Loral Fairchild Corp. v. Victor Co. of Japan, Ltd.

931 F. Supp. 1014, 1996 U.S. Dist. LEXIS 14204, 1996 WL 402622

Opinion

OPINION

RADER, Circuit Judge (Sitting by Designation).

In this patent ease, Loral Fairchild Corporation sued numerous Japanese electronics companies for infringement of United States Patents 3,931,674 (the ’674 patent) and 3,896,485 (the ’485 patent). In August 1995, the court severed and stayed all of the non-manufacturing consumer defendants, leaving six core groups of manufacturing defendants — Sony, Sanyo, Toshiba, Hitachi, NEC, and OKI. On October 11, 1995, the court ordered separate trials for each of the six groups of manufacturing defendants and scheduled trial dates for each.

Over the course of five weeks in January and February, 1996, Loral Fairchild Corporation (Loral) and Sony Electronics, Inc. and Sony Corporation (Sony) tried the issues of ownership, validity, and infringement of the two patents to a jury. On February 14,1996, the jury returned a verdict finding that Loral owns the patents, that Sony did not prove invalidity of the asserted claims, and that Loral had proven infringement. The jury also found that Sony induced infringement of the ’674 patent but did not induce infringement of the ’485 patent. After the verdict, Sony moved for judgment as a matter of law (JMOL) or for a new trial.

Loral and Sony then tried the remaining liability issues to the court on April 15 and 16,1996. This Opinion is the court’s decision on Sony’s post-trial motion as well as the court’s findings of fact and conclusions of law on prosecution history estoppel, equitable es-toppel, and laches.

Because no reasonable jury could find that Sony infringed the two patents, this court grants Sony’s motion for judgment as a matter of law. In addition, this court finds prosecution history estoppel prevents Loral from asserting the doctrine of equivalents to cover Sony’s accused products. However, this court finds that neither equitable estop-pel nor laches bar recovery in this suit.

BACKGROUND

This court has issued several opinions in this case which discuss the technology, the patents, and the claims. See Loral Fairchild Corp. v. Victor Co. of Japan, 911 F.Supp. 76 (E.D.N.Y.1996) (imposing discovery sanction); Loral Fairchild Corp. v. Victor Co. of Japan, 906 F.Supp. 813 (E.D.N.Y.1995) (granting defendants’ motion for summary judgment on marking); Loral Fairchild Corp. v. Victor Co. of Japan, 906 F.Supp. 798 (E.D.N.Y.1995) (construing disputed claims of the patents). This opinion builds on the foundation of those earlier decisions. This opinion assumes familiarity with those preceding decisions.

The ’485 Patent

Claim 1 of the ’485 patent reads:

Structure which comprises:

a. a light sensing element comprising a first region of semiconductor material overlaid by a first electrode separated from said semiconductor material by insulation, said light sensing element being capable of containing a charge packet;

b. an adjacent region of said semiconductor material disposed for receiving said charge packet from said light sensing element;

c. means for controlling the transfer of said charge packet from said light sensing element to said adjacent regions; and,

d. charge sink means having a contact for applying a bias thereto buried within said semiconductor material and disposed for receiving excess charge accumulated in said light sensing element, said charge sink means extending laterally from said contact toward said light sensing element while beneath the surface of said semiconductor material.

This court interpreted the disputed portions of this claim language. Loral, 906 F.Supp. at 807-12. Summarizing this court’s detailed claim construction, “charge sink means” describes a means-plus-function limitation. In accord with the specification, this court determined that the charge sink means is a region of semiconductor material doped opposite from its surrounding semiconductor material. The charge sink means does not include its associated depletion region. Thus, the charge sink means is a separate and distinct structure within the semiconductor substrate.

Claim 1 also provides a description of the location of this structure. The language identifies the “charge sink means” structure as “buried within said semi-conductor material.” Further the claim places this structure “beneath the surface of said semi-conductor material.” The “buried within” and “beneath the surface” language means the charge sink structure is surrounded by and submerged within the semiconductor material.

This court construed “contact” to mean one or more electrical paths for charge to flow to and from the charge sink means. Lastly, this court interpreted the phrase “extending laterally from said contact toward the light sensing element” as conveying more information about the location of the charge sink structure. This directional language requires the charge sink structure to stretch from the contact toward the light sensing element. Thus, this buried structure extends parallel to the semiconductor’s top surface and perpendicular from a contact toward the light sensing element. Figure 1 of the ’486 patent provides an excellent depiction of the location of the charge sink structure.

During the course of the claim construction trial, the court heard extensive testimony about the reasons for a “buried” charge sink means. Semiconductor devices sense light with a structure that converts light waves into electrical charges. The light sensing element in the device, however, has a certain charge capacity. Intense light can cause the element to exceed its capacity causing excess charge to spill out of the light sensing structure and flow into adjacent light sensing elements.. The result of this overflow is blooming — a spreading distortion of light on .the detected image. To prevent blooming, the chip includes a structure to capture and dissipate excess charge from the light sensing elements. These charge capturing devices prevent the excess charge from spilling over into adjacent light sensing elements. Initially, chip designers placed these charge drains next to the light sensing elements on the surface of the chips. See C.H. Sequin, Blooming Suppression in Charge Coupled Area Imaging Devices, The Bell System Technical Journal, October 1972; United States Patent 3,866,067. However, surface drains occupied valuable light sensing space on the surface of the chip. To achieve higher resolution in light detection, designers sought to remove charge drains from the chip surface leaving more room for light sensors. Early’s ’485 patent represents a step toward solving this problem by placing a drain structure below the surface of the chip to capture the excess charge below the light sensing element. [See Transcript of September 19, 1995, Hearing, at 32-33]. Consequently, the claims of the ’485 patent cover a structure entirely beneath the surface of the chip.

After construing the claims, the court invited the parties to renew motions for summary judgment on the asserted claims of the ’485 patent. The court granted Sony’s motion for summary judgment of no literal infringement. However, the court denied the motion on the issue of infringement under the doctrine of equivalents. Additionally, the court granted Sony’s motion for sanctions against Loral for discovery abuses. In its order, the court precluded Loral from deviating from its long-standing position that claims 7 and 8 describe only the operation of the structure in claim 1. Loral, 911 F.Supp. at 80. Under this ruling on the eve of trial, this court prevented Loral from asserting that claims 7 and 8 describe a “means” without many of the structural limitations in claim 1.

At trial the parties disputed whether Sony’s accused CCD devices satisfied elements (a) and (d) of claim 1 of the ’485 patent under the doctrine of equivalents. The parties also disputed whether Sony’s devices infringed the method claims 7 and 8 in the ’485 patent. The jury found that Loral had proven that Sony’s devices infringed all of the asserted claims.

The ’674 Patent

Loral asserts infringement of claim 1 of the ’674 patent, which describes a process for making a self-aligned, gapless charge-coupled device (CCD). Claim 1 reads:

1. A process for fabricating a charge coupled device structure in a semiconductor substrate, comprising the steps of

[1] selectively applying at least one layer of insulation material to said semiconductor substrate;

[2] selectively forming a plurality of spaced-apart first gate electrodes on the uppermost surface of said at least one layer of insulation material;

[3] forming a first insulation layer over said plurality of first gate electrodes;

[4] forming implanted barrier regions in said semiconductor substrate in the intervals between said plurality of spaced-apart first gate electrodes, the edges of said implanted barrier regions being aligned with the vertical edges of the insulation layer on the respective first gate electrodes;

[5] selectively forming a plurality of second gate electrodes on said uppermost surface of said at least one insulating layer between said plurality of spaced-apart first gate electrodes, each of said second gate electrodes substantially occupying the space between adjacent first gate electrode; and

[6] connecting each of said second gate electrodes to and individual adjacent first gate electrode to form a composite electrode for a charge coupled element.

United States Patent 3,931,674 (bracketed numbers added to facilitate reference).

As with the ’485 patent, this court construed the disputed portions of claim 1. Loral, 906 F.Supp. at 803-07. In particular, this court construed the claim to require formation of the insulation layer over the first gate electrodes before implantation of the barrier regions. In other words, the claim requires a process sequence with step three preceding step four. Id. at 806.

This court also invited the parties to renew motions for summary judgment on the ’674 patent. On December 14, 1995, this court ruled that elements. 1, 2, and 6 of claim 1 of the ’674 patent read on Sony’s processes. 1 On that same day, this court granted Sony’s motion for summary judgment of no literal infringement of claim 1 because its processes all use a naked gate as a mask. In other words, Sony’s process performs claimed step four before step three. However, this court denied all the motions for summary judgment on infringement under the doctrine of equivalents.

The case proceeded to trial on Loral’s claim that Sony’s processes infringe claim 1 of the ’674 patent under the doctrine of equivalents. At trial, Loral called several witnesses to testify about the insubstantiality of the changes between the claimed invention and Sony’s processes.

