Opinion for the court filed by Circuit Judge RADER. Opinion concurring in part and dissenting in part filed by Circuit Judge PAULINE NEWMAN.
RADER, Circuit Judge.The United States Supreme Court vacated the judgment in Litton Systems, Inc. v. Honeywell, Inc., 87 F.3d 1559, 39 USPQ2d 1321 (Fed.Cir.1996) (Litton I), and remanded to this court for further consideration in light of Warner-Jenkinson Co. v. Hilton Davis Chemical Co., 520 U.S. 17, 117 S.Ct. 1040, 137 L.Ed.2d 146 (1997). See Honeywell, Inc. v. Litton Sys., Inc., — U.S.-, 117 S.Ct. 1240, 137 L.Ed.2d 323 (1997). After applying the principles of Warner-Jenkinson, this court affirms-in-part, reverses-in-part, vacates-in-part, and remands.
Effect of Vacatur
This court first considered an appeal by Litton, Inc. (Litton), from a decision of the United States District Court for the Central District of California. After a jury verdict in favor of Litton, the trial court granted-in-part Honeywell, Inc.’s (Honeywell’s) motion for judgment as a matter of law (JMOL). The district court determined that: (1) Honeywell’s hollow cathode process did not literally infringe Litton’s U.S. Patent No. Re. 32,849 (the ’849 reissue), but upheld the jury’s verdict of infringement under the doctrine of equivalents; (2) Honeywell’s radio frequency (RF) ion beam process did not infringe, either literally or under the doctrine of equivalents, the ’849 reissue; (3) the ’849 reissue was invalid for obviousness; (4) Litton did not prove its claims of intentional interference with contractual relations and intentional interference with prospective economic advantage under state law; (5) inequitable conduct precluded enforcement of the ’849 reissue; and (6) the doctrine of intervening rights applied. See Litton Sys., Inc. v. Honeywell, Inc., 1995 WL 366468 (C.D.Cal.1995). In the alternative, the trial court granted Honeywell’s motion for a new trial on damages, but not liability. See id.
In Litton I, this court reversed the trial court’s grant of JMOL. This court determined that: (1) substantial evidence supported the jury’s findings that Honeywell’s hollow cathode and RF processes infringed the ’849 reissue; (2) the ’849 reissue was neither invalid for obviousness nor unenforceable for inequitable conduct; and (3) substantial evidence supported the jury’s finding of Honeywell’s liability under both state law claims. See Litton I, 87 F.3d at 1559. Litton I also affirmed the trial court’s grant of a new trial on damages and its order on intervening rights. See id.
*1453Litton’s patents claim a method for coating a substrate with multiple layers of optical materials. The method uses an ion beam from a Kaufinan-type ion beam source to sputter deposit the optical materials on the substrate. The result is an almost perfectly reflective mirror, an essential component for ring-laser gyroscopes (RLGs). RLGs control navigation in aircraft. Honeywell’s alleged infringing methods for making these mirrors use ion beams from hollow cathode and RF ion beam sources. Litton I, 87 F.3d at 1563-66, gives a more detailed discussion of the factual background of the case.
Because Warner-Jenkinson addressed the doctrine of equivalents, particularly the application of prosecution history estoppel, this court reinstates, without further analysis, the parts of Litton I: (1) reversing the trial court’s grant of JMOL with respect to the invalidity of the ’849 reissue on grounds of obviousness; (2) reversing the determination of unenforceability for inequitable conduct; (3) affirming the intervening rights order; and (4) affirming the grant of a new trial on damages (if liability is found on remand for any or all of the federal and state law claims). The Supreme Court’s opinion in Wamer-Jenkinson does not affect these aspects of Litton I. See Syntex Ophthalmics, Inc. v. Novicky, 767 F.2d 901, 902, 226 USPQ 952, 952 (Fed.Cir.1985).
This court applies the doctrines of Warner-]enkinson to the remaining issues. The standard of review with respect to the remaining issues remains unchanged. See, e.g., Martin v. Telectronics Pacing Sys., Inc., 105 F.3d 1090, 1092 (6th Cir.1997), cert. denied, — U.S. -, 118 S.Ct. 850, 139 L.Ed.2d 751 (1998).
Claim Construction
In this case, the question of infringement turns primarily on the interpretation of the phrase “Kaufman-type ion beam source” in claim 1 of the '849 reissue. Litton contends that the appropriate interpretation of “Kaufinan-type ion beam source” encompasses any broad-beam, multi-apertured, gridded ion beam source. Litton’s proposed construction, however, is inconsistent with the prosecution history of the ’849 reissue. See York Prods., Inc. v. Central Tractor Farm & Family Ctr., 99 F.3d 1568, 1575, 40 USPQ2d 1619, 1624 (Fed.Cir.1996) (“In a literal infringement analysis, prosecution history is relevant to claim coverage.”); see also Howes v. Medical Components, Inc., 814 F.2d 638, 645-46, 2 USPQ2d 1271, 1275-76 (Fed.Cir.1987).
In the course of prosecuting the ’849 reissue, Litton argued that the term “ion beam source” in its original claims could not “properly be construed to refer to any other ion beam gun but the Kaufman gun.” Paper No. 15 at 8. Moreover, a declaration accompanying Litton’s remarks plainly stated: “Those skilled in the coating arts ... would reasonably construe these claims to refer ... only to the Kaufinan-type ion-beam guns referred to in the specification of this application.” Paper No. 16 at 7. Thus, Litton defined “ion beam source” to mean only the Kaufinantype gun. This definition acquires even more credibility when Litton later amended its claims to cover a “Kaufinan-type ion beam source.” If, as Litton insisted, one of skill could only construe the broad term to mean a Kaufinan-type gun, certainly the specific term encompasses nothing more.
