, inter alia, of claim 2 of Marcillat and claim 1 of Boura. Analog's accused devices are fabricated using semiconductor processing techniques and use a deposited doped polysilicon layer for those structures corresponding to the test body and blades of the claims. Analog moved for summary judgment of noninfringement and for construction of certain disputed limitations of the claims. See Sextant Avionique, S.A. v. Analog Devices, Inc., No. C-95-2838 SI (N.D.Cal. Feb. 6, 1997) (order granting summary judgment). The court construed six limitations, four of them in favor of Analog's interpretation,2 and thereafter granted summary judgment of no literal infringement to Analog. Particularly relevant here, the court construed the term "metallization" to mean a deposited metallic material, and in so doing noted the relevance of the prosecution history in which Marcillat distinguished his claims over Rudolf's use of doping to achieve suitable conductivity. See id. at 14-15. Because Analog did not use a deposited metallic layer, but instead used a doped polycrystalline silicon layer, the court held that Analog's device did not literally infringe any claim of the Marcillat patent.

Analog later successfully moved for judgment as a matter of law, arguing that prosecution history estoppel barred Sextant from asserting that Analog's devices infringed either patent under the doctrine of equivalents. Concerning the Marcillat patent, the court concluded that

Marcillat added the "metallization" limitations to his patent to distinguish Rudolf and by so doing, he surrendered an accelerometer in which the test body and attachment means are conductive without the addition of an additional deposited metal layer. The Court furthermore finds that Analog's accused devices are conductive without the addition of a deposited metal layer and therefore fall within the scope of the "metallization" estoppel. Accordingly, Sextant is estopped from claiming infringement under the doctrine of equivalents as to the "metallization" element of the Marcillat patent claims.

Sextant Avionique, S.A. v. Analog Devices, Inc., No. C-95-2838 SI, slip op. at 9 (N.D.Cal. Sept. 29, 1997) (order granting judgment as a matter of law). The court also concluded that Marcillat's addition of the "made from one and the same crystalline wafer" and "flat fixed part" limitations also created estoppels which precluded Sextant's equivalence theories against Analog's accused devices. See id. at 9-13.

The court next addressed the impact of prosecution history estoppel on the Boura patent and concluded that the estoppels present in the Marcillat file history were applicable to limit the scope of equivalence of the same claim terms in the Boura patent. The court explained that

Analog argues that Boura's amendments to his claim language in response to the section 112 [rejection] should create an estoppel. The Court agrees with Analog that under the unusual facts of this case, prosecution history estoppel applies. See Pall Corp. v. Micron Separations, Inc., 66 F.3d 1211, 1220 (Fed.Cir.1995) ("[change made in order to more particularly point out the applicant's invention are] not presumed to raise an estoppel, but is reviewed on its facts, with the guidance of precedent.")

On having his claims rejected as indefinite by the Examiner, Boura incorporated the claim terms of the Marcillat patent which had recently been allowed and then argued that these claim terms addressed the Examiner's concerns. Of particular importance is the timing of Boura's amendments. The claims of the Marcillat patent had been allowed on October 22, 1986 after Marcillat had narrowed and redrafted his claims to the satisfaction of the Examiner. Boura incorporated the terms from the Marcillat patent (i.e. "metallization," "made from one and the same crystalline wafer," and "flat fixed part") into his own patent application on May 15, 1987. It appears that Boura's intent in adding these terms was to secure quick allowance of the Boura application and to avoid the objections raised during Marcillat's prosecution of his patent. Under these circumstances, the Court finds that Boura should be estopped from asserting infringement under the doctrine of equivalents against Analog as to these claim elements[.] Because Sextant is estopped from asserting infringement under the doctrine of equivalents as to the individual "metallization," "made from one and the same wafer," and "flat fixed part" limitations of the Boura claims, the court GRANTS judgment in favor of Analog on Sextant's doctrine of equivalents claim as regards the Boura patent.

Id. at 14-15 (brackets in quote from Pall in original).

Sextant appealed the noninfringement rulings to this court. We have jurisdiction pursuant to 28 U.S.C. § 1295(a)(1) (1994).

Judgment as a matter of law (JMOL) is appropriate when "a party has been fully heard on an issue and there is no legally sufficient evidentiary basis for a reasonable jury to find for that party on that issue." Fed.R.Civ.P. 50(a)(1). We review a district court's decision on a motion for JMOL de novo, reapplying the JMOL standard. See Markman v. Westview Instruments, Inc., 52 F.3d 967, 975, 34 USPQ2d 1321, 1326 (Fed.Cir.1995) (en banc), aff'd, 517 U.S. 370, 116 S. Ct. 1384, 134 L. Ed. 2d 577, 38 USPQ2d 1461 (1996). Summary judgment is appropriate when there are no genuine issues of material fact and the moving party is entitled to judgment as a matter of law. Fed.R.Civ.P. 56(c). We similarly review a district court's grant of summary judgment de novo, reapplying the summary judgment standard. See Conroy v. Reebok Int'l, Ltd., 14 F.3d 1570, 1575, 29 USPQ2d 1373, 1377 (Fed.Cir.1994).B. Claim Construction and Literal Infringement