In addition to arguing that claim 1 is invalid, Sony contends its processes do not infringe claim 1 under the doctrine of equivalents. It, too, called several witnesses to testify about the substantiality of the differ- enees between its processes and the claimed process. Sony alleges that reversal of steps 3 and 4 can cause the barrier region to extend a significant distance under the first gate electrodes in the resulting CCD structure. Sony presented evidence at trial that this undercutting phenomenon can cause “glitches” in a chip’s operation. At the time of the patent, the use of an insulated gate as a mask, performing step 3 before 4, helped prevent undercutting. With the advance of technology in later years, Sony has developed others ways to prevent undercutting. 2

Sony also alleges its processes do not satisfy the fifth limitation of the claim, which requires formation of the second gate electrodes between the first gate electrodes on the upper surface of the same continuous insulation layer with the first gate electrodes. At trial, Sony offered transmission electron micrographs (TEMs) to demonstrate that its processes do not form the second gate electrodes exactly on the same surface as the first gate electrodes. [STX 12, 13, 14, and 15]. In its case-in-chief, Loral relied on scanning electron micrographs (SEMs) to show substantial compliance with that claim limitation. According to Sony’s expert, Mr. Denboer, however, SEMs do not sufficiently magnify the structure to reveal layer differentiation. [TR 2808-09].

In addition, Sony argues that prior art to the ’674 patent describe its processes and thus limits application of the doctrine of equivalents. In particular, Sony argues work at Hughes Aircraft Company by Drs. Erb and Su, which culminated in a December 3, 1973, article, describes Sony’s manufacturing processes. As such, Sony merely practices known prior art processes in the public domain to manufacture its accused CCDs. Sony called Drs. Erb and Su to testify about their work at Hughes during the relevant time period.

Prosecution History Estoppel

Beyond its arguments that the prior art limits the scope of the doctrine of equivalents and that its processes substantially differ from the claimed process, Sony argues the prosecution history prevents Loral from asserting infringement against Sony’s processes and CCDs. The parties filed pretrial motions on the issue of prosecution history estoppel. This court denied Sony’s motion and submitted the issue to the jury over the objection of both Loral and Sony. This court, however, advised the parties that it would treat the jury’s verdict on this issue as advisory only. Indeed this court devoted significant time to resolution of the prosecution history estoppel issue in a separate hearing and argument several months after the jury trial.

Loral argues claim 1 of the ’674 patent describes a process which implants barrier regions either before or after insulation of the first gate electrodes. Loral contends that the claim can cover a process which performs step four before step three. Sony contends Loral surrendered during prosecution any deviation from the process sequence.

The ’485 patent also presents prosecution history estoppel questions. Sony argues that the prosecution process confined the ’485 patent to a charge sink means buried within the substrate, not in contact with the surface, and extending laterally from the contact toward the light sensing element. Loral claims the prosecution history does not preclude Loral from asserting a range of equivalents that would include Sony’s substrate drain.

In this case, the trial record thoroughly examined the prosecution history of both the ’674 and ’485 patents. Beyond examination of the file wrapper, this court and the jury heard extensive inventor and expert testimony about the administrative acquisition of these patents. In its advisory capacity, the jury found that: (1) Sony did not prove by a preponderance of the evidence that Dr. Amelio surrendered or abandoned a 4r-be-fore-3 process sequence during acquisition of the ’674 patent; and (2) Sony did not prove by a preponderance of the evidence that Dr. Early abandoned or surrendered a charge sink that covers Sony’s substrate drain during the ’485 prosecution history.

The Equitable Defenses

Finally, Sony contends the equitable defenses of laches and estoppel bar recovery by Loral in this case. Sony had claimed unen-forceability for inequitable conduct but waived that defense before the trial on the equitable issues. The court heard testimony and received exhibits dining the course of a two-day trial in April, 1996.

DISCUSSION

This Opinion will first address Sony’s renewed motion for judgment as a matter of law. This analysis will include conclusions on prosecution history estoppel. The Opinion will then discuss this court’s findings and conclusions on the equitable issues presented in the bench trial.

Sony filed a renewed motion for judgment as a matter of law under Fed.R.Civ.P. 50(b). Sony challenges the jury’s verdict on the issues of ownership, infringement, and inducement to infringe. Curiously, Sony does not challenge the jury’s verdict on validity.

A Rule 50 motion tests the sufficiency of evidence in support of a jury verdict. Mi lbank, Tweed, Hadley & McCloy v. Boon, 13 F.3d 537, 542 (2d Cir.1994); accord Litton Sys., Inc. v. Honeywell, Inc., 87 F.3d 1559, 1566 (Fed.Cir.1996). In evaluating a Rule 50 motion, this court reviews the record to ascertain whether a reasonable jury could have reached the verdict in the case. Milbank, 13 F.3d at 542. Where substantial evidence supports its findings, this court must uphold the jury’s verdict. Goodwall Constr. Co. v. Beers Constr. Co., 991 F.2d 751, 754 (Fed.Cir.1993). Moreover, this court must view the evidence in the light most favorable to Loral, drawing reasonable inferences favorable to Loral, without determining credibility of witnesses, and without substituting its choice for that of the jury. Wang Labs, Inc. v. Toshiba Corp., 993 F.2d 858, 863 (Fed.Cir.1993). In sum, this court must deny a renewed JMOL motion unless the movant shows that no reasonable jury could have reached its result. Goodwall, 991 F.2d at 754. In determining that no reasonable jury could have reached its result, the evidence must be “so overwhelming that reasonable and fair minded persons could only have reached the opposite result.” Sorlucco v. New York City Police Dep’t, 971 F.2d 864, 871 (2d Cir.1992) (quoting Stubbs v. Dudley, 849 F.2d 83, 85 (2d Cir.1988), cert. denied, 489 U.S. 1034, 109 S.Ct. 1095, 103 L.Ed.2d 230 (1989)).

With this lofty standard in mind, this court discerns substantial evidence in support of the jury’s verdict on ownership. Loral called Steven Wahl, the general counsel involved in the asset transfers at issue, to testify about the various transfers of interests. Loral also offered the various agreements purporting to transfer the patents from company to company.

Based on the testimony and the document, a reasonable jury could have concluded that Loral Fairchild owned the patents in suit. Both inventors assigned the patents to Fair-child Camera & Instrument Corporation (FCIC). FCIC was renamed Fairchild Semiconductor Corporation (FSC). [TR 954-55, PTX 1037, PTX 41]. By written agreement and subsequent written assignment, FSC transferred title to the patents to Fairchild Weston Systems, Inc. (FWSI). Schedule A to those documents listed the patents. Loral Fairchild acquired the assets of FWSI. [TR 955-60, 1072-75, PTX 1037, PTX 41, PTX 70]. In 1989, FWSI separately assigned the patents to Loral Fairchild. [TR 960-61, PTX 1037, PTX 41].

Because neither party challenges the jury instructions, this court now turns to the core issue in this ease — infringement of the asserted claims of the two patents under the doctrine of equivalents.

Sony’s Motion for JMOL on the %85 Patent

Sony moves for judgment as a matter of law that its accused devices do not infringe the ’485 patent under the doctrine of equivalents. The jury returned a verdict finding that Sony had infringed the asserted claims of the ’485 patent under the doctrine of equivalents. Thus, the jury upheld Loral’s position that Sony’s Hole Accumulated Diode (HAD) sensor and substrate drain infringe elements (a) and (d) of claim 1 of the ’485 patent.

However, no reasonable jury could find Sony’s accused devices infringe any asserted claim of the ’485 patent. Infringement of a claim requires,, “without exception, that an accused device contain each limitation or its equivalent.” Athletic Alternatives, Inc. v. Prince Mfg., Inc., 73 F.3d 1573, 1582 (Fed.Cir.1996). The doctrine of equivalents grants no license to ignore claim limitations. Dolly, Inc. v. Spalding & Evenflo Cos., 16 F.3d 394, 397 (Fed.Cir.1994); accord Maxwell v. J. Baker, Inc., 86 F.3d 1098, 1105 (Fed.Cir.1996); see Texas Instruments Inc. v. United States Int’l Trade Comm’n, 988 F.2d 1165, 1171 (Fed.Cir.1993); see also Unique Concepts, Inc. v. Brown, 939 F.2d 1558, 1563 (Fed.Cir.1991) (“Our interpretation gives full effect to the recitation of two distinct elements in the claimed structure: linear border pieces and right angle comer border pieces.”). Sony’s devices do not contain a charge sink means as defined by the ’485 claims.

Claim 1 of the ’485 patent defines the charge sink means as a distinct structure of semiconductor material doped opposite from its surrounding semiconductor material. The claim also locates that structure within the semiconductor material. According to the claim, the charge sink means must extend parallel to the top surface of the semiconductor and perpendicular to a contact toward a light sensing element. Sony’s accused devices contain no such structure as a matter of law. See Pennwalt Corp. v. Durand-Wayland, Inc., 833 F.2d 931, 935 (Fed.Cir.1987) (en banc), cert. denied, 485 U.S. 961, 108 S.Ct. 1226, 99 L.Ed.2d 426 (1988); Dolly, 16 F.3d at 398.

In a Sony device, the entire substrate functions as a charge sink — a so-called substrate drain. Because Sony’s charge sink structure is the entire substrate, it has no structure distinct from the surrounding substrate. Without this structure, nothing is doped opposite from the surrounding semiconductor material. Nothing is “buried within” the semiconductor material. Nothing extends from the contact toward the light sensing element beneath the surface of the semiconductor material. Moreover, Sony’s substrate is in contact with a surface of the semiconductor (indeed it forms at least one surface) and thus cannot be “buried within” the semiconductor material. In sum, nothing meets element d of claim 1 of the ’485 patent. Loral’s effort to write this limitation out of the claim by way 6f the doctrine of equivalents violates the all'elements rule.