At column 4, lines 44-57, the reissue’s specification describes a Kaufinan-type ion beam source:
The ion beam gu[n] 4 is a commercially available ion [e]mitting ap[p]aratus generally known in the art as a Kauffman [sic] type ion beam gun. The gun’s cathode 6 is a therm[i]onic emitter, i.e., it emits electrons by passing an electric current through it which heats the wire. The cathode 6 emits electrons which are accelerated towards the anode 8. The electrons being accelerated from the cathode to the anode strike argon atoms and in so doing dislodge electrons from the argon. The results are positively charged argon ions which are accelerated away from the anode and towards the grids 12 and 14. Perma*1454nent bar magnets 10 attached to the anode introduce a magnetic field into the area between the cathode and the anode....
(Emphasis added.) Thus, this court interprets the phrase “Kaufman-type ion beam source” to include a thermionic (hot-wire) cathode, an anode, grids, and magnets.
This court detects no legally significant distinction between the phrases, “Kaufman-type ion beam source” and “Kaufman-type ion beam gun.” During the prosecution of the reissue, Litton used the terms “gun” and “source” interchangeably. For example, in response to one of the examiner’s rejections, Litton stated: “Applicants need not add the words ‘Kaufman gun’ or ‘Kaufman source’ to claim 1 because claim 1 cannot properly be construed to refer to any other ion beam gun but the Kaufman gun.” Paper No. 15 at 8.
In sum, after consideration of the primary sources for construing patent claim meaning, this court interprets the phrase “Kaufman-type ion beam source” to encompass any ion beam gun with the four stated components: a hot-wire cathode, an anode, grids, and magnets.
Literal Infringement
Literal infringement requires that the accused device contain each limitation of the claim exactly; any deviation from the claim precludes a finding of literal infringement. See Cole v. Kimberly-Clark Corp., 102 F.3d 524, 532, 41 USPQ2d 1001, 1007 (Fed.Cir.1996), cert. denied, — U.S. -, 118 S.Ct. 56, 139 L.Ed.2d 20 (1997). Because Honeywell’s hollow cathode process does not employ a hot-wire cathode and its RF ion beam process does not use a cathode at all, this court affirms the trial court’s grant of JMOL that the accused hollow cathode and RF ion beam processes do not literally infringe the asserted claims of the ’849 reissue.
Infringement Under the Doctrine of Equivalents
Because Honeywell’s processes do not literally infringe Litton’s claimed method, this court now considers the issue of infringement under the doctrine of equivalents. Litton asserts that Honeywell’s hollow cathode and RF ion beam sources are equivalents of the “Kaufman-type ion beam source” in claim 1. Honeywell contends that they are not equivalents, and, more importantly, that Litton’s admissions and amendments during patent prosecution estop it from asserting any equivalents to the “Kaufman-type ion beam source.”
In reconciling the uncertainty surrounding application of the doctrine of equivalents with the definitional and public-notice functions of the statutory claiming requirement, the Supreme Court endorsed an element-by-element analytical framework for infringement. See Warner-Jenkinson, 520 U.S. at -, 117 S.Ct. at 1049; see also Pennwalt Corp. v. Durand-Wayland, Inc., 833 F.2d 931, 935, 4 USPQ2d 1737, 1741 (Fed.Cir.1987) (in bane). In this framework, “[e]ach element contained in a patent claim is deemed material to defining the scope of the patented invention, and thus the doctrine of equivalents must be applied to individual elements of the claim, not to the invention as a whole.” Warner-Jenkinson, 520 U.S. at -, 117 S.Ct. at 1049. In applying this method, the Court cautioned against a range of equivalents with “such broad play as to effectively eliminate that element in its entirety.” Id. The essential inquiry is whether “the accused product or process contain[s] elements identical or equivalent to each claimed element of the patented invention.” Id. at-, 117 S.Ct. at 1054.
As discussed above, the “Kaufman-type ien beam source” phrase means a source with a hot-wire cathode, an anode, grids, and magnets. The jury’s special verdict form indicates that it adopted a construction of “Kaufman-type ion beam source” as “[a]ny broad-beam, multiapertured, gridded ion beam source, which includes any hollow cathode gun and any radio frequency gun.” Therefore, the jury based its finding of infringement under the doctrine of equivalents on an erroneous claim construction.
*1455Under an element-by-element analysis, an error in claim construction propagates into the jury’s equivalence determination. The fact finder must determine whether a specific feature of the accused device or process is equivalent to a given claim element, as properly construed. Due to reliance on an erroneous claim construction, this court vacates the jury’s verdict on the issue of infringement under the doctrine of equivalents for both the hollow cathode and RF processes.
This same claim interpretation error cannot be found, however, in the district court’s grant of JMOL that Honeywell’s RF source could not infringe under the doctrine of equivalents. The district court adopted an interpretation of the “Kaufman-type ion beam source” identical to that which this court has adopted. In denying infringement under the doctrine of equivalents, the district court reasoned—despite evidence of interchangeability and insubstantial differences— that the “Kaufman-type ion beam source” limitation could not embrace a range of equivalents that extended to a source that worked in a substantially different way. See 1995 WL 366468 at *46; Litton I, 87 F.3d at 1572.
In light of the additional clarification supplied by case law in the intervening period, this court vacates the district court’s grant of JMOL to allow it to consider other rationales for judgment as a matter of law—for both the hollow cathode and RF processes. Specifically, this court provides the district court an opportunity to consider: (1) whether, as outlined below, prosecution history estoppel precludes infringement under the doctrine of equivalents, (2) whether prior art precludes infringement under the doctrine of equivalents, see General Am. Transp. Corp. v. Cryo-Trans, Inc., 93 F.3d 766, 771, 39 USPQ2d 1801, 1804 (Fed.Cir.1996), cert. denied, — U.S. -, 117 S.Ct. 1334, 137 L.Ed.2d 493 (1997); Wilson Sporting Goods Co. v. David Geoffrey & Assocs., 904 F.2d 677, 684, 14 USPQ2d 1942, 1948 (Fed.Cir.1990), and (3) whether, under the reasoning of Warner-Jenkinson, application of the doctrine of equivalents would “effectively eliminate [an] element in its entirety, Warner-Jenkinson, 520 U.S. at -, 117 S.Ct. at 1049. In sum, this court remands to permit examination of the application of the doctrine of equivalents.