"An infringement analysis entails two steps. The first step is determining the meaning and scope of the patent claims asserted to be infringed. The second step is comparing the properly construed claims to the device accused of infringing." Markman, 52 F.3d at 976, 34 USPQ2d at 1326. The first step, claim construction, is a question of law, which we review de novo. See Cybor Corp. v. FAS Techs., Inc., 138 F.3d 1448, 1456, 46 USPQ2d 1169, 1174 (Fed.Cir.1998) (en banc). The second step is factual. See North Am. Vaccine, Inc. v. American Cyanamid Co., 7 F.3d 1571, 1574, 28 USPQ2d 1333, 1335 (Fed.Cir.1993). When construing a claim, a court principally consults the evidence intrinsic to the patent, viz., the claims themselves, the written description portion of the specification, and the prosecution history. See Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1582-83, 39 USPQ2d 1573, 1576-77 (Fed.Cir.1996).

Sextant argues that the court erred in its construction of the four claim limitations that were construed against it, viz., "metallization," "one and the same crystalline wafer," "flat fixed part," and "two parallel blades." Because we conclude that the district court did not err in its construction of the term "metallization," and therefore that summary judgment of no literal infringement was properly granted to Analog, we need not address Sextant's arguments with respect to the latter three limitations.

Sextant's claim construction arguments hinge on an ironic contention: that the evidence intrinsic to its own patent is ambiguous and insufficient to construe the claims, thus requiring resort to extrinsic evidence such as expert testimony. We disagree that the intrinsic evidence is insufficient to construe the term "metallization." We start first with the claim language. See Vitronics, 90 F.3d at 1582, 39 USPQ2d at 1576. The term "metallization" certainly connotes the use of a metallic material. While Sextant was free to act as its own lexicographer and to define this term in a manner different from its plain meaning, see id. ("Although words in a claim are generally given their ordinary and customary meaning, a patentee may choose to be his own lexicographer and use terms in a manner different from their ordinary meaning, as long as the special definition of the term is clearly stated in the patent specification or file history."), it has not done so here. The Marcillat specification discloses only the deposition of metallic materials, see Marcillat patent, col. 8, ll. 1-8 ("The metal deposited on the pendular structure ... may be gold or a gold alloy such as a chromium-gold alloy for example. However, ... these circuits are preferably made from silver or aluminum ....") (emphasis added), and does not otherwise suggest that the term "metallization" can include conductive materials generally or materials made conductive by doping. In fact, the specifications of both Marcillat and Boura use the concepts of "metallization" and "conductors" in related, but somewhat different, contexts, suggesting that they do not have identical meanings. See, e.g., Marcillat patent, col. 7, ll. 3-5 ("this conducting strip being connected to a metallized one of the thin faces of the flexible blade"); Boura patent, col. 3, ll. 20-22 ("[T]he edges of the mobile teeth are covered by metallizations which extend the conducting layer at right angles.").

The prosecution history of the Marcillat patent, in which the applicant distinguished his "metallized" structure over the "doped silicon" structure of Rudolf, strengthens our conclusion that the term "metallization" denotes only metallic materials, not doped silicon materials. The extrinsic evidence proffered by Sextant cannot alter the construction indicated by the intrinsic evidence, thereby injecting an ambiguity where none exists. See Bell & Howell Document Management Prods. Co. v. Altek Sys., 132 F.3d 701, 706, 45 USPQ2d 1033, 1038 (Fed.Cir.1997) (holding that extrinsic evidence may not be relied upon during claim construction when the intrinsic evidence is sufficient to construe the claim). Accordingly, the district court did not err in its construction of the term "metallization," and it therefore did not err in granting summary judgment that Analog's device, which uses doped polysilicon, did not literally infringe any of the claims of the Marcillat or Boura patents.

As we recently restated en banc,

An accused device that does not literally infringe a claim may still infringe under the doctrine of equivalents if each limitation of the claim is met in the accused device either literally or equivalently. Prosecution history estoppel provides a legal limitation on the application of the doctrine of equivalents by excluding from the range of equivalents subject matter surrendered during prosecution of the application for the patent. The estoppel may arise from matter surrendered as a result of amendments to overcome patentability rejections, or as a result of argument to secure allowance of a claim. Prosecution history estoppel is a legal question subject to de novo review on appeal.

Cybor, 138 F.3d at 1459-60, 46 USPQ2d at 1177-78 (citations omitted).

Sextant makes several arguments concerning the district court's alleged "misapplication" of the law of prosecution history estoppel following the Supreme Court's decision in Warner-Jenkinson Co. v. Hilton Davis Chemical Co., 520 U.S. 17, 117 S. Ct. 1040, 137 L. Ed. 2d 146, 41 USPQ2d 1865 (1997). We have considered these arguments, but find them unpersuasive. Again, we focus exclusively on the "metallization" limitation because it is dispositive.