Figure 1 of the ’485 patent — relied upon extensively by both parties to describe element d of claim 1 — depicts well the charge sink defined by the claims. This figure shows an elliptical region (14) with an N + doping, opposite from the surrounding P material (11, 12). This elliptical structure extends from the contact (16) toward the light sensing element (28a). The Sony accused devices have no such structure. Instead the substrate beneath the light sensor, represented by the P region (11,12) in Figure 1 of the ’485 patent, functions to remove excess charge from the light sensing region and prevent blooming. While looking to Figure 1 to enlighten the meaning of the claim terms, this court recognizes that this drawing represents only one embodiment of the claims.

Attempting to identify a charge sink means within the Sony device, Loral arbitrarily draws an ellipse in a region of semiconductor material under the Sony fight sensing element. This ellipse is not a distinct structure in a Sony device at all, but merely ink marks on a graphic representation of the Sony device. In the actual Sony devices, no distinct structure separate from the surrounding semiconductor material exists at the location Loral highlights. Thus, the Sony device lacks a vital limitation in claim 1 of the ’485 patent. Because Sony does not make, use, or sell the structure in claim 1, it also cannot infringe claims 7 and 8. These claims describe only methods of operating the structure in claim 1.

In addition, this court concludes that the ’485'prosecution history estops Loral from asserting that Sony’s substrate drain satisfies element d of claim 1 under the doctrine of equivalents. Claim 1 of the ’485 defines the “charge sink” structure as disposed in a specifically defined location. The inventor added these limitations defining the charge sink structure and its location in response to examiner rejections in light of prior art. Loral, 906 F.Supp. at 811.

When first filed, element (d) of the application claim read:

charge sink means buried within said semiconductor material and disposed in proximity to said light sensing element for receiving excess charge accumulated in said light sensing element.

[STX 504 at E18]. The examiner objected to this language because it did not recite the location of the charge sink region other than within the semiconductor material and close to the light sensor. This description read on the prior art which included semiconductors with charge sinks located on the top surface of the device. [STX 504 at E24],

In response to the rejection, Loral amended element d to read:

charge sink means buried within said semiconductor material and disposed for receiving excess charge accumulated in said light sensing element, said charge sink means being located beneath and not in contact with the surface of said semiconductor material.

[STX 504 at E29]. In another amendment Loral amended element d to read:

charge sink means buried within said semiconductor material and disposed for receiving excess charge accumulated in said fight sensing element, said charge sink means extending laterally toward said fight sensing element while beneath the surface of said semiconductor [sic] material.

[STX 504 at E35]. The examiner required further structural definition of the location of the charge sink structure before allowing the claims to issue in the ’485 patent. In particular, the examiner required the claim to specify a “contact” and to define the spatial orientation of the charge sink structure in relation to the contact. [STX 504 at E40]. These additional requirements from the examiner came in a telephone conference with Loral’s attorney. [Id. ] Loral’s attorney agreed to these amendments to secure issuance of the ’485.

To overcome an examiner’s rejection based on prior art, Loral changed the scope of its claim. See Pall Corp. v. Micron Separations, Inc., 66 F.3d 1211, 1219 (Fed.Cir.1995). This court finds that Loral narrowed its claims to avoid rejections based on prior art and to obtain the ’485 patent. Thus, Loral relinquished during prosecution a range of equivalents which might embrace more than the specific structure described in the claim language. In other words, a charge sink only infringes if it is a separate structure-buried within the substrate, extending laterally from a contact toward the light sensor, and avoiding all contact with the surface. Loral cannot recapture under the doctrine of equivalents claim scope which might encompass the Sony device.

In addition to the missing charge sink structure, Sony’s semiconductors also contain no distinct contact structure. In its claim interpretation opinion, this court defined the term “contact” as an electrical path or paths. Loral 906 F.Supp. at 811-12. Within the meaning of the claims, the term “contact” describes a path formed by conductive structure extending from a surface of the semiconductor to make ohmic contact with the charge sink structure. As the language of the claim indicates, this structure’s perpendicular extension to a surface of the semiconductor provides a reference point for the lateral extension of the charge sink structure toward the fight sensor.

As with the charge sink means itself, Loral attempts to manufacture a contact structure in the Sony device by drawing potential contact fines from the imaginary charge sink ellipse on a diagram of a Sony device. As aptly noted by Sony counsel at oral argument on this motion, Loral’s attempt to create structure in Sony’s substrate resembles the argument that a chair exists in a block of metal because one can draw the outline of the chair on one face of the block. Loral then draws an infinite number of contacts running downward from the imaginary ellipse toward the bottom of the substrate. These imaginary fines are not contact structures as defined by the ’485 claims. The ’485 claims require a defined electrical structure to reach from a surface to the charge sink structure. ■ This electrical structure in the claims also serves to define the location of the charge sink, which stretches from that structure toward the light sensor. No such electrical structures exist in the Sony device. The Sony device uses the entire substrate as both the charge sink and the contact. Once again, the Sony device simply does not include the “contact” structure limitation in the ’485 claims.

In sum, the accused Sony devices contain no distinct charge sink or contact structures in their substrate drain. The substrate of the Sony device is simply not a charge sink structure nor a contact structure as defined in the claims. These limitations missing from the Sony devices prevent infringement under the doctrine of equivalents as a matter of law. Dolly, 16 F.3d at 400; Maxwell, 86 F.3d at 1105. Consequently, this court will grant Sony’s motion for judgment as a matter of law of noninfiingement of the asserted claims of the ’485 patent.

Although the absence of a charge sink means alone precludes infringement, this court also addresses Sony’s HAD sensor and element a of claim 1. Claim 1 of the ’485 patent requires a “light sensing element.” The claim further describes this element as a MOS (metal-oxide-semiconductor) structure “capable of containing a charge packet.” The claim describes the MOS structure in specific terms: “a first region of-semiconductor material overlaid by a first electrode separated from said semiconductor material by insulation.” An external voltage or potential enables the MOS structure to contain a charge packet.

In a Sony device, the light sensing element is not a MOS structure. Instead the Sony CCDs have a photodiode sensor. This type of sensor, called a HAD sensor, is a hole accumulation layer within the semiconductor material. Nothing in the Sony device is an electrode separated from the top of the semiconductor material by insulation. The HAD sensor lacks an electrode and lies within the semiconductor material, not on top. Moreover, nothing in the Sony sensor requires the application of an external voltage to contain a charge packet.

As with the charge sink means structure, no fight sensor structure in the Sony device corresponds to the MOS structure on top of the semiconductor material — specific requirements of the ’485 claims. “[E]quivalency cannot embrace a structure that is specifically excluded from the scope of the claims.” Dolly, 16 F.3d at 400. Therefore, as a matter of law, the Sony devices do not satisfy element a of claim 1 under the doctrine of equivalents.

Beyond the problem of a missing structural claim element, no reasonable jury could apply the doctrine of equivalents to find that the HAD sensor satisfies the fight sensor requirement of the ’485 patent. In the first place, as mentioned, the structure of the HAD sensor varies substantially from the claimed structure. In addition, these structural differences cause the HAD sensor to work in a different way to reach a different result. The HAD sensor works in a different way because it requires no external bias. It also achieves a different result because without an electrode to block some fight, the Sony sensor is more sensitive to blue frequencies of fight.

To show infringement by equivalents, Loral produced, among other evidence, statements from Sony that its HAD sensor works as a “virtual gate.” [PTX 98, TR 3795]. In the circumstances of this case, Sony’s rhetoric, designed to draw broad analogies to reliable MOS technology, has no legal significance. Sony’s rhetoric does not admit that the HAD sensor possesses the structure required by the claims of the ’485 patent. Moreover, in the face of the overwhelming evidence of the structure of Sony’s HAD sensor, this rhetoric cannot substitute for the undisputable evidence of the structure of Sony’s HAD sensor and the requirements of claim 1 of the ’485 patent. Therefore, this evidence does not alter the legal conclusion that the fight sensor defined in claim 1 of the ’485 patent cannot have an equivalent in the Sony CCDs.

In conclusion, primarily because Sony’s devices clearly and unequivocally lack a charge sink structure, this court grants Sony’s mo tion on that sufficient ground. That single, vital missing claim limitation precludes infringement by equivalents as a matter of law. In addition, this court concludes, based on similar reasoning, that Sony’s devices lack a contact structure and a light sensor structure as defined in the claims. As a matter of law, the ’485 patent has no equivalent in the Sony CCD devices. Because the court previously ruled that all the asserted claims of the ’485 patent rise or fall with infringement of claim 1, no reasonable jury could find infringement of any of the asserted claims of the ’485 patent.

Sony’s Motion for JMOL on the ’674, Patent

The jury returned a verdict finding that Sony had infringed claim 1 of the ’674 patent under the doctrine of equivalents. The jury also reported that Sony did not prove prosecution history estoppel and that Loral was not asserting infringement of a process covered by the prior art.

The issue of infringement of claim 1 of the ’674 patent turns entirely on the prior art status of a December 2, 1973 article by Dr. Darrel Erb and the work leading up to that article. The jury found in a special verdict that neither the article nor the antecedent work was in the prior art. Without the article or the work in the prior art, substantial evidence supports the jury’s verdict of infringement under the doctrine of equivalents. However, if the Erb article or the preceding work is prior art to the ’674 patent, no reasonable jury could find infringement of claim 1 of the ’674 patent.

Several facts are not in dispute. Dr. Gilbert Amelio filed the ’674 patent on February 8, 1974. Dr. Erb published his article more than two months earlier in the International Electron Devices Meeting Technical Digest.