Prosecution History Estoppel
In Warner-Jenkinson, the Supreme Court changed some aspects of the law of prosecution history estoppel. However, the extent of this change was limited. Honeywell argues that Wamer-Jenkinson held that if a claim element has been added by amendment for reasons of patentability, prosecution history estoppel automatically bars all equivalents for that element, regardless of whether the administrative record before the Patent and Trademark Office (PTO) shows that the applicant surrendered all coverage beyond the literal claim scope. Honeywell’s position, however, would both bar after-arising equivalents expressly approved by the Supreme Court and bar any equivalents whatsoever to the vast majority of claim limitations amended during patent prosecution.
This court determines in this opinion that the Supreme Court did not in fact effect such a sweeping change. Instead the Supreme Court adhered to the longstanding doctrine that an estoppel only bars recapture of that subject matter actually surrendered during prosecution. The common practice of amending a claim during prosecution, even amending to overcome prior art, does not necessarily surrender all subject matter beyond the literal scope of the amended claim limitation.
In Warner-Jenkinson, the Supreme Court first rejected several arguments for negation of the doctrine of equivalents. See 520 U.S. at---, 117 S.Ct. at 1047-48. Next the Court considered doctrines which restrain application of the doctrine. See id. at---, 117 S.Ct. at 1049-51. At this juncture, the Court discussed the “well-established limit on non-literal infringement, known variously as ‘prosecution history estoppel’ and ‘file wrapper estoppel.’ ” Id. at *1456-, 117 S.Ct. at 1049. The Court noted: “In each of our cases ... prosecution history estoppel was tied to amendments made to avoid the prior art, or otherwise to address a specific concern—such as obviousness.... ” Id. Thus, the Court reaffirmed that the reason for claim amendments remains relevant to application of an estoppel. See id. at-, 117 S.Ct. at 1050. A claim amendment to avoid prior art prevents recapture of subject matter the applicant surrendered to obtain patent protection. “Where the reason for the [claim] change was not related to avoiding the prior art ... [the prosecution history] does not necessarily preclude infringement by equivalents____” Id. at---, 117 S.Ct. at 1050-51. The Supreme Court thus noted that an amendment during prosecution does not foreclose all application of the doctrine of equivalents. Rather a court must inquire into the reasons for an amendment before invoking an estoppel.
In Warner-Jenkinson, however, the patent prosecution record disclosed no reason for Hilton Davis’s claim amendments. See id. at-, 117 S.Ct. at 1051. In this context, the Court established a presumption that when an applicant narrows a claim element during prosecution, a trial court should presume that the applicant did so for a reason related to patentability. See id. Consequently, where a patent owner cannot show a reason for the amendment other than patent-ability, “a court should presume that the purpose behind the ... amendment is such that prosecution history estoppel would apply.” Id. at ——•, 117 S.Ct. at 1054. Thus, the Supreme Court articulated an additional rule to trigger, in applicable circumstances, prosecution history estoppel. But that is all the Supreme Court articulated. Although Warner-Jenkinson supplied new guidance about when prosecution history estoppel might apply, the Court did not change the scope or effect of the estoppel.
In accord with the Supreme Court’s understanding, this court has repeatedly stated that application of prosecution history estoppel does not necessarily limit a patentee to the literal language of the amended element—even when an amendment has been made to overcome the prior art. See La-Bounty Mfg., Inc. v. United States Int’l Trade Comm’n, 867 F.2d 1572, 1575-76, 9 USPQ2d 1995, 1998-99, (Fed.Cir.1989) (reversing an administrative law judge’s ruling otherwise); Dixie USA, Inc. v. Infab Corp., 927 F.2d 584, 588, 17 USPQ2d 1968, 1970-71 (Fed.Cir.1991) (confirming that as a general principle, “a total preclusion of equivalence should not apply”). Thus, an amendment to claim language in response to prior art “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. It is not fatal to application of the doctrine itself.” Hughes Aircraft Co. v. United States, 717 F.2d 1351, 1363, 219 USPQ 473, 481 (Fed.Cir.1983); see also Modine Mfg. Co. v. United States Int’l Trade Comm’n, 75 F.3d 1545, 1555-56, 37 USPQ2d 1609, 1616 (Fed.Cir.), cert. denied, 518 U.S. 1005, 116 S.Ct. 2523, 135 L.Ed.2d 1048 (1996).
Despite the Supreme Court’s overall reasoning, Honeywell argues that Warner-Jenkinson eliminated any scope for prosecution history estoppel. In other words, according to Honeywell, any claim language amended during prosecution for reasons related to patentability has no range of equivalents. To support its transformation of infringement rules, Honeywell points to the following language:
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 [sic] equivalents as to that element.
Warner-Jenkinson, 520 U.S. at -, 117 S.Ct. at 1051 (emphasis added).
Read in context, this passage does not effect the sweeping change advocated by Honeywell. As noted above, the entire con*1457text of the Warner-Jenkinson opinion shows that the Supreme Court approved the PTO’s practice of requesting amendments with the understanding that the doctrine of equivalents would still apply to the amended language. As noted above, this court had repeatedly articulated that rule. In that context, it is telling that the Supreme Court was careful to avoid disturbing basic rules and assumptions. Footnotes 6 and 7 further evince an intent “[t]o [not] change so substantially the rules of the game.” Warner-Jenkinson, 520 U.S. at- n. 6, 117 S.Ct. at 1050 n. 6. Footnote 6 appears to explain further the Court’s rejection of the petitioner’s proposed any-type-of-amendment, no-equivalents rule. Footnote 7 explains that “where a change is made to overcome an objection based on the prior art,” a court should explore “the reason (right or wrong) for the objection and the manner in which the amendment addressed and avoided the objection.” Id. at-, 117 S.Ct. at 1051 n. 7. These footnotes evince the Court’s intent to change as little as possible.