[w]hen determining whether prosecution history estoppel applies to limit the doctrine of equivalents, a court must examine the reason why an applicant amended a claim. See Warner-Jenkinson, 520 U.S. 17, 117 S.Ct. at 1049, 137 L. Ed. 2d 146, 41 USPQ2d at 1872. If such examination indicates that a patent applicant has made a substantive change to his claim that clearly responds to an examiner's rejection of that claim as unpatentable over prior art, prosecution history estoppel applies to that claim; only the question of the scope of the estoppel remains. No presumption needs to be applied in such a case because the reason for the amendment is clear.

Bai v. L & L Wings, Inc., 160 F.3d 1350, 1355, 48 USPQ2d 1674, 1677 (Fed.Cir.1998). In this case, original claim 7 of Marcillat was drawn broadly to capacitor plates. This claim was rejected over Rudolf, to which Marcillat responded by amending his claim to include the "metallization" limitation and by distinguishing Rudolf on the basis that it disclosed the use of doped silicon. Marcillat's amendment was thus clearly responsive to the examiner's prior art rejection. Under such circumstances, prosecution history estoppel plainly applies.

As to the scope of estoppel, Sextant contends that Analog's accused devices differ too substantially from the device disclosed in Rudolf for prosecution history estoppel to preclude a conclusion that Analog infringes the Marcillat patent under the doctrine of equivalents. Our case law recognizes that the scope of prosecution history estoppel "may have a limiting effect within a spectrum ranging from great to small to zero. The effect may or may not be fatal to application of a range of equivalents broad enough to encompass a particular accused product." Hughes Aircraft Co. v. United States, 717 F.2d 1351, 1363, 219 U.S.P.Q. (BNA) 473, 481 (Fed.Cir.1983). The scope of estoppel, i.e., what subject matter has been surrendered during prosecution by the patentee, is to be viewed from the vantage point of a reasonable competitor of the patentee, see Haynes Int'l v. Jessop Steel Co., 8 F.3d 1573, 1578, 28 USPQ2d 1652, 1656 (Fed.Cir.1993), on rehearing, 15 F.3d 1076, 29 USPQ2d 1958 (Fed.Cir.1994), and is determined with reference to the prior art and any amendments and/or arguments made in an attempt to distinguish such art, see Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 72 F.3d 857, 864, 37 USPQ2d 1161, 1165 (Fed.Cir.1995), vacated on other grounds, 520 U.S. 1111, 117 S. Ct. 1240, 137 L. Ed. 2d 323 (1997).

Here, it is clear that Sextant is estopped from successfully asserting that Analog's devices employing doped silicon are equivalent to the "metallization" of the claims. As noted above, Marcillat's "metallization" limitation was added in response to the examiner's citation of Rudolf, which disclosed the use of doped silicon. In addition to the amendment, Marcillat argued that its claim was distinguishable over Rudolf because Rudolf's capacitor plate "comprise[d] monocrystalline silicon heavily doped with boron ... without metallizations." Joint App. at A2224. The effect of Marcillat's amendment and argument was clearly to surrender silicon materials made conductive by doping. Analog's doped-polysilicon devices thus fall within the scope of the estoppel and therefore do not infringe the Marcillat patent under the doctrine of equivalents. Accordingly, the district court did not err in granting judgment to Analog as a matter of law on Sextant's claim that Analog infringed the Marcillat patent under the doctrine of equivalents.

Sextant next argues that the court's application of prosecution history estoppel was "demonstratively erroneous as to scope because it is apparent that Analog is not practicing the surrendered prior art." Sextant's Opening Brief at 39. We disagree. If Analog were practicing the prior art, it would have had a complete defense to Sextant's infringement charge, and prosecution history would be irrelevant. Case law has established that the prior art and prosecution history estoppel provide independent "policy oriented" limitations on the doctrine of equivalents. See Loctite Corp. v. Ultraseal Ltd., 781 F.2d 861, 870, 228 U.S.P.Q. (BNA) 90, 96 (Fed.Cir.1985), overruled on other grounds by Nobelpharma AB v. Implant Innovations, Inc., 141 F.3d 1059, 46 USPQ2d 1097 (Fed.Cir.1998) (en banc); see also Haynes Int'l, Inc. v. Jessop Steel Co., 8 F.3d 1573, 1579, 28 USPQ2d 1652, 1656-57 (Fed.Cir.1993), on rehearing, 15 F.3d 1076, 29 USPQ2d 1958 (Fed.Cir.1994). Therefore, the fact that Sextant is not practicing the prior art is not fatal to the application of prosecution history estoppel.

Sextant contends that the district court had no basis to hold that prosecution history estoppel applies to limit the range of equivalents otherwise affordable to the term "metallization" in the claims of Boura, and it makes two arguments in support of that contention. Sextant first argues that Boura amended claim 1 to overcome the examiner's indefiniteness rejection under 35 U.S.C. § 112, p 2, which Sextant characterizes as the "principle rejection" pending at that time. Sextant cites the Supreme Court's opinion in Warner-Jenkinson to support its contention that amendments made in response to § 112 rejections do not give rise to prosecution history estoppel. Sextant next argues that the court erred in "incorporating" the file history of Marcillat into the file history of Boura, thereby raising in Boura the estoppel that was present in Marcillat. Sextant contends that such action was improper because the two applications were not formally related, and therefore a reasonable competitor desiring to assess the claim scope of Boura would not investigate the file history of Marcillat.