The Erb Article recites the following process:

The lightly doped p layer is formed by implanting boron or by epitaxy. After regrowing the gate oxide, a layer of silicon nitride is deposited. This is followed by the deposition and etch of the polycrystalline silicon storage electrodés. To make the device operate as a two phase CCD the wafer is then implanted with an n-type impurity to partially compensate the p-type layer in the areas which are not masked by the silicon gate electrodes. The device is then oxidized to form the insulator over the polycrytalline silicon electrodes. The nitride, which is exposed in the interelectrode gaps, resists oxidation and does not get thicker. Finally, a layer of aluminum is deposited and etched giving the transfer electrodes.

[STX 322]. This passage recites the process in claim 1 of the ’674 patent, with the exception of a reversal of steps 3 and 4 of the claimed process. In addition to this close recitation of the process claimed in the ’674 patent, the Erb article included six figures showing a cross section of the CCD device produced by the Erb method and some oscilloscope readings from the Erb CCD in operation.

Overwhelming evidence showed conclusively that the Erb article describes or, at least, renders obvious Sony’s accused processes. With the Erb article in the prior art, no reasonable jury could find that Sony infringes claim 1 of the ’674 patent.

On its face the document discloses a process for making a two-phase, buried channel CCD shift register using a naked gate as a mask. Testimony at trial showed conclusively that the document describes the significant process steps used to make the Sony process and would thus at a minimum render obvious the Sony processes. [TR 2341-44, 2608, 3160, 3214]. As explained by Dr. Robert Bower, Sony’s expert witness, by process step numbers, the Erb article describes a process sequence of 1, 2, 4, 3, 5, and 6. [TR 3160]. Sony’s processes follow that very same sequence. [TR 3160-61]. Moreover, one of ordinary skill in the art using the Erb article would know how to use the process to make CCDs. Thus, the article describes the significant process steps used to make Sony’s CCD devices.

Loral attempts to argue that the Erb article is nonenabling and therefore ineffective as prior art for purposes of the doctrine of equivalents. A non-enabling document, however, still qualifies as prior art for obviousness purposes. Symbol Technologies, Inc. v. Opticon, Inc., 935 F.2d 1569, 1578 (Fed.Cir.1991).

As such, under the Federal Circuit’s decision in Wilson Sporting Goods Co. v. David Geoffrey & Assoc., 904 F.2d 677, 683-85 (Fed.Cir.), cert. denied, 498 U.S. 992, 111 S.Ct. 537, 112 L.Ed.2d 547 (1990), the Erb article would (if it is prior art), as a matter of law, bar Loral’s claim of infringement of the ’674 patent against Sony’s processes. See Conroy v. Reebok Int'l, Ltd., 14 F.3d 1570, 1576 (Fed.Cir.1994). Loral may not assert a range of equivalents which embraces prior art. Because Sony’s process replicates the steps of the Erb article, Loral could not assert the doctrine of equivalents to cover the Sony process because that accused process antedated the claims. Consequently, with Erb in the prior art no reasonable jury could find infringement of claim 1 of the ’674 patent by equivalents. See E.I. du Pont de Nemours & Co. v. Phillips Petroleum Co., 849 F.2d 1430, 1436-37 (Fed.Cir.), cert. denied, 488 U.S. 986, 109 S.Ct. 542, 102 L.Ed.2d 572 (1988).

Interestingly, Loral also argues that because the Erb article describes mainly structure, it would not render obvious a process to make that structure. This argument, however, runs counter to Loral’s position on inducement to infringe. In support of the jury’s verdict of inducement of claim 1 of the ’674 patent, Loral strenuously urges that Sony reverse engineered or copied a Loral chip. That copying, Loral argues, supports the jury’s verdict that Sony induced infringement of claim 1 of the ’674 patent, which describes a process to make a CCD. Thus, Loral’s inducement argument admits that one may determine the process used to make a chip by looking at the resulting structure. Moreover, testimony from both sides clearly indicated that the structure of a chip may reveal the significant process steps used to make that chip. [See TR 1380-83, 2350],

Erb Article as Prior Art—Section 102(a)

The issue then becomes whether the Erb article or the work at Hughes culminating in the Erb article constitutes prior art for application of the doctrine of equivalents. As a publication, 35 U.S.C. § 102(a) (1994) makes the Erb article prior art for assessing infringement under the doctrine of equivalents. As the patentee, Loral bears the burden of persuasion on all issues relating to infringement under the doctrine of equivalents. At trial, Loral sought to show that Amelio invented the subject matter of the ’674 patent before publication of the Erb article. Thus, the critical date which Loral must antedate is December 3,1973.

As recently clarified by the Federal Circuit, Loral’s attempt to antedate a publication must satisfy priority rules encountered primarily in the interference context. Mahurkar v. C.R. Bard, Inc., 79 F.3d 1572, 1576-77 (Fed.Cir.1996). In Mahurkar, the Federal Circuit stated the proof required to antedate a reference. An inventor may gain priority by showing that he reduced the invention to practice before the publication date of the reference. Id. at 1577. Alternatively, the inventor may gain priority by showing his conception of the invention before publication of the reference, coupled with reasonable diligence from a date before publication to reduction to practice. Id. at 1577-78. Here, the patentee, Loral, bears the burden of persuasion on the status of the Erb article and the Erb/Su work as prior art by a preponderance of the evidence. Wilson, 904 F.2d at 685.

The initial priority inquiry is reduction to practice. To show actual reduction to practice, an inventor must demonstrate that the invention works for its intended purpose. Scott v. Finney, 34 F.3d 1058, 1061 (Fed.Cir.1994). Depending on the invention and the problem it solves, test results may be necessary to show the invention worked for its intended purpose. Mahurkar, 79 F.3d at 1578. As noted by Scott and Mahurkar, actual reduction to practice for a complex invention such as the CCD process in this case requires test results that demonstrate the processes will work for their intended purpose. Scott, 34 F.3d at 1062, Mahurkar, 79 F.3d at 1578; Koval v. Bodenschatz, 463 F.2d 442, 447 (CCPA 1972) (“[T]here must be a functional relationship between the test conditions and the intended functional setting.”); see also Holmwood v. Sugavanam, 948 F.2d 1236, 1238-40 (Fed.Cir.1991); DSL Dynamic Sciences Ltd. v. Union Switch & Signal, Inc., 928 F.2d 1122, 1125 (Fed.Cir.1991); Elmore v. Schmitt, 278 F.2d 510, 513 (CCPA 1960).

Loral contends Dr. Amelio reduced to practice the process claimed in the ’674 patent at least as early as September 1973. Loral points to a mask overlay set, a run sheet, and testimony from Dr. David Wen as evidence of actual reduction to practice. Dr. Wen explained that the mask overlays were a set of transparencies or stencils used to “define the actual features that you will put on a piece of silicon to manufacture ... a CCD device.” [TR 1747]. Each sheet or transparency corresponds to a different process step. [TR 1748]. He explained that the particular set of transparencies, dated September 14, 1973, corresponded to a linear array which “basically follows the Amelio process.” Id. (emphasis added). According to Dr. Wen, these stencils suggest Fairchild had completed the design of the device and were about to begin manufacturing. [TR 1749],

Dr. Wen explained that the run sheets identify specific process steps to make a CCD. [TR 1750-53]. The particular run sheets offered by Loral had several months identified on them. However, none of the sheets identified the year. Dr. Wen believed they were generated in the “1974 time frame.” [TR 1751]. He based this belief on the fact that one particular document “describes a process sequence for one of the experiments that ... was run at that time. This particular experiment is to produce some surface channel CCD devices.” [TR 1751]. Apparently, Dr. Wen ran some experiments in March 1975 on a CCD produced in late 1974. [TR 1752-53],

This evidence does not corroborate Loral’s allegation that Dr. Amelio had reduced his invention to practice by September 1973. At best, as recognized by Dr. Wen, the mask overlay set demonstrates Fairchild researchers had performed some design work. The overlays do not show that Dr. Amelio had run the process claimed in the ’674 patent and knew it would work for its intended purpose.

The run sheets show nothing about activities occurring at Fairchild in September 1973. As Dr. Wen testified, the run sheets detail process steps performed “in the 1974 time frame” — too late to antedate Dr. Erb’s activities. Moreover, Dr. Wen explained that the run sheets refer to experiments with a surface channel CCD. [TR 1752]. The process claimed in the ’674 patent, however, relates to production of a buried channel CCD. Consequently, the run sheets bear no clear connection to the- mask overlay set. Even if the run sheets depict performance of the process steps in the mask overlay set, Fairchild did not run the process steps until 1974. Dr. Amelio could not have known the process would work for its intended purpose until after he ran the process and tested CCDs produced by that process. In simple terms,- the mask overlay set and run sheets do not supply enough evidence to permit a reasonable jury to find actual reduction to practice as early as September 1973.

In an effort to bolster its case, Loral also notes that on September 18,1973, Dr. Kim, a Fairchild researcher, gave a presentation at a conference in San Diego, California at which he discussed the process claimed in the ’674 patent. [PTX 1057]. Again, however, nothing in the record about this speech suggests Dr. Amelio had run the process and had tested CCDs produced from that process. In sum, Dr. Kim’s speech does not prove that Dr. Amelio knew the process would work for its intended purpose.