Also, the accumulated case law of the Supreme Court and this court on the scope of prosecution history estoppel contradict Honeywell’s reading of a single isolated sentence. For example, Keystone Driller Co. v. Northwest Engineering Corp., 294 U.S. 42, 47-48, 55 S.Ct. 262, 264-65, 79 L.Ed. 747 (1935), and Smith v. Magic City Kennel Club, 282 U.S. 784, 788-90, 51 S.Ct. 291, 293-94, 75 L.Ed. 707 (1931), cited by the Warner-Jenkinson Court, acknowledge the concept that an estoppel has an appropriate scope. In both cases, the Court inquired into the scope of the original claim and the prior art before estopping the patentee from asserting the equivalence of a particular feature of the accused device. If an amendment related to patentability automatically precluded all equivalents to the amended language, the Court’s inquiry would have been irrelevant.
Honeywell’s interpretation also reads too narrowly the Supreme Court’s endorsement of “after-arising equivalents.” The Supreme Court expressly rejected the argument that the doctrine of equivalents should be limited to equivalents that are disclosed within the patent itself or equivalents known at the time the patent was issued. See Warner-Jenkinson, 520 U.S. at---, 117 S.Ct. at 1052-53. Nowhere did the Supreme Court suggest that such after-arising equivalents would be barred as a matter of law whenever a claim is amended to overcome prior art before the examiner. Such an interpretation is also inconsistent with the Supreme Court’s admonition that a “brighter line for determining whether a patentee is estopped under certain circumstances is not a sufficient reason for adopting such a rule.” Id. at-n. 6,117 S.Ct. at 1050 n. 6.
Finally, the Supreme Court did not find an estoppel based on the record before it, but remanded to this court, which, in turn, remanded to the district court for fact finding on application of the presumption. Therefore, the Supreme Court did not reach the question of the proper scope of estoppel for an amended limitation. In sum, 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.
The Supreme Court began its discussion of prosecution history estoppel by acknowledging that in each of its cases estoppel “was tied to amendments made to avoid the prior art, or otherwise to address a specific concern—such as obviousness—that arguably would have rendered the claimed subject matter unpatentable.” Warner-Jenkinson, 520 U.S. at-, 117 S.Ct. at 1049. Thus, from the outset, the Court recognized that the Warner-Jenkinson standard “related to patentability” encompassed amendments “made to avoid the prior art.” The Court then observed:
It is telling that in each case this Court probed the reasoning behind the Patent Office’s insistence upon a change in the claims. In each instance, a change was demanded because the claim as otherwise *1458written was viewed as not describing a patentable invention at all—typically because what it described was encompassed within the prior art.
Id. at-, 117 S.Ct. at 1050.
The Supreme Court then proceeded to recognize that “there are a variety of other reasons why the PTO may request a change in claim language.” Id. For this proposition, the Court cited the United States’ amicus brief, which refers to amendments made to overcome indefiniteness and nonenablement rejections. See Brief for United States as Amicus Curiae at 22-23, Warner-Jenkinson, — U.S. -, 117 S.Ct. 1040 (1997) (No. 95-728), available in 1996 WL 172221. Thus, the Court concluded, “[o]ur prior cases have consistently applied prosecution history estoppel only where claims have been amended for a limited set of reasons, and we see no substantial cause for requiring a more rigid rule invoking an estoppel regardless of the reasons for a change.” Warner-Jenkinson, 520 U.S. at-, 117 S.Ct. at 1050.
Although not automatically erecting an estoppel, an amendment made for reasons other than patentability may still give rise to an estoppel. This court has acknowledged that even arguments made during prosecution without amendments to claim language—if sufficient to evince a clear and unmistakable surrender of subject matter-may estop an applicant from recapturing that surrendered matter under the doctrine of equivalents. See Hoganas AB v. Dresser Indus., Inc., 9 F.3d 948, 952, 28 USPQ2d 1936, 1939 (Fed.Cir.1993). Estoppel by clear and unmistakable surrender without claim amendments may arise even when the arguments to the examiner were not necessary to distinguish prior art. Texas Instruments Inc. v. United States Int’l Trade Comm’n, 988 F.2d 1165, 1174-75, 26 USPQ2d 1018, 1025 (Fed.Cir.1993). This principle presupposes that the applicant has made the surrender unmistakable enough that the public may reasonably rely on it. By logical extension, if an applicant makes an amendment unrelated to patentability which evinces an unmistakable surrender, that action will preclude recapture of the surrendered subject matter under the doctrine of equivalents.
Prosecution History of the ’849 Reissue
Turning to application of the doctrine of prosecution history estoppel in this case, this court first recognizes that Litton amended its claims. The applicant amended claim 1 to add language about the “Kaufman-type ion beam source.” Thus, this court must examine whether that amendment occurred for a reason related to patentability. The applicant filed the ’849 reissue on July 2, 1985. At that time, independent claim 1 read as follows:
1. A method of fabricating multiple layer optical films, said multiple layer optical films comprising optical layers having different indices of refraction comprising:
bombarding targets obliquely with an ion beam in a vacuum chamber to sputter deposit a plurality of optical film layers on a base;
controlling the atmosphere inside the vacuum chamber to provide sufficient gas to sustain the ion beam and the proper amount of oxygen to accomplish proper stoichiometry of the thin films; and
depositing multiple layers of different materials on said base by varying the targets being bombarded by the ion beam; and
continuously rotating said base during the deposition of said multiple optical layers.