The Supreme Court in Warner-Jenkinson held that whether prosecution history estoppel applied to limit the doctrine of equivalents depended on the reason disclosed in the prosecution history for the amendment to the claims. See Warner-Jenkinson, 520 U.S. at 30, 117 S.Ct. at 1049, 137 L. Ed. 2d 146, 41 USPQ2d at 1871 ("petitioner reaches too far in arguing that the reason for an amendment during patent prosecution is irrelevant to any subsequent estoppel."). If the claims were amended for a reason "related to patentability," prosecution history estoppel applies. See 520 U.S. at 29-35, 117 S.Ct. at 1049-51, 137 L. Ed. 2d 146, 41 USPQ2d at 1871-73; Bai, 160 F.3d at 1355, 48 USPQ2d at 1677. On the other hand, if the claims were amended for a reason that was not "related to patentability," prosecution history estoppel does not apply absent a "clear and unmistakable surrender" of certain subject matter. See Warner-Jenkinson, 520 U.S. at 31-33, 117 S.Ct. at 1050, 137 L. Ed. 2d 146, 41 USPQ2d at 1872; Litton Sys., Inc. v. Honeywell, Inc., 140 F.3d 1449, 1458, 46 USPQ2d 1321, 1327 (Fed.Cir.1998).3 Finally, if the patent prosecution record does not disclose the reason for an amendment, a court must presume that the amendment was made for purposes of patentability and that prosecution history estoppel applies, and provide the patentee with an opportunity to rebut that presumption. See Warner-Jenkinson, 520 U.S. at 33-35, 117 S.Ct. at 1051, 137 L. Ed. 2d 146, 41 USPQ2d at 1873.4

The Court did not resolve the issue whether an amendment made in response to a § 112 rejection is an amendment made for "reasons of patentability" that gives rise to prosecution history estoppel. See Bai, 160 F.3d at 1355, 48 USPQ2d at 1678 (noting that the "meaning of the word 'patentability' was not made clear [by the Supreme Court in Warner-Jenkinson ], but the context of the court's discussion was patentability over the prior art."); Litton, 140 F.3d at 1457, 46 USPQ2d at 1327 (noting that " 'related to patentability' encompassed amendments 'made to avoid the prior art.' ").

On the one hand, it is clear that a patent applicant must comply with § 112 in order to obtain a patent; failure to meet the requirements of that section results in the denial of a patent, and a basis for denial of a patent is a form of "patentability" requirement. On the other hand, the patent statute uses the title "Conditions for patentability" as the heading for the novelty and nonobviousness provisions, see 35 U.S.C. §§ 102, 103, whereas § 112 only refers to the requirements of the specification. See also and compare 35 U.S.C. ch. 10, §§ 101-105 (1994) (entitled "Patentability of Inventions") with id. ch. 11, §§ 111122 (entitled "Application for a Patent"); id. § 282(2) & (3) (drawing a distinction between invalidity based on "any ground specified in part II [§§ 100-212] of this title as a condition for patentability " and "failure to comply with any requirement of sections 112 or 251 of this title.") (emphasis added). Thus, there are arguments on both sides of the issue.

In any event, we need not resolve this issue because it is not necessary to resolve this case. Although neither the parties nor the district court emphasized that the § 112 rejection in Boura was combined with a § 103 rejection over the Deval reference, that was the fact, and Sextant's contention that Boura amended his claims to "overcome a § 112 rejection, not a § 103 rejection," Sextant's Reply Brief at 7, is belied by the prosecution history. First, the addition of the "metallization" limitation was not responsive to the examiner's § 112, p 2 rejection and cannot be said to have been added for purposes of clarity or precision of the claim language. The § 112 rejection was accompanied by the following remarks from the examiner: "Claim 1 would appear to be an improvement type claim, but it is not clear what the improvement is. See 37 C.F.R. § 1.75(e). Consistent terminology should be used. For example, line 2 recites a 'pendular structure,' whereas lines 7-8 recite a 'pendular element.' In line 13, 'recall' should be --return--." Joint App. at A2301. Second, Boura's remarks accompanying his amendment were directed to distinguishing the claims over Deval, albeit in respects not relevant to the "metallization" limitation. Thus, we are not presented with the situation in which an applicant has amended his claims for the purpose of overcoming a § 112, p 2 rejection. Cf. Litton, 140 F.3d at 1461, 46 USPQ2d at 1329-30.