Loral also points to a proposal submitted by Fairchild to the United States Air Force in November 1973. Fairchild sought funding from the government to develop a CCD device for the Government. [TR 1764], Fair-child indicates in the proposal that it would use its isoplanar gapless buried channel technology. Fairchild also represents to the Air Force that as of November 1973, “Fairchild’s Isoplanar gapless buried channel technology ... is already showing reliable performance and high yield.” [TR1758-59, 1763-65, PTX 16 at § 4, p. 4-5]. According to Dr. Wen, that process is the same as the process claimed in the ’674 patent. [TR 1764]. The proposal goes on to discuss transfer efficiency, a key parameter in CCD devices. [PTX 16 at § 4, p. 20]. Again, Fairchild represents to the Air Force that it has solved the transfer efficiency problem encountered with surface channel technology through the use of its buried channel technology. [Id. ].

The Air Force proposal contains only con-clusory statements about yield and performance of a process to make LARAM devices. No process steps, no data, no test parame-' ters, no test results, and no dates appear in the proposal. Moreover, nothing in the proposal suggests Dr. Amelio ran tests in September 1973. The conclusory statements in the Air Force proposal do not corroborate that Dr. Amelio had run the process and had tested the CCDs produced by the process. In sum, the Fairchild proposal does not show, with the degree of specificity required for reduction to practice in a priority contest, that Dr. Amelio knew that the process would work for its intended purpose.

Loral also points to a newspaper article and data sheet regarding the CCD 110, a 256 element linear CCD device. [PTX 127 and 128]. The March 1974 data sheets describe the characteristics of a recently developed device for potential customers. [TR 1766], According to Dr. Wen, Fairchild would have needed two months of testing on a working device to generate the information in the data sheets. [TR 1767]. Fairchild then announced availability in á publication of the 256 element linear array. [PTX 128, TR 1768-69]. Again, this information does not corroborate any testing from September 1973 that would permit the trier of fact to find reduction to practice at that time.

In sum, Loral does not present sufficient evidence of actual reduction to practice before September 1973. Although presenting circumstantial evidence of activities at Fair-child, none of this evidence shows that Dr. Amelio had actually tested a CCD device made by the claimed process to verify its workability. Loral’s evidence lacks specifics about test results or about time frame or about the types of processes producing the CCDs. When measuring reduction to practice in a priority contest, this specific evidence is critical. Naber v. Cricchi, 567 F.2d 382, 386 (CCPA 1977) (explaining that evidence must show “specific acts at specific times directed at reduction to practice”), cert. denied, 439 U.S. 826, 99 S.Ct. 98, 58 L.Ed.2d 119 (1978); Litchfield v. Eigen, 535 F.2d 72, 76-77 (CCPA 1976); National Tractor Pullers Ass’n v. Watkins, 205 USPQ 892, 912 (N.D.Ill.1980). Without these specifics, Loral does not supply enough evidence to support a showing of actual reduction to practice before September 1973.

Without evidence of an actual reduction to practice, Loral can rely upon the alternative course of conception and diligence to antedate the Erb reference. Conception requires that an inventor have formed in his or her mind a definite and permanent idea of the complete and operative invention. Burroughs Wellcome Co. v. Barr Labs., Inc., 40 F.3d 1223, 1228 (Fed.Cir.1994), cert. denied, — U.S. -, 116 S.Ct. 771, 133 L.Ed.2d 724 (1996). The idea must be so clearly defined in the inventor’s mind that only ordinary skill, without extensive research or experimentation, would be necessary to reduce the invention to practice. Id.

According to Loral, Amelio conceived of the subject matter of the ’674 patent no later than July 1973. Loral presents no contemporary evidence of conception directly from Dr. Amelio, but instead supplies entries in another researcher’s laboratory notebook and a memorandum written by that same Fairchild researcher to Dr. Amelio. First, Loral cites the. laboratory notebook with Dr. Pao’s signature and a date of July 3, 1973. [PTX 15]. Dr. Pao worked at Fair-child Camera and Instrument Corporation during the relevant period under Dr. Amelio.

On the excerpted pages, Dr. Pao lists twenty-two process steps, which include those claimed in the ’674 patent. Dr. Pao also states that implantation of the barrier regions may be performed either before or after oxidation of the first gate electrodes. After listing the process steps, Dr. Pao depicts a CCD that would result from performance of the listed process.

Loral also points to a memorandum from Dr. Pao to Dr. Amelio dated July 13, 1973. In that memorandum, Dr. Pao proposes a process to make a particular device using the process claimed in the ’674 patent. [PTX 17]. Loral offered Dr. Amelio’s deposition testimony during the course of trial. In his deposition for this case, Dr. Amelio testified that the process sequence described in the Pao notebook was the same as his process. [TR 1660]. He further testified that he conceived of implanting the barrier regions either before or after oxidation at the time of the entries in Dr. Pao’s notebook. [TR 1662]. According to Dr. Amelio, Fairchild experimented with both techniques in the 1971-75 time frame. [TR 1662],

Finally, Loral cites an entry in Dr. Choong-Ki Kim’s notebook dated October 5, 1973. [PTX 129], After describing a problem encountered by Fairchild, Dr. Kim states: “A solution to this problem was invented by Gilbert F. Amelio.” He then lists a series of process steps, which include those claimed in the ’674 patent.

This evidence provides substantial support for a finding that Amelio conceived of the subject matter of the ’674 patent at least as early as July 1973. While Loral does not supply documentary evidence from Amelio, the circumstantial evidence repeatedly refers to the inventor and repeats his invention, thus corroborating that Dr. Amelio was in possession of the process steps in his claimed invention in July 1973. For instance, Dr. Pao recites Dr. Amelio’s process in July 1973. Dr. Kim’s notebook entry a few months later identifies the process as Amelio’s. A jury could reasonably find from this evidence that Amelio conceived the subject matter of the ’674 patent at least as early as that time period. ■

Although Loral can supply evidence to support conception before publication of the Erb article, that' showing is only a part of the equation. Loral must also show reasonable diligence from a time just before publication of the Erb article through the filing date of the patent. See Griffith v. Kanamaru, 816 F.2d 624, 625-26 (Fed.Cir.1987). An inventor must provide specific details on activity during the critical period. General testimony that the inventor worked continuously and diligently will not suffice. Moreover, the inventor must corroborate evidence of reasonable diligence. Gould v. Schawlow, 363 F.2d 908, 919 (CCPA 1966).

Once again, Loral offers no reliable evidence at all—let alone documentary evidence during the relevant period—from Dr. Amelio himself about his diligence. Instead, to show Amelio’s reasonable diligence from late 1973 through the filing date of the patent, Loral points to the March 1974 data sheet and the March 1974 newspaper article announcing the availability of commercial CCDs made by the Amelio process. This evidence is insufficient as a matter of law. Neither the article nor the data sheet detail any activity during the critical period. While Dr. Wen explained that the data sheet could not have been prepared without several months of reliable testing, such generalized statements do not satisfy the stringent diligence requirements. Naber, 567 F.2d at 386, Gould, 363 F.2d at 919; Ireland v. Smith, 97 F.2d 95, 100 (CCPA 1938) (“We may surmise that appellant was probably diligent ... but mere surmise cannot take the place of proof.”). At no point does Loral present signed laboratory notebooks from Dr. Amelio describing his continual, uninterrupted diligence. At no point does Loral present a series of test results or activities showing continual, uninterrupted diligence. At no point does Loral present Dr. Amelio to testify of his diligence, which would still, of course, require further corroboration. Gould, 363 F.2d at 919-20; accord Holmwood, 948 F.2d at 1239. In fact, other than Dr. Wen’s testimony about what must have been occurring, Loral presents absolutely no evidence at all of any activities occurring at Fairchild between the publication date of the Erb article and the filing date of the ’674 patent.

In sum, based on Loral’s evidence, no reasonable jury could find Amelio proceeded with reasonable diligence from late 1973 to his reduction to practice. To antedate the Erb reference, Loral would need to show Amelio’s diligence from before Erb’s publication until his constructive reduction to practice upon filing his patent application. Without evidence of diligence following publication, Dr. Amelio cannot antedate Dr. Erb’s article. In re Mulder, 716 F.2d 1542, 1544-45 (Fed.Cir.1983).

While presenting its story about development of the process claimed in the ’674 patent, Loral suffers from a lack of evidence to support that story. The law on priority regards oral testimony with great skepticism. Mahurkar, 79 F.3d at 1577 (collecting cases). For that reason, a party asserting priority must supply corroborating evidence. Id. More than twenty years have passed since the events of invention and patent filing. No doubt many documents that would support Loral’s story have been lost. Without that critical evidence, however, the law leaves the parties with the filing date of the ’674 patent and the publication date of the Erb article. Consequently, this court concludes the Erb article is prior art as a matter of law for purposes of determining infringement under the doctrine of equivalents.

This court reaches this conclusion despite the brief period for which Dr. Amelio must show diligence. Dr. Erb published his article on December 1973. Dr. Amelio filed his patent application on February 8, 1974. Although Loral presents its story that Dr. Amelio must have been busy during that brief period, the law requires more than a story; it requires relevant evidence with corroboration. In the absence of that evidence, this court must conclude that the law prevents Dr. Amelio from showing reasonable diligence from before Dr. Erb’s publication until constructive reduction to practice.

Erb/Su Work as Prior Art—Section 102(g)

As noted above, because Dr. Amelio cannot antedate Dr. Erb’s December 1973 publication, the Erb article stands as prior art under section 102(a) and precludes infringement as a matter of law. In the alternative, however, his court also finds no reasonable jury could conclude the Erb/Su work is not prior art to the ’674 patent for assessing infringement under the doctrine of equivalents. The work of Drs. Erb and Su culminating in the December 1973 article qualifies as prior art under 35 U.S.C. § 102(g) (1994). This work antedates Dr. Amelio’s inventive activities.