(Emphasis added.) In the next month, August 1985, Litton submitted a patentability report citing eighty-two references. Litton took pains to separately list and distinguish each of these references. Often Litton explained that a given reference “does not disclose Applicant’s [Litton’s] claimed ion beam sputtering techniques.” The list of references included an article by G.E. Lane and J.C. Anderson, “The Nucleation and Initial Growth of Gold Films Deposited onto Sodium *1459Chloride by Ion-Beam Sputtering,” that appeared in Thin Solid Films in 1975. The Lane and Anderson (Lane) article disclosed RF ion beam sputtering of gold films onto substrates at low deposition rates.
In October 1985, one of Litton’s competitors filed a protest under 37 C.F.R. § 1.291 (1985). The protest highlighted two references: (1) an article by W. Laznovsky entitled “Advances in Low Energy Ion Beam Technology,” in Research/Development in 1975; and (2) French Patent No. 2,129,996, whose counterpart issued in 1975 as U.S. Patent No. 3,925,187 to Bernard. The Laznovsky article described a device for ion beam sputtering deposition with a rotating target in an atmosphere controlled for oxygen loss. The Bernard patent was included in the patentability report and disclosed an apparatus for forming multiple layer coatings on microelectronic or semiconductor devices with a duoplasmatron ion beam. The protest for the first time brought Laznovsky to the examiner’s attention.
In December 1985, after receipt of the protest, the examiner rejected all the claims as obvious under section 103. In addition to extensive reliance on the Laznovsky article, the examiner rejected the claims “under 35 U.S.C. 103 as ... unpatentable over the combination of admitted prior art contained in [the] ‘Patentability Report.’ ” See Paper No. 6.
In its June 1986 response, Litton acknowledged that all the prior art in the patentability report, as well as the Laznovsky article, occasioned the obviousness rejection. Paper No. 10 at 4-5. Indeed Litton again a second time listed by number each of the eighty-two references previously cited along with the Laznovsky article and expressly distinguished its claims from each piece of prior art. Beyond the express reference-by-reference distinctions, Litton further distinguished the invention from these eighty-three references as follows:
What is novel and unobvious about Applicants’ invention is the use of ion-beam sputtering techniques, and particularly the use of Kaufman gun ion beam techniques, to form multiple-layer optical films on a base with optical layers having different indices of refraction____
Id. at 5 (emphasis added). At this point, Litton also included for the first time its “Kaufman-type” limitation in several dependent claims.
Nonetheless, the examiner issued a final rejection again under section 103 on July 3, 1986. The examiner focused on one ion beam source in the prior art—the duoplasmatron source of the Bernard patent. The rejection highlighted the discontinuity between the claims’ general requirement for an “ion beam source” and Litton’s arguments directed to a more narrow “Kaufman-type ion beam source”:
As for applicants’ arguments based upon an alleged unexpected result obtained by use of a Kauffman [sic] type ion gun rather than a duoplasmatron type ion gun, it is noted that most of the claims are not limited to any particular type of ion gun; thus applicants are arguing a limitation not in the claims. Moreover, Kauffman [sic] type ion sources are standard in the art; their use in ion beam sputtering apparatus constitutes nothing unobivous [sic]. See e.g. Laznovsky at page 47, column 1. Lastly, applicants merely allege an unexpected result; no comparative data are supplied.
Paper No. 11 at 6. In conclusion, the examiner again a second time rejected Litton’s claims “over the combination of references as made in the first Office action.” In rejecting again over all eighty-three references, the examiner pointed out as noted above that Litton had not limited its claims “to any particular type of ion gun.”
In December 1986, Litton replied: “Contrary to the contentions in Paper No. 11, Applicants’ claims relate only to Kaufinantype ion-beam sputtering because duoplasmatron sources are today, and were in 1978, devices that no one of ordinary skill in the art would use for forming multiple layer optical coatings.” Paper No. 15 at 7. Litton argued that the examiner had read claim 1 *1460broadly to include duoplasmatron ion beam sources when it “[could not] properly be construed to refer to any other ion beam gun but the Kaufman gun.” Id. at 8.
In addition, Litton submitted three declarations under 37 C.F.R. § 1.132 supporting its nonobviousness contentions. These declarations highlighted the unexpected results from the Kaufman-type guns and the slow deposition rates of the duoplasmatron. In particular, the Baumeister declaration contended:
Those skilled in the coating arts, in reading. claims 1 and 3-10 of this application for reissue, knowing that duoplasmatron sources are unworkable for making multiple layer optical coatings of any kind, let alone the multiple layer optical coatings referred to in claims 1 and 3-10, would reasonably construe these claims to refer not to duoplasmatron sputtering sources, but only to the Kaufman-type ion-beam guns referred to in the specification of this application.
Paper No. 16 at 7. Baumeister also included as an exhibit a chapter of James M.E. Harper’s Thin Film Processes, entitled “Ion Beam Deposition” and published in 1978. The Harper chapter discussed several ion beam sources, particularly the Penning, Kaufman, and hollow cathode ion beam sources.
Recognizing that the examiner had rejected over all eighty-three references, however, Litton again a third time cited and individually distinguished each of the cited references. Indeed for emphasis, Baumeister also resubmitted all eighty-three references with his declaration as evidence that he had considered and distinguished the relevant prior art.
Apparently, the declarations persuaded the examiner. After a January 1987 telephone communication, he made the following notes:
The examiner initiated the interview to inform applicant’s attorney that allowance of the application was possible if all the claims were limited to Kaufman gun sources being used in the claimed process. Unless the claims were so limited, the examiner indicated that a 35 U.S.C. 112 Par 2 rejection[,] on the basis that applicants were not claiming what they regarded as their invention in view of their argument contained in paper no. 15 filed 12/16/86[,] would be made in the next office action on the merits. The paper # 15 overcomes the other previously made objections.
Paper No. 17. Thus, at this point, the examiner conditioned allowance of the claims on an amendment limiting the invention to Kaufman gun sources.
In its March 1987 response, Litton treated the examiner’s threat as a formal section 112 rejection. Litton incorporated the Kaufman-type ion beam source limitation into claim 1 (the “Kaufman-type amendment”), stating that the amendment was made “to overcome the rejections of these claims under 35 USC § 112, second paragraph.” With this action, Litton submitted for still a fourth time all eighty-three references and a number-by-number explanation of the distinctions between its invention and each of those references.