Instead, this case involves the situation in which the prosecution history of Boura does not disclose the reason for the addition of the "metallization" limitation. Like the 6.0 pH limitation in Warner-Jenkinson, addition of this limitation was not necessary to overcome the prior art rejection based on Deval that was pending at the time of the amendment. Limiting the claims of Boura to capacitor plates comprising "metallizations" could not have distinguished over Deval because Deval disclosed this exact arrangement: "It is necessary to form capacitor electrodes on the confronting faces of pendulum 11 and stator 10 (if the latter is insulating). All electrodes may be formed as a thin metal layer. The layer may be obtained by evaporation and deposition under vacuum for example." Deval '474 patent, col. 2, ll. 41-45 (emphasis added). Furthermore, and not surprisingly, Boura did not attempt to argue that the claims were distinguishable from Deval on the basis of the inclusion of the term "metallization" in the claims. Boura instead argued that his claim as amended was distinguishable from Deval on the basis of the location of the metallization, i.e., because Deval disclosed capacity electrodes formed on the sides of the structure, while the claim as amended claimed an edge-wise configuration. See Joint App. at A2310 ("In contrast [ ] with the Applicant's sensor, [Deval's] capacity electrodes of the test body are formed by a thin layer of metal which covers the two sides of this test body and does not consist in [sic] metallizations formed on edge portions of this test body."). Thus, while the prosecution history discloses the reason for the addition of the "edge" limitation, it does not disclose the reason that the term "metallization" was added to the claims of Boura. Accordingly, we must, under Warner-Jenkinson, rebuttably presume that this limitation was added for a reason "related to patentability" and that prosecution history estoppel applies to limit the range of equivalents that might otherwise be affordable to the "metallization" limitation.

Sextant has not rebutted this presumption. As we have already noted, it is Sextant's contention that Boura amended his claims to overcome a § 112 rejection, a reason which Sextant opines is not "related to patentability." However, we have already dismissed this contention as unsupported by and even contrary to the prosecution history. Because Sextant has not established that Boura did not amend his claims for reasons unrelated to patentability, we conclude that the presumption has not been rebutted and that prosecution history estoppel applies to the "metallization" limitation.5

Having concluded that prosecution history estoppel applies under operation of the Warner-Jenkinson presumption, we are left with the question of the scope of that estoppel. See Bai, 160 F.3d at 1355, 48 USPQ2d at 1677 (after a court finds that prosecution history estoppel applies, "only the question of the scope of the estoppel remains."). Because none of our cases prior to the instant case has applied the Warner-Jenkinson presumption or addressed the scope of any estoppel arising therefrom, we consult the Supreme Court's discussion of the presumption for guidance:

We are left with the problem, however, of what to do in a case like the one at bar, where the record seems not to reveal the reason for including the lower pH limit of 6.0. In our view, holding that certain reasons for a claim amendment may avoid the application of prosecution history estoppel is not tantamount to holding that the absence of a reason for an amendment may similarly avoid such an estoppel. Mindful that claims do indeed serve both a definitional and a notice function, we think the better rule is to place the burden on the patent-holder to establish the reason for an amendment required during patent prosecution. The court then would decide whether that reason is sufficient to overcome prosecution history estoppel as a bar to application of the doctrine of equivalents to the element added by that amendment. Where no explanation is established, however, the court should presume that the PTO had a substantial reason related to patentability for including the limiting element added by amendment. In those circumstances, prosecution history estoppel would bar the application of the doctrine equivalents as to that element. The presumption we have described, one subject to rebuttal if an appropriate reason for a required amendment is established, gives proper deference to the role of claims in defining an invention and providing public notice, and to the primacy of the PTO in ensuring that the claims allowed cover only subject matter that is properly patentable in a proffered patent application. Applied in this fashion, prosecution history estoppel places reasonable limits on the doctrine of equivalents, and further insulates the doctrine from any feared conflict with the Patent Act.

Warner-Jenkinson, 520 U.S. at 33-34, 117 S.Ct. at 1051, 137 L. Ed. 2d 146, 41 USPQ2d at 1873 (emphasis added).

Our cases have addressed the meaning to be attributed to the Court's statement that prosecution history estoppel "bars" any equivalency with respect to the limitation. See generally Litton Sys., Inc. v. Honeywell, Inc., 145 F.3d 1472, 47 USPQ2d 1106 (Fed.Cir.1998) (denial of suggestion for rehearing en banc) (Plager, Clevenger, and Gajarsa, JJ., dissenting); Hughes Aircraft Co. v. United States, 148 F.3d 1384, 47 USPQ2d 1542 (Fed.Cir.1998) (denial of suggestion for rehearing en banc) (Clevenger and Gajarsa, JJ., dissenting). In Litton, a case dealing not with the Warner-Jenkinson presumption but with an amendment clearly made for reasons related to patentability, we rejected the accused infringer's argument that "any claim language amended during prosecution for reasons related to patentability has no range of equivalents." Litton, 140 F.3d at 1456, 46 USPQ2d at 1326 (emphasis added). We reasoned that the Supreme Court had endeavored "to not change so substantially the rules of the game" concerning the scope of estoppel, id. at 1457, 46 USPQ2d at 1326 (brackets and quotations omitted), and that Supreme Court case law predating Warner-Jenkinson embraced the concept of scope in the context of amendments made to avoid prior art, see id. at 1457, 46 USPQ2d at 1326 (citing Keystone Driller Co. v. Northwest Eng'g Corp., 294 U.S. 42, 47-48, 55 S. Ct. 262, 79 L. Ed. 747 (1935) and Smith v. Magic City Kennel Club, 282 U.S. 784, 788-790, 51 S. Ct. 291, 75 L. Ed. 707 (1931)). Thus, we concluded for amendments made for reasons related to patentability that "[a] careful reading of the Supreme Court's opinion in context shows that Warner-Jenkinson did not effect a change in the scope of subject matter precluded by an estoppel, but only in the circumstances that may trigger an estoppel." Id., 140 F.3d 1449, 46 USPQ2d at 1327.