Drs. Erb and Su had conceived of the process described in claim 1 of the ’674 patent by November 7, 1972, and reduced it to practice no later than August 1973. Moreover, their publication of that work in December 1973 obviates any credible claim of abandonment, suppression, or concealment. As such, the Erb/Su work qualifies as prior art under section 102(g).

Applying the reduction to practice rules noted above, Sony states that Erb and Su reduced their idea to practice no later than August 1973. Dr. Erb testified that the conference which published his paper would only accept articles reflecting successful experiments on a working device. Indeed the Erb article includes three oscilloscope printouts, which show output from a working CCD produced with the process described in the paper (with implantation of the barrier regions before insulation of the first gate electrodes). An entry from Dr. Erb’s notebook indicated a September 20 deadline for submission of conference papers. To meet the September 20 deadline, Drs. Erb and Su testified that they would have completed the successful tests on the working device in August 1973. Sony presented no further evidence for its August 1973 reduction to practice.

Sony presented no documentary evidence from the July to August 1973 time period to corroborate its August 1973 actual reduction to practice. As explained above, priority rules require corroboration of an actual reduction to practice. Hahn v. Wong, 892 F.2d 1028, 1032-33 (Fed.Cir.1989); see Gortatowsky v. Anwar, 442 F.2d 970, 972 (CCPA 1971). Other than Erb and Su, the “inventors,” no one testified regarding events at Hughes during July and August 1973. At most, their testimony works backward from the December 1973 article to suggest that inventive activity must have been underway in July and August. This type of evidence is insufficient as a matter of law. Dr. Erb conceded that no notebook entries reflect work on a CCD in that time frame. [TR 2752],

For its part, Loral contends Erb and Su finalized their work on the December article only after Erb heard Dr. Kim’s presentation at the San Diego conference. 3 As acknowledged by Dr. Erb, his notebooks contain his sketchy notes of Dr. Kim’s lecture at the San Diego conference. [PTX 1057]. After returning from this conference, Dr. Erb calculated a way to implant phosphorous — a calculation necessary to design the CCD structure in the December 1973 paper. In addition, Dr. Erb cites Dr. Kim’s San Diego paper as a reference in the back of his own December paper. [TR 2753-54], This evidence supports Loral’s position that Erb completed his article in a grace period after the September conference. Moreover, as Loral showed, before the San-Diego conference Dr. Erb primarily worked with four-phase CCD devices. [PTX 709, TR 3890-91],

Drawing all inferences in Loral’s favor, this court concludes that no reasonable jury could find that Erb and Su actually reduced to practice the process described in the December paper as early as August 1973. At-the same time, Loral concedes Dr. Erb must have mailed the article to the conference within two to three weeks of the San Diego conference. Therefore, before that point, Drs. Erb and Su must have tested the working device because the article includes those test results in the form of the oscilloscope readings. These test results show an actual reduction to practice. Thus, Drs. Erb and Su must have reduced to practice the process described in the paper by the end of September 1973. The data in Dr. Erb’s notebook immediately following the San Diego- conference supports this date for actual reduction to practice. [PTX 703],

In any event, the record shows that Drs. Erb and Su actually reduced their process to practice at the end of September 1973. This process, except for reversal of process steps 3 and 4, replicates the process claimed in the ’674 patent. Drs. Erb and Su described implantation of the barrier regions before insulating the first gate electrodes. Their September 1973 reduction to practice antedates Dr. Amelio’s constructive reduction to practice in February 1974. Because Dr. Amelio cannot show reasonable diligence from before this date to his filing date, the Erb/Su work stands as prior art. Moreover, as this court will now examine, Drs. Erb and Su can show a conception which antedates Dr. Amelio’s failed attempts to prove conception and diligence.

Applying the same conception rules recited above, Sony argues Dr. Steven Su (a listed co-author of the Erb article) conceived of the idea described in the December 1973 article on November 7, 1972. Unlike the dearth of evidence from Dr. Amelio, Dr. Su testified at trial and produced a remarkably detailed, contemporary laboratory notebook showing a buried channel device. [PTX 705 at 91-93]. Extensive testimony at trial verified that one of skill could deduce the process of manufacture from the CCD cross sections and other information in Dr. Su’s notebook. 4

Loral challenged the timing of the November 1972 entry with an allegation that Su added key information to the November 7 entry at a later date. This charge came from Loral’s expert, Ms. Manoliu, who offered her opinion that entries in a different color of ink showed that Dr. Su must have doctored his notebook. [TR 3883], Dr. Su, however, categorically denied adding the information at a later date. [TR 2584, 2594-95]. Indeed, corroborating that testimony, the notebook shows that Dr. Su consistently used different colors of ink throughout his well-organized and tidy notebook. Moreover, Loral contends that the alleged doctoring attempts to hide that Dr. Su only conceived of a stepped oxide semiconductor device. To the contrary, the companion drawing in Dr. Su’s notebook shows the first and second gate electrodes on the same surface, which verifies Dr. Su’s understanding of the same surface concept. [PTX 705].

In addition, Dr. Erb’s notebook contains a November 14,1972 entry expressly recognizing Dr. Su’s work with buried channel technology. [PTX 700 at 87]. This entry in Dr. Erb’s notebook adds support-to the timing of Dr. Su’s November entry. The weight and veracity of the notebook evidence, even accounting for the sole opinion about doctoring, would compel a reasonable jury to find that Dr. Su conceived the Erb/Su invention no later than November 1972. Thus, as a matter of law, Drs. Erb and Su conceived their invention before Dr. Amelio.

Because Drs. Erb and Su were both first to conceive and first to reduce to practice their invention, their work is prior art for application of the doctrine of equivalents to Sony devices. See Steinberg v. Seitz, 517 F.2d 1359, 1364 (CCPA 1975) (noting that diligence is only relevant where the first to conceive is also the second to reduce to practice). Thus, no reasonable jury could find that the Erb/Su work was not prior art. Finally, as noted earlier, publication of the paper precludes any finding that Erb and Su abandoned, suppressed, or concealed their work.

In summary, the record shows that Erb and Su conceived of their invention in November 1972. They then proceeded to reduce it to practice in late September 1973. Because the most favorable view of the evidence sets Dr. Amelio’s conception in July 1973 and his reduction to practice (constructive) in February 1974, the Erb/Su work qualifies as prior art. With the Erb/Su work as prior art, no reasonable jury could find infringement under the doctrine of equivalents. The record shows that Sony practiced the process disclosed in the Erb prior art, namely implantation of barrier regions before insulation of the first gate electrodes. Because the doctrine of equivalents cannot extend exclusive patent rights to encompass art already in the public domain, Wilson, 904 F.2d at 683-84, the law will not permit a jury to find that Sony infringed the ’674 patent by using the Erb process.

Putting aside the issue of sufficiency of the evidence, the jury made a finding that the Erb article was not prior art.. The closing argument may help explain why the jury made a finding contrary to the law. During closing argument, Loral argued that Dr. Erb had abandoned his inventive work. Although the court barred Loral’s witness, Mr. Cecil Quillen, from testifying to the issue of abandonment, Loral’s counsel during his reply closing argument argued strenuously that Hughes and Dr. Erb had abandoned the work published in Dr. Erb’s paper. Counsel argued:

The Erb article was not published until some three months after [Dr.] Kim[’s] public disclosure of the Amelio process and Hughes abandoned the Erb/Su work. If you look at PX 1055, it was abandoned. What did they do? They went to a stepped oxide.

If Erb and Su ever had a self-aligned implanted barrier, it was abandoned in favor of something else.

[TR 4274]. Loral apparently based this abandonment argument on Hughes’ decision to pursue other technology or Dr. Erb’s decision to publish, rather than seek a patent for, his invention. This argument, to which Sony had no opportunity to respond, is incorrect as a matter of law. International Glass Co. v. United States, 408 F.2d 395, 403 (Ct.Cl.1969) (per curiam). Dr. Erb did not abandon his invention by publication. Inventors may reap their public reward for innovation as they see fit, either by publication or by seeking exclusive legal rights. Nonetheless, because the jury instructions mentioned abandonment in its recitation of priority rules, the jury may well have seized this incorrect ar gument (bolstered by Sony’s inability to respond) to deny Dr. Erb’s work its proper-prior art status.

The court also notes that the parties offered extensive evidence at trial on the same surface limitation in claim 1 of the ’674 patent, including TEMs and SEMs. However, in the light of the court’s claim construction on this limitation and the evidence offered by Loral during trial, the court finds substantial evidence supports the jury’s verdict on this limitation.

Sony has moved for a new trial in the alternative. Fed.R.CivJ?. 59 gives this court authority to grant new trials on all or part of the issues tried to a jury “for any of the reasons for which new trials have heretofore been granted in actions at law in the courts of the United States.” Under this rule, a court may grant a new trial where the verdict is against the great weight of the evidence. Byrd v. Blue Ridge Rural Elec. Coop., Inc., 356 U.S. 525, 540, 78 S.Ct. 893, 902, 2 L.Ed.2d 953 (1958). Even where substantial evidence might support a jury’s verdict, a court may grant a new trial. Litton, 87 F.3d at 1576; Song v. Ives Labs., Inc., 957 F.2d 1041, 1047 (2d Cir.1992); Dyer v. MacDougatt, 201 F.2d 265, 272 (2d Cir.1952) (Hand, J.). As the United States Court of Appeals for the Second Circuit has expressly recognized, Rule 59 authorizes this court to sit as the thirteenth juror in deciding a motion for a new trial. Dyer, 201 F.2d at 272. In other words, this court may weigh the evidence and need not view it in the light most favorable to the verdict winner. Song, 957 F.2d at 1047. Because the record lacks sufficient evidence to support the jury’s verdict, the court, in the alternative, grants Sony’s motion for a new trial on infringement of the asserted claims of the ’674 and ’485 patents under the doctrine of equivalents.