The examiner allowed the amended claims and closed prosecution on the merits on June 26, 1987. Claim 1, as amended, read as follows:
1. A method of fabricating multiple layer optical films, said multiple layer optical films comprising optical layers having different indices of refraction comprising:
bombarding targets obliquely with an ion beam produced by or derived from a Kaufinan-type ion beam source in a vacuum chamber to sputter deposit a plurality of optical film layers on a base;
controlling the atmosphere inside the vacuum chamber to provide sufficient gas to sustain the ion beam and the proper amount of oxygen to accomplish proper stoichiometry of the thin films; and
*1461depositing multiple layers of different materials on said base by varying the targets being bombarded by the ion beam; and
continuously rotating said base during the deposition of said multiple optical layers.
(Emphasis added.) Claim 1 is representative of the other claims; wherever the phrase “an ion beam” had appeared, Litton substituted “an ion beam produced by or derived from a Kaufman-type ion beam source.”
Even this action, however, did not complete the prosecution history. After another technical examiner interview in April 1987, Litton submitted again the entire list of the eighty-three references complete with distinctions from its invention. This final June 1987 submission was the fifth time (sixth counting the Baumeister declaration) that Litton had raised and expressly distinguished its invention from each of the eighty-three references. At that point, an inequitable conduct investigation began. The investigation ended just before issuance of the ’849 reissue on January 31,1989.
Effect of the Kaufman-type Amendment
In view of the foregoing prosecution history, this court now considers whether the Kaufman-type amendment was related to patentability. Litton argues that the amendment came after the applicants had “overcome the other previously made [obviousness] rejections.” Paper No. 17 at 1. Litton characterizes its amendment as a response to a rejection under 35 U.S.C. § 112, ¶2 because the “applicants were not claiming what they regarded as their invention.” Id. Litton cites Warner Jenkinson for the proposition that amendments in response to section 112 rejections are not related to patentability. Because its amendment responded to a section 112 rejection, Litton argues that its amendment was not “related to patentability” and this court should not presume that the amendment created an estoppel.
Although amendments made in response to indefiniteness and enablement rejections are generally not made “in response to the prior art,” the amendment made in response to the section 112 rejection at issue here, a “regards as his invention” rejection, was related to patentability. Without evidence to the contrary, an examiner generally should presume that a claim recites what the applicant regards as his invention. See In re Moore, 58 C.C.P.A. 1042, 439 F.2d 1232, 1235, 169 USPQ 236, 238 (CCPA 1971). Moreover, “some material submitted by [the] applicant, other than his specification ” must warrant this type of section 112 rejection. In re Conley, 490 F.2d 972, 976, 180 USPQ 454, 457 (CCPA 1974). Thus, an examiner generally makes a “regards as his invention” rejection only as an applicant’s position becomes clear over the course of prosecution. In other words, this rejection almost always follows some other rejection of the inventive material set forth in the claim. See, e.g., In re Prater, 56 C.C.P.A. 1381, 415 F.2d 1393, 162 USPQ 541 (CCPA 1969) (followed section 101 non-statutory subject matter rejection).
Consequently, this court cannot ignore the rejections which preceded this “regards as his invention” rejection. This particular “regards as his invention” rejection followed a series of obviousness rejections. In effect, the examiner threatened to reject again for obviousness unless the applicant restated its claim to match the scope of its narrow arguments for patentability. In this context, the section 112 rejection carried the same message as the prior obviousness rejection. An obviousness rejection is of course made in response to prior art. Consequently, this court determines that Litton made its amendment for reasons related to patentability.
Scope of the Estoppel
Because Litton’s amendment was made for reasons related to patentability, prosecution history estoppel applies to the phrase “Kaufman-type ion beam source.” As previously indicated, however, estoppel does not bar Litton from invoking the doctrine of equivalents altogether with respect to the “Kaufman-type ion beam source” limitation. Ac*1462cordingly, this court now considers the scope of what Litton actually surrendered.
As 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. See Mark I Mktg. Corp. v. R.R. Donnelley & Sons Co., 66 F.3d 285, 291, 36 USPQ2d 1095, 1100 (Fed.Cir.1995). As noted earlier, either amendments or arguments made by an applicant may be the basis for this conclusion.
When prosecution history estops a patentee, the court ascertains the scope of the estoppel in several ways. First, “a patentee is estopped from recovering through equivalency that which was deemed unpatentable in view of the prior art.” Pall Corp. v. Micron Separations, Inc., 66 F.3d 1211, 1219 (Fed.Cir.1995). In other words, when an applicant, in response to an examiner’s prior art rejection, amends a claim by substituting one limitation for another, the applicant cannot later assert that the original limitation is an equivalent of the substituted limitation. Thus, the doctrine prevents the applicant from completely recapturing the subject matter rejected by the examiner.
In addition, when an applicant narrows a claim element in the face of an examiner’s rejection based on the prior art, the doctrine estops the applicant from later asserting that the claim covers, through the doctrine of equivalents, features that the applicant amended his claim to avoid. A patentee is also estopped to assert equivalence to “trivial” variations of such prior art features. See Southwall Techs., Inc. v. Cardinal IG Co., 54 F.3d 1570, 1580, 34 USPQ2d 1673, 1680 (Fed.Cir.1995). Depending on the facts of the case, an amendment may also limit the patentee to its literal claim scope. See Wang Lab., Inc. v. Toshiba Corp., 993 F.2d 858, 867-68, 26 USPQ2d 1767, 1775-76 (Fed.Cir.1993); Wang Lab., Inc. v. Mitsubishi Elecs. America, 103 F.3d 1571, 1577-78, 41 USPQ2d 1263, 1268-69 (Fed.Cir.1997).