However, a similarly careful reading shows that the Supreme Court announced a different rule of scope to be applied when a presumption arises. The Court's language is unequivocal: "In those circumstances [i.e., when the Warner-Jenkinson presumption applies], prosecution history estoppel would bar the application of the doctrine of equivalents as to that element. " Warner-Jenkinson, 520 U.S. at 33, 117 S.Ct. at 1051, 137 L. Ed. 2d 146, 41 USPQ2d at 1873 (emphasis added). This language was not used by the Court in its foregoing discussion of the effect of amendments clearly made for reasons "related to patentability," i.e., for the purpose of distinguishing the claims over the prior art. See Litton, 140 F.3d at 1456-58, 46 USPQ2d at 1326-27 (rejecting the notion that prosecution history arising from an amendment made for reasons related to patentability "bars" the application of the doctrine of equivalents as to the amended limitation). Instead, that language was confined to the discussion of the operation of the presumption, suggesting that it is only operative in that circumstance.

This reading of Warner-Jenkinson is strengthened by consideration of the policies that the Court desired to foster in that case. Central to the Court's rationale for creating the presumption was the idea that "claims ... serve both a definitional and public notice function" and that the public should be able to assess the literal scope of claims and effective extension of claim scope under the doctrine of equivalents with reasonable certainty. See Warner-Jenkinson, 520 U.S. at 33-35, 117 S.Ct. at 1051, 137 L. Ed. 2d 146, 41 USPQ2d at 1873.6 This policy of "public notice" is heavily implicated in the circumstance in which the presumption is operative because, by definition, it is unclear to one reading the prosecution history why a particular amendment was made. Accordingly, the Court, through the operation of the presumption, placed the burden on the patentee to clarify his reasons. The patentee's failure to so clarify and thereby rebut the presumption should not work to the detriment of the public by allowing an uncertain range of equivalency to remain as to the limitation at issue.

Moreover, and as a practical matter, anyone desiring to assess the possible scope of an estoppel arising from the operation of the Warner-Jenkinson presumption faces a difficult task. As we noted previously in our discussion of the Marcillat patent, the scope of prosecution history estoppel has traditionally been determined with reference to the prior art and any amendments and/or arguments made in an attempt to distinguish such art. See page 15, supra. However, in the circumstance in which a presumption is operative, because the record is unclear concerning the reason for an amendment, it is unlikely that the prior art or the patentee's arguments, if any, will be of assistance in an estoppel analysis. The lack of relevant information concerning the scope of surrendered subject matter is further exacerbated if the patentee does not provide clarification during rebuttal. Indeed, this case is an example of the futility of a scope analysis in the circumstance in which the presumption is operative. Unguided by the prosecution history, the prior art, applicant's argument during prosecution, and sufficient evidence in rebuttal to the presumption, we have no way to set reasonable limits on how far beyond the literal scope of the term "metallization" the estoppel will allow the doctrine of equivalents to reach. We can expect reasonable competitors assessing the file history of Boura to be equally puzzled as to the scope of any potential estoppel. Thus, it is logical and fair that prosecution history estoppel arising from the operation of the Warner-Jenkinson presumption allows the doctrine of equivalents no room to operate. Finding the Supreme Court's language clear, we hold that in circumstances in which the Warner-Jenkinson presumption is applicable, i.e., where the reason for an amendment is unclear from an analysis of the prosecution history record, and unrebutted by the patentee, the prosecution history estoppel arising therefrom is total and completely "bars" the application of the doctrine of equivalents as to the amended limitation. The district court thus correctly rendered judgment that Analog was not liable to Sextant under the doctrine of equivalents because Sextant could not show equivalency to the "metallization" limitation.7 Because this estoppel bars infringement under the doctrine of equivalents, it is unnecessary for us to address the scope of any other possible estoppels applicable to Boura.

We have considered the parties' remaining arguments, but find them either without merit or unnecessary to resolve given our disposition of the case.

The district court did not err in its construction of the "metallization" limitation, which appears in the claims of both the Marcillat and Boura patents. The court did not err in granting summary judgment that Analog's devices do not literally infringe the claims of either of these patents. Moreover, the court did not err in concluding that prosecution history estoppel precluded Sextant from asserting that Analog's use of doped silicon is equivalent to the "metallization" limitation of the Marcillat claims. Finally, the court's judgment that Analog does not infringe under the doctrine of equivalents is affirmed because the unrebutted operation of the Warner-Jenkinson presumption concerning the "metallization" limitation bars the operation of the doctrine of equivalents as to that element. The judgment of the district court is therefore

AFFIRMED.