For the reasons stated above, with the Erb article as prior art, no reasonable jury could find that Sony’s processes infringe claim 1 of the ’674 patent under the doctrine of equivalents. The court will therefore grant Sony’s motion for judgment as a matter of law on this issue. In the alternative, the court grants Sony’s motion for a new trial because the great weight of the evidence is against the verdict. Finally, because the court .finds no infringement of claim 1 of the ’674 patent as a matter of law, the court also reverses the jury’s verdict finding inducement to infringe of this claim. Without direct infringement, a party cannot be liable for inducement to infringe. MetCoil Sys. Corp. v. Korners Unlimited, Inc., 803 F.2d 684, 687 (Fed.Cir.1986).

Prosecution History Estoppel on ’674

As indicated, this court submitted the legal issue of prosecution history estoppel to the jury for an advisory verdict. The jury determined that the prosecution history did not limit application of the doctrine of equivalents. Because prosecution history estoppel is a question of law, LaBounty Mfg., Inc. v. United States Int’l Trade Comm’n, 867 F.2d 1572, 1576 (Fed.Cir.1989), this court applies independently the law to the file wrapper evidence in this case.

Prosecution history estoppel places a limitation on the permissible scope of equivalents. Charles Greiner, 962 F.2d at 1036. An applicant’s- statements during prosecution can estop the patentee from later reclaiming as an equivalent coverage surrendered to acquire the patent. Pall, 66 F.3d at 1218; see Wang Labs., 993 F.2d at 868. Stated otherwise, a patentee cannot “recapture through equivalence ... coverage given up [by argument or amendment] during prosecution.” Hormone Research Found., Inc. v. Genentech, Inc., 904 F.2d 1558, 1564 (Fed.Cir.1990) (quoting Loctite, 781 F.2d at 870).

This estoppel limits the application of the doctrine of equivalents in two primary circumstances. In the classic format, estop-pel arises “when a change of claim scope is made in order to overcome an examiner’s rejection based on prior art.” Pall, 66 F.3d at 1219. In addition, the estoppel may arise (without a rejection based on prior art) when an', applicant unmistakably and clearly disavows claim coverage during prosecution in an effort to obtain the patent. See Southwall Technologies, Inc. v. Cardinal IG Co., 54 F.3d 1570, 1580 (Fed.Cir.), cert. denied, — U.S. -, 116 S.Ct. 515, 133 L.Ed.2d 424 (1995). In this latter type of estoppel, how ever, this disavowal must be unequivocal and unmistakable:

While it is true that a patentee’s unmistakable assertions to the PTO in support of patentability, whether or not they were truly required to secure allowance of the claim, will estop the patentee from obtaining protection of the subject matter surrendered thereby, ... it is difficult to conclude, in light of the truly equivocal nature of the claim phrase [at issue] and the conflicting evidence in the prosecution history, that [the patentee] surrendered protection ... with the clarity that prosecution history estoppel requires.

Athletic Alternatives, 73 F.3d at 1582 (emphasis added); accord Texas Instruments, Inc. v. United States Int’l Trade Comm’n, 988 F.2d 1165, 1174 (Fed.Cir.1998); Southwall, 54 F.3d at 1581.

The prosecution history estoppel issue in this case once again focuses on the sequence of the ’674 process steps. As construed by this court, the ’674 patent discloses an express sequence to make a CCD. Under the ’674 process, an insulated gate serves as a mask during implantation of the barriers. Thus, step three—insulation (or oxidation) of the gate—precedes step four—implantation of the barriers. The Sony process implants the barrier before insulating the gate. Sony thus uses step four before step three. Or, in other words, Sony uses a naked gate (not an insulated gate) as a mask for implanting the barriers. Thus, for this issue, the question is whether Amelio relinquished during prosecution patent coverage for naked gate processes in a manner sufficient to create an estop-pel.

With those basic propositions in mind, this court examines the administrative process that yielded the ’674 patent. As originally filed, claim 1 of the ’674 patent read, in pertinent part:

1. A process for fabricating a semiconductor structure comprising the steps of: selectively forming a plurality of spaeed-apart first gate electrodes;

selectively forming a second gate electrodes between and spaced slightly apart from adjacent ones of said first gate electrodes; and forming barrier regions in a semiconductor substrate beneath said second gate electrodes, said barrier regions extending in said substrate beyond a boundary defined by the outer perimeter of said second gate electrodes toward but not reaching an outer boundary defined by the outer perimeter of said first gate electrodes, thereby providing barrier regions having outer boundaries precisely aligned between the boundaries of said first and said second gate electrodes.

Claim 2, as originally filed, read:

2. The process as recited in claim 1 and further comprising the step of forming a first insulation layer over said first gate electrodes prior to the step of forming barrier regions in said semiconductor substrate, whereby the outer lateral surfaces of said first insulation layer are aligned •with the corresponding surface of said barrier regions.

As originally filed, application claim 1 provided no explicit order or sequence of process steps. The language of this original claim covered both naked and insulated gate masking processes. In fact, original application claim 2 specified the sequence of insulating the electrode “prior to” formation of barrier regions in the semiconductor substrate. To underscore that claim 1 included naked gate processes, the latter, more specific, dependent claim 2 expressly covered insulated gate processes.

After filing the application claim 1 without a sequence, the applicant then filed a preliminary amendment, before an examiner’s review, cancelling claims 1-10 and adding claims 17-27. Amended claims 17 and 18 read:

17. A process for fabricating a charge coupled device structure in a semiconductor substrate, comprising the steps of:

selectively applying at least on layer of insulation material to said semiconductor substrateQ]

selectively forming a plurality of spaced-apart first gate electrodes on the uppermost surface of said at least one layer of insulation material;

forming implanted barrier regions in said semiconductor substrate in the intervals between said plurality of spaeed-apart first gate electrodes, the edges of said implanted barrier regions being aligned with the respective first gate electrodes; and selectively forming a plurality of second gate electrodes on said uppermost surface of said at least one insulating layer between said plurality of spaced-apart first gate electrodes, one of said second gate electrodes substantially occupying the space between each of said plurality of first gate electrodes.

18. The process as recited in claim 17 further comprising the step of forming a first insulation layer over said plurality of first gate electrodes prior to the step of forming implanted barrier regions in said semiconductor substrate, whereby the edges of said implanted barrier regions are aligned with the vertical edges of the first insulation layer on the respective first gate electrodes.

In making these amendments the applicant stated in the remarks section, that these changes were made “in order to ensure that they particularly point out and distinctly claim the invention taught by the application.” Application claim 17 still recited a naked gate process because it envisioned formation of barrier regions “in intervals between ... spaced-apart first gate electrodes, the edges of said implanted barrier regions being aligned with the respective first gate electrodes.” Moreover application claim 18 specified an insulated gate process as an additional separate claim. Thus, in its early iterations, Amelio’s claims seemed to claim both a naked gate and an insulated gate process.

The examiner rejected claims 17 and 18 as vague and indefinite under 35 U.S.C. § 112 and as obvious under 35 U.S.C. § 103. In addition, the examiner rejected claim 17 as being anticipated under 35 U.S.C. § 102. The examiner explained that claims 17-20 were “unpatentable over Walden [United States Patent 3,852,799] in view of Boleky [United States Patent 3,745,647].” The examiner explained: “It is considered obvious that the oxide surrounding Walden’s first electrodes 46, 49 etc. may be used as [a] mask in forming his regions 43, 45 ete[.], as in Boleky at 20.” In addition, the examiner explained the- reasons for the section 112 rejection:

In claim 17, the edges of the [implanted barrier] regions are not aligned with the first gate electrodes but are aligned with the oxide surface covering the first gate electrodes and it is not clear that only one second electrode occupies the spaces between all of the first gate electrodes. Also, no apparent relationship has been set forth between the two sets of gate electrodes and the barrier region.

Thus, the PTO rejected Amelio’s claims as unpatentable over prior art. In response to this Office Action, the applicant amended claim 17 and cancelled claim 18. Specifically, the applicant narrowed claim 17 to cover only insulated gate masking:

17. (amended) A process for fabricating a charge coupled device structure in a semiconductor substrate, comprising the steps of

selectively apply at least one layer of insulation material to said semiconductor substrate;

selectively forming a plurality of spaced-apart first gate electrodes on the uppermost surface of said at least one layer of insulation material;

forming a first insulation layer over said plurality of first gate electrodes;

forming implanted barrier regions in said semiconductor substrate in the intervals between said plurality of spaced-apart first gate electrodes, the edges of said implanted barrier regions being aligned with the vertical edges of the insulation layer on the respective first gate electrodes; and selectively forming a plurality of second gate electrodes on said uppermost surface of said at least one insulating layer between said plurality of spaced-apart first gate electrodes, one of said second gate electrodes substantially occupying the space between each of said plurality of first gate electrodes[.], each of said second gate electrodes comprising in combination with an individual adjacent first gate elec- trade a composite electrode for a charge coupled element.