In addition, as noted earlier, an applicant’s arguments may constitute a clear and unmistakable surrender of subject matter. Such arguments preclude recapture of that subject matter. See Texas Instruments, 988 F.2d at 1174-75 (holding that applicants had clearly represented that “same-side gating” did not work and could not be asserted as an equivalent). As noted above, this type of estoppel can arise regardless of the Warner-! enkinson presumptions. See ante at 1456. Of course, applicants commonly make arguments in combination with an amendment, as in this case. In such circumstances, the scope of estoppel is a product of the effects of both factors working in concert.*
According to these principles, Litton’s conduct clearly estops it from asserting that a duoplasmatron source is equivalent to a “Kaufman-type” source. This estoppel falls within one of the categories mentioned above—an amendment which narrows a claim element to avoid prior art. Duoplasmatron sources were in the prior art specifically cited by the examiner in his rejection of claim 1. In response, Litton narrowed its claim language “ion beam source” to “Kaufman-type ion beam source” by making an amendment this court has determined to be “related to patentability.” This record suffices to exclude duoplasmatron sources from the permissible range of equivalents.
*1463Determining whether Litton’s conduct estops it from claiming that either Honeywell’s hollow cathode or its RF ion beam source is an equivalent presents a more complicated inquiry. The examiner did not cite a hollow cathode ion beam source nor an RF ion beam source. Therefore, Litton did not amend its claim to avoid these sources. However, this alone does not preclude estoppel.
Presented with section 103 rejections to all its claims based upon the many references which Litton repeatedly cited and expressly distinguished, Litton demonstrated patentability over the prior art by submitting declarations attesting to the unexpected results obtained from the Kaufman-type source. See Paper Nos. 16 at 5 (Baumeister declaration) and 16A at 3 (declaration of Dr. Samuel Lu). Moreover, Litton argued in unmistakable terms that the phrase “ion beam source” referred only to a “Kaufman-type ion beam source.” See, e.g., Paper No. 15 at 8. The record shows Litton made that argument in the face of at least five other types of ion beam sources among the eighty-three references and the Harper chapter. These included a hollow cathode source discussed in the Harper chapter and an RF ion beam source described in the Lane article (Reference No. 73 among Litton’s oft-cited references).
On the unique facts of this case—Litton’s unmistakable arguments that its claims encompassed only the “Kaufman-type ion beam source,” the five (or six) unambiguous declarations by Litton and its affiants expressly distancing its invention from the references before the examiner, and Litton’s amendment—one of ordinary skill in the art would reasonably conclude that Litton surrendered the other ion beam sources disclosed in the references before the examiner. Contrary to what the dissent suggests, these references were central to the course of the prosecution. Moreover, this result does not automatically, without more, mandate that in every case an estoppel arises based on any reference before an examiner. Rather, in this unique case, Litton repeatedly referred expressly to the many references it presented to the examiner, distinguished each reference by number over and over, and then further insisted that its claims encompassed only processes with Kaufman-type sources. To confirm that its invention was different from these references, Litton further amended its claim—in the face of its knowledge of all the ion beam sources disclosed in these references—in response to the examiner’s rejection. Under these telling circumstances, the administrative record estops Litton from asserting the equivalence of any ion beam source before the examiner.
In view of Honeywell’s accused devices, the hollow cathode source disclosed in the Harper chapter and the RF source disclosed in the Lane article are particularly relevant to the inquiry of estoppel here. The Harper chapter provided a detailed discussion of the Kaufman ion source and then suggested substituting a hollow cathode for “the [hot-wire] cathode in applications where lifetime or contamination from the cathode are important.”
The Lane article described an ion beam source comprising an RF discharge maintained in an argon-filled glass cylinder. The RF discharge was maintained by means of one external coil connected to a generator and regulated by means of a second coil, wound around the first and connected to a capacitor. Two focusing electrodes achieved the ion extraction. A plate at the end of the cylinder with a central hole 3/8 inch in radius served as one electrode; the second electrode was located within the cylinder and contained a second hole 1/4 inch in radius, aligned 1/4 inch from the plate. Conceivably, the two focusing holes are analogous to the Kaufman-type grids, which have numerous holes.
Other sources before the examiner may be relevant. However, the trial court made no findings about the relationship, if any, between any of the sources before the examiner and the sources in Honeywell’s accused processes. For example, if the pertinent differences between either (1) the Harper source and Honeywell’s hollow cathode source or (2) *1464the Lane source and Honeywell’s RF source are trivial, then the surrender' of each disclosed source necessarily includes a surrender of its corresponding accused source. See Southwall, 54 F.3d at 1580. Alternatively, if the differences between either (1) the Harper source and Honeywell’s hollow cathode sources or (2) the Lane source and Honeywell’s RF sources are not trivial, then Lit ton did not surrender the accused source by surrendering the corresponding disclosed source.
Yet whether Honeywell’s hollow cathode and RF ion beam sources constitute trivial variations of the Harper and Lane sources, respectively, are factual determinations that the district court never addressed. The same is true for any other sources before the examiner. Because the trial record is silent on these issues, this court cannot conclude as a matter of law whether the record estops Litton from asserting that the accused hollow cathode and RF sources are equivalents of the Kaufman-type limitation. See Hormone Research Found. v. Genentech, Inc., 904 F.2d 1558, 1564, 15 USPQ2d 1039, 1044 (Fed.Cir.1990) (“[T]he scope of estoppel can depend on factual questions regarding the prosecution history, which may be disputed____”).
Accordingly, this court remands for a determination of the factual issues underlying prosecution history estoppel: namely, whether the accused hollow cathode source or the accused RF source constitutes a trivial variation of any of the sources before the examiner. If the district court determines that the variations are trivial for either of the accused sources, it should enter judgment as a matter of law that the process using that accused source does not infringe under the doctrine of equivalents. However, if the district court concludes that there is no estoppel, then it should proceed with the factual determination of infringement under the doctrine of equivalents.