SMITH, Senior Circuit Judge, dissenting in part.

I agree with the majority's analysis of the district court's claim construction and I agree that the district court properly granted summary judgment that Analog's device did not literally infringe the claims of either the Marcillat or Boura patents. I also agree with the majority that Sextant is estopped by the prosecution history of the Marcillat patent from asserting that Analog's device infringes the claims of that patent under the doctrine of equivalents. I dissent, however, from the majority's holding that the Warner-Jenkinson presumption estops Sextant from relying on the doctrine of equivalents with respect to the Boura patent.

The majority affirms the district court's summary judgment that the Boura patent is not infringed under the doctrine of equivalents based on the following reasoning. (1) Boura did not disclose the reason for adding the "metallization" limitation and therefore, under Warner-Jenkinson, we must presume that the limitation was added for reasons related to patentability. (2) Sextant has not rebutted this presumption. (3) Therefore, the Warner-Jenkinson presumption applies and the application of the doctrine of equivalents is barred, so no equivalents of "metallizations" are covered by the Boura patent's claims.

I disagree that the Warner-Jenkinson presumption applies here. The Boura prosecution history clearly shows the reason the "metallization" limitation was added: Boura distinguished his invention from the Deval prior art patent by requiring that the metallizations be on the edge, not the side. In his response to the first Office action, Boura replaced original, rejected claim 1 with new claim 10 containing the "metallization" limitation. Boura distinguished the newly claimed invention from that of the Deval reference as follows: "In contrast [ ] with the Applicant's sensor, in this cited patent, the capacity electrodes of the test body are formed by a thin layer of metal which covers the two sides of this test body and does not consist in metallizations formed on edge portions of this body. This is quite different from the Applicant's sensor in which these capacity electrodes are formed on edge portions of the flat fixed part" (emphasis added).

The quoted passage makes it clear that Boura added the "metallization" limitation in order to contrast his invention with, and distinguish it from, that disclosed by Deval, not by requiring metallizations per se but by requiring that the metallizations be on the edge of the test body and not on the side.1 Therefore, the record shows that the "metallization" limitation was added for reasons related to patentability and the Warner-Jenkinson presumption should not apply.

The majority acknowledges that Boura added the "metallization-on-the-edge" limitation in response to a rejection over Deval, but splits the limitation into two parts: "metallization" and "on-the-edge." Since Deval discloses capacitors formed from metallizations, the majority reasons, that part of the limitation could not have been added in order to distinguish the invention from the prior art. Since the record does not disclose any other reason for adding the "metallization" limitation, the majority applies the Warner-Jenkinson presumption, finds it unrebutted, and bars application of the doctrine of equivalents.

Our cases applying the Warner-Jenkinson presumption do not support such a delicate parsing of claim limitations. For example, in the recent case of Bai v. L & L Wings, Inc., 160 F.3d 1350 (Fed.Cir.1998) , we held that the Warner-Jenkinson presumption did not apply. In Bai, the applicant claimed a target game including a fabric-covered, dish-shaped glove. Original claims 1-3 were rejected over a combination of four prior art patents, including one (Hartel) that disclosed a dish-shaped structure. In response, Bai amended claim 1 to include all the limitations of original claims 1-3, along with two additional limitations: that the glove be "hemispherical" and that it include a "strap mounted on the rear side." Id. at 1352 and n. 1. The amended claim was allowed.

The district court concluded that Bai had added the "hemispherical" limitation in response to the obviousness rejection, despite Bai's argument that the limitation was added merely to clarify the claimed structure and not to distinguish over the prior art. Id. at 1353, 1354. On appeal, we agreed that the amendment "was clearly an attempt by Bai to differentiate his invention over the Hartel patent." Id. at 1355.

We reached this conclusion despite the fact that Bai had added the "strap" limitation at the same time as the "hemispherical" limitation. The Bai court never considered whether the "hemispherical" limitation was necessary to distinguish the prior art in view of the simultaneous "strap" limitation, and in fact rejected Bai's argument that "the amendment could not have been made to overcome prior art because it was not necessary to overcome the prior art rejection." Id. at 1356. The Bai court held that Bai was estopped from raising the issue of the necessity of the "hemispherical" limitation. "Whether or not the 'hemispherical' limitation was superfluous, the prosecution record clearly reveals that Bai made this amendment to overcome prior art." Id., citing American Permahedge v. Barcana, 105 F.3d 1441, 1446 ("[C]lear assertions made during prosecution in support of patentability, whether or not actually required to secure allowance, may create an estoppel.").