Regardless of the triviality of the pertinent differences between Honeywell’s RF source and any of the sources before the examiner, Honeywell nonetheless claims it should prevail on a different theory. Honeywell maintains that Litton should be es-topped from asserting the equivalence of at least the accused RF ion beam source because, at the time of the Kaufman-type amendment, one of ordinary skill in the art knew that Honeywell’s RF source could have been used as a substitute in Litton’s patented process. Honeywell argues that that knowledge should be imputed to Litton, and therefore Litton’s arguments and its amendment manifest a surrender of any potential equivalents covering Honeywell’s RF process.
Honeywell points to a number of references in the trial court record—but not the administrative record before the PTO—establishing that RF ion beam sources similar to Honeywell’s were in existence at the time of the Kaufman-type amendment. For example, the article, “Radio Frequency Ion Sources for Electrostatic Propulsion” by Horst W. Loeb, was distributed to one of Litton’s named inventors at a 1971 symposium. The Loeb reference disclosed a method of ion extraction from an RF discharge plasma using aligned, perforated plates, which resemble grids under Honeywell’s ex post reading. Another reference is entitled Operation of Broad-Beam Sources, by Harold R. Kaufman, the inventor of the Kaufman gun, and Raymond S. Robinson. The Kaufman book indicates that, by at least 1987, a few gridded RF ion beam sources had been used in industrial applications.
If one of ordinary skill in the art knew of the interchangeability of Honeywell’s RF source when the Kaufman-type amendment was made, so Honeywell’s argument goes, then a person of ordinary skill reasonably would conclude from the patent’s prosecution history that this source, too, had been surrendered. See Mark I Mktg., 66 F.3d at 291-92.
However, even assuming arguendo that one of ordinary skill in the art would have known of the interchangeability of Honeywell’s RF source at the time of the Kaufman-type amendment, this argument must fail. To accept this argument would be essentially *1465to accept the premise that when an applicant makes an amendment to an element to overcome the prior art, he has automatically surrendered all then-known substitutes to the amended element. Because known interchangeability is often synonymous with equivalence, the result would generally be to allow only after-arising equivalents for amended elements. This is simply a variation of the no-equivalents theory already rejected by the Supreme Court and this court and must itself be rejected.
Reliance on this theory would not comport with the principles on which prosecution history estoppel is based. An applicant cannot surrender that which he does not know, but this theory would impute the knowledge of those of ordinary skill in the art to the applicant and deem it surrendered. Because the prosecution history is objective evidence of what knowledge the applicant has of the art, prosecution history estoppel should remain limited to the prosecution history.
State Law Claims
In addition to the findings of infringement of Litton’s patent, the jury found Honeywell liable for two state law torts, intentional interference with contractual relations and intentional interference with prospective economic advantage. In making its liability determinations, the jury determined that Honeywell’s activities were not privileged because Honeywell used “wrongful means.” The jury’s special verdict does not reveal what it considered to be “wrongful means.” The district court granted Honeywell’s motion for JMOL on the state law claims. Although an appellate court must sustain a jury verdict against a JMOL if there are any reasonable grounds for the verdict, see Litton I, 87 F.3d at 1573 n. 3, an appellate court must also vacate a jury verdict and remand for a new trial if a jury may have relied on an impermissible basis in reaching its verdict, see Richards v. Michelin Tire Corp., 21 F.3d 1048, 1055 (11th Cir.1994). Because the jury may have relied on its finding of patent infringement as the “wrongful means,” and because this court vacates and remands on the issue of patent infringement, it must do the same for the state law claims. Accordingly, this court reverses the trial court’s grant of JMOL, vacates the jury’s verdict, and remands for further proceedings in accordance with state law.
Conclusion
This court: (1) reverses the trial court’s grant of JMOL and reinstates the jury’s verdict that the asserted claims of the ’849 reissue are not invalid for obviousness; (2) reverses the trial court’s determination of inequitable conduct; (3) affirms the trial court’s order finding intervening rights; (4) affirms the trial court’s grant of JMOL that the accused hollow cathode and RF ion beam processes do not literally infringe the asserted claims of the ’849 reissue; (5) vacates the trial court’s grant of JMOL that the accused RF ion beam process does not infringe the asserted claims of the ’849 reissue under the doctrine of equivalents, vacates the jury’s verdict on that issue, and remands for further proceedings; (6) vacates the jury’s verdict that the accused hollow cathode process infringes the asserted claims of the ’849 reissue under the doctrine of equivalents and remands for further proceedings; (7) reverses the trial court’s grant of JMOL with respect to the torts of intentional interference with contractual relations and intentional interference with prospective economic advantage, vacates the jury’s verdict on that issue, and remands for further proceedings; and (8) affirms the trial court’s grant of a new trial on damages, if necessary.
Costs
Each party shall bear its own costs.
AFFIRMED-IN-PART, REVERSED-IN-PART, VACATED-IN-PART and REMANDED.
Contrary to the suggestions in the dissenting opinion, it is not necessary that a reference be specifically cited by the examiner as the reason for a rejection in order for it to give rise to an estoppel. Arguments made by an applicant in an information disclosure statement or otherwise during prosecution may form the basis of an estoppel without regard to whether the argument was made in response to a rejection or the prior art was cited by the examiner. See Lockwood v. American Airlines, Inc., 107 F.3d 1565, 1575, 41 USPQ2d 1961, 1969 (Fed.Cir.1997); Ekchian v. Home Depot, Inc., 104 F.3d 1299, 1304, 41 USPQ2d 1364, 1368 (Fed.Cir.1997); Texas Instruments, Inc. v. United States Int'l Trade Comm’n, 988 F.2d 1165, 1174, 26 USPQ2d 1018, 1025 (Fed.Cir.1993); Coleco Indus., Inc. v. United States Int'l Trade Comm'n, 65 C.C.P.A. 105, 573 F.2d 1247, 1256-57, 197 USPQ 472, 479-80 (CCPA 1978).