The majority in this case, by contrast, reasons that because the "metallization" limitation was disclosed in the prior art, it could not have been necessary to distinguish Boura's invention and therefore was not added for a reason related to patentability. The Bai court rejected the very logic relied on by the majority in this case. Panel decisions are binding on future cases, unless overturned by the court en banc. See UMC Elecs. Co. v. United States, 816 F.2d 647, 652 n. 6 (Fed.Cir.1987) ("A panel of this court is bound by prior precedential decisions unless and until overturned en banc."). The majority's approach today is foreclosed by Bai.2

As a colleague so eloquently stated, "[w]e require practitioners who appear before this court to apply rigorous logic and consistency in their advocacy; we should require no less of ourselves whenever we are providing directions to the bar." Litton Systems, Inc. v. Honeywell, Inc., 145 F.3d 1472, 1478 (Fed.Cir.1998) (Gajarsa, J., dissenting from denial of rehearing en banc). The inconsistency engendered by the majority's approach in this case does a disservice to the development of our jurisprudence.

In addition, I am uneasy with application of the Warner-Jenkinson presumption for the first time on appeal, because I do not believe that Sextant has had an adequate opportunity to rebut the presumption. This court should not lightly conclude that the record presents no reason for a claim amendment, long after the parties have exhausted their opportunities to present evidence and arguments. This is especially so when that conclusion requires the court to limit patent claims to their literal terms. Here, the district court concluded that the record did in fact establish the reason for the amendment in the Boura patent--"to secure quick allowance of the Boura application and to avoid the objections raised during Marcillat's prosecution of his patent." Thus, the court never applied the Warner-Jenkinson presumption to the Boura patent's claims and never presented Sextant with an opportunity to rebut it.3 To apply the presumption first in this court, with the effect of completely barring the doctrine of equivalents, is unfair to litigants and adds to the unpredictability of patent litigation. From this point on, litigants will have to worry about the Warner-Jenkinson presumption being applied for the first time on appeal, and their claims being limited to their literal terms, because this court disagrees with the trial court on the reason a claim was amended. This development cannot be welcomed by anyone.

Regardless of the propriety of applying the Warner-Jenkinson presumption for the first time on appeal, the presumption should not even apply in this case, since the Boura prosecution history shows that the "metallization (on the edge, not the side)" limitation was added to distinguish the claimed accelerometer from the Deval reference. Since the presumption does not apply, there was no burden on Sextant to rebut it and no bar to application of the doctrine of equivalents.

I also disagree with the district court's treatment of the Boura patent. The district court held that the Marcillat patent's prosecution history applied to the Boura patent's claims, even though the Marcillat and Boura applications were not related by continuation or division and shared no common inventors, because the Boura claims were amended to correspond to the language of the Marcillat claims after the Marcillat claims were indicated allowable. The court found that the timing of the amendment and the similarity of the claim language indicated that Boura added the terms "to secure quick allowance of the Boura application and to avoid the objections raised during Marcillat's prosecution of his patent." The court concluded that the Marcillat patent's prosecution history created an estoppel with respect to the Boura patent's claims.

We have recently confirmed that "[a]s a basic proposition, the standard for determining whether subject matter has been relinquished is whether one of ordinary skill in the art would objectively conclude from the prosecution history that an applicant surrendered it." Litton Systems, Inc. v. Honeywell, Inc., 140 F.3d 1449, 1462 (Fed.Cir.1998) (citing Mark I Mktg. Corp. v. R.R. Donnelley & Sons Co., 66 F.3d 285, 291 (Fed.Cir.1995)). See also Desper Products, Inc. v. QSound Labs, Inc., 157 F.3d 1325, 1338 (Fed.Cir.1998) ("In determining the scope of what, if any, subject matter has been surrendered, the standard is an objective one: what would a reasonable competitor reading the prosecution history conclude has been surrendered."); Wang Labs., Inc. v. Mitsubishi Elecs. America, Inc., 103 F.3d 1571, 1578 (Fed.Cir.1997) ("We examine the statements and actions of the patentee before the PTO during prosecution and ask what a competitor reasonably may conclude the patentee surrendered to gain issuance of the patent." (citation omitted)).

We have found a basis for prosecution history estoppel in statements made by patent applicants in a variety of contexts during prosecution. See, e.g., Haynes Int'l, Inc. v. Jessop Steel Co., 8 F.3d 1573, 1579 (Fed.Cir.1993) and cases cited therein. But we have never based a finding of prosecution history estoppel on a statement made by a different applicant during prosecution of an unrelated application, nor have we ever hinted that competitors should examine the file histories of applications other than the one in suit in order to determine the scope of equivalents that may be accorded the patentee. The district court's importation of the Marcillat patent's prosecution history into the Boura patent was erroneous and should be reversed.

The district court's holding that the prosecution history of the Marcillat patent estops Sextant from asserting the doctrine of equivalents in the Boura patent is contrary to our precedent and should be reversed. In addition, the majority's holding that the "metallization" part of Boura's "metallization-on-the-edge" limitation is subject to the Warner-Jenkinson presumption is contrary to our precedent and to the facts of this case. I would reverse the district court's holding with respect to the Boura patent and remand for further proceedings to determine whether Analog's device infringes the claims of the Boura patent under the doctrine of equivalents, as those equivalents are limited by the prosecution history of the Boura patent and our controlling precedents. I therefore dissent from that part of the panel's disposition of this case.