(bracketed material deleted, underlined material added). In the Remarks section of the amendment, the applicant explained his effort to overcome the obviousness rejection:

It is submitted that the thrust of the Bo-leky reference is the lateral inward diffusion of the dopant impurities as shown by the points of penetration 40. The use of the insulated gate as a mask shows a technique that is generally known in the semiconductor art. The present invention is directed to a process sequence which includes this masking feature as one step but which further includes the unique and distinctive steps of connecting a single first gate electrode to a single adjacent second gate electrode to form a composite gate electrode. Thus, not only is self-alignment of the implanted barriers obtained, but a [composite] electrode for inclusion in a charge-coupled element is formed. This sequence of steps goes far beyond the teaching of Boleky and, therefore, is not rendered obvious by it. (emphasis added)

Thus, when the PTO rejected claim 17, Amelio changed claim 17 to specify only the use of insulated gates as a mask. [STX 499 at A55]. As amended, claim 17 covered only “forming a first insulation layer over said plurality of first gate electrodes” followed by a step of “forming implanted barrier regions ..., the edges of said implanted barrier regions being aligned with the vertical edges of the insulation layer on the respective first gate electrodes.” In remarks, the applicant characterized claim 17 as “directed to a process sequence” which included the specific “masking feature” of an insulated gate as “one step.” Moreover, the applicant closes his explanation of his amendment by clarifying: “this sequence of steps [insulation before implantation] goes far beyond the teaching of Boleky [the prior art invoked by the PTO] and, therefore, is not rendered obvious by it.” (emphasis added).

In other words, the applicant expressly narrowed his' claim to a process sequence of step 3 before step 4 (insulated gate masking) in response to a PTO prior art rejection. Moreover, the applicant narrowed the scope of his claims to abandon naked gate masking. Thus, the applicant invoked the classic prosecution history estoppel rule precluding recapture of abandoned claim scope.

Loral now protests that the applicant did not amend his claims to overcomé prior art rejections, but only to overcome the examiner’s § 112 rejection. The applicant’s double reference to the process sequence and direct reference to the Boleky prior art belies this protestation. Moreover another related incident also reveals the applicant’s motivation to overcome prior art during prosecution. While prosecuting the ’674 patent, Fairchild, Loral’s predecessor, proseeutéd as well a divisional apparatus application based on the application that ripened into the ’674 patent. In that divisional application, Fairchild used claim language which encompassed not only structures made by the claimed insulated gate processes but also structures made by naked gate processes. The examiner rejected these apparatus claims as anticipated by the Erb article. 5

Despite this rejection of the apparatus application as anticipated by Erb, Fairchild did not cite the Erb article to the' examiner for the related process application. Because the only difference between the Erb process and the ’674 process is naked as opposed to insulated gate masking, the applicant’s failure to cite Erb in the ’674 examination suggests that Fairchild relied during prosecution on that difference to avoid citing Erb as relevant prior art. Thus, in light of the duty to disclose material information during proseeu tion, see Norton v. Curtiss, 433 F.2d 779, 793-94 (CCPA 1970), a detailed examination of the public prosecution record shows that Fairchild relied again on the process sequence (step 3 before 4) to avoid potentially crippling prior art.

In sum, the applicant undeniably changed the original claims to exclude naked gate processes and specifically include only insulated gate processes. This change came in response to a rejection based on prior art. When explaining these changes, the applicant expressly relied upon a process sequence to distinguish over prior art. With this narrowing of claim scope, the applicant persuaded the PTO to issue a patent. After thus abandoning processes with implantation before insulation. Loral cannot now recapture that claim scope as an equivalent.

In the alternative, this court examines Loral’s contention that Amelio’s attorney amended the claims solely to address the section 112 rejection and not to distinguish over the prior art. At most this argument would remove Loral from the classic prosecution history estoppel category and require a showing of unmistakable surrender or abandonment of the naked gate processes. An examination of the administrative record, however, discloses sufficient clarity and un-mistakability to invoke an estoppel under Loral’s alternative argument as well.

The claims themselves show an unmistakable change in scope. Amelio did narrow his claims from language that included naked gate masking to language that clearly precluded implantation before insulation. Moreover, Amelio’s attorney explained unmistakably that this change claimed only a particular chronological sequence of steps. In this context, the related divisional apparatus application shows that Fairchild relied more than once on that sequence in an effort to obtain the ’674 patent. Under this alternative form of prosecution history estoppel, Loral also cannot recapture claim scope unmistakably disavowed during prosecution.

In sum, the public may justifiably rely on the prosecution history to interpret the scope of claim language. See Hoganas AB v. Dresser Indus., Inc., 9 F.3d 948, 952 (Fed.Cir.1993). Where an applicant has surrendered claim scope to obtain exclusive rights, the public may rely on that abandonment. See id. Thus, this court concludes that the prosecution history prevents Loral from asserting infringement of claim 1 of the ’674 patent under the doctrine of equivalents.

Sony’s Defense of Laches

Having resolved the issues tried to the jury, this court turns next to resolution of the equitable issues tried to the bench over the course of two days in April 1996. As noted above, Sony contends laches and equitable estoppel bar Loral’s recovery in this case. Equitable estoppel and laches fall within the sound discretion of the trial court. Aukerman Co. v. R.L. Chaides Constr. Co., 960 F.2d 1020, 1041 (Fed.Cir.1992) (en banc); Jamesbury Corp. v. Litton Indus. Prods., Inc., 839 F.2d 1544, 1553 (Fed.Cir.), cert. denied, 488 U.S. 828, 109 S.Ct. 80, 102 L.Ed.2d 57 (1988). Both defenses ultimately depend on underlying factual determinations. Hemstreet v. Computer Entry Systems Corp., 972 F.2d 1290, 1292 (Fed.Cir.1992).

The defense of laches arises when a patentee “neglect[s] or delay[s] ... bringing suit to remedy an alleged wrong, which taken together with lapse of time and other circumstances, causes prejudice to the adverse party and operates as an equitable bar.” Gasser Chair Co., Inc. v. Infanti Chair Mfg., Corp., 60 F.3d 770, 773 (Fed.Cir.1995) (quoting Aukerman, 960 F.2d at 1028-29). To show laches, the defendant must prove by a preponderance of the evidence: (1) unreasonable and inexcusable delay in filing suit, and (2) material prejudice resulting from the delay. Aukerman, 960 F.2d at 1032-33. A rebuttable presumption of laches arises in patent infringement cases for delays longer than six years. Id. at 1034. Nonetheless, as an equitable doctrine, laches does not operate on mechanical rules. Id. at 1032 (internal quotations and citations omitted). The reasonableness of a delay in bringing suit depends on the facts and circumstances of each case. Gasser, 60 F.3d at 773; Aukerman, 960 F.2d at 1032; Meyers, 912 F.2d at 1462-63.

In determining when the clock begins ticking for laches, this court must consider when the patent owner knew or should have known of defendant’s alleged infringement. Jamesbury, 839 F.2d at 1552. Beginning in 1979, Sony had a license to practice the ’674 process. Therefore, when the license expired in July 1985, Fairchild knew or should have known of Sony’s continued use of the ’674 process. The record corroborates that conclusion. Dr. Amelio’s deposition testimony, for- instance, revealed his perception of the number of companies infringing his process in 1982: “My impression was that virtually everyone was using the process by that time.” [STX 2012], Furthermore, Dr. Wen, Director of Engineering for Loral Fairchild, and Dr. Dyck, Fairchild Camera and Instrument’s CCD Division, also testified about their suspicions of widespread infringement. [Wen Dep. 1/27/93 at 142; Dyck Dep. 12/3/93 at 517-18].

During this court’s bench trial, Sony sought to show that Loral knew or should have known of Sony’s alleged infringement as early as Í982. Loral responded that its separate divisions—Fairchild Camera and Instrument Corporation in Palo Alto, California, and Fairchild Weston Systems in New York—did not share confidential information. Accordingly, Loral argues that the patent holder, Fairchild Weston, did not know or have a reason to know of the infringement. 6

This debate, however, was largely irrelevant because the ’674 is a process patent. 7 Before enactment of the 1988 Process Patent Amendment Act. Process Patent Amendment Act of 1988, Pub.L. No. 100-418, § 9006(a), 102 Stat. 1107, 1567 (1988), the Patent Act provided no remedy against a manufacturer who used a patented process outside the United States to fabricate a product for sale within the United States. Therefore, before the 1989 effective date of the 1988 Act, a patent owner could not seek damages or injunctions under the Patent Act for extraterritorial use of processes patented in the U.S.

Before 1989, the U.S.Code only offered a very limited opportunity for relief to a paten-tee against importation of products using a patented process. This limited opportunity was a Section 337 action against the alleged infringer in the U.S. International Trade Commission. See 19 U.S.C. §§ 1337 and 1337(a). Despite the grave limitations of this purported remedy, Sony contends that Fair-child’s decision to forego a Section 337 action after expiration of its license agreement in July 1985 gives rise to laches.

The inadequacy of a pre-1989 Section 337 remedy belie Sony’s contention. See S.Rep. No. 83, 100th Cong., 1st Sess. 30 (1987); H.R.Rep. No. 60, 100th Cong., 1st Sess. 5 (1987) (Congress found existing remedies available through the International Trade Commission (ITC), 19 U.S.C. § 1337 (1988), inadequate because, among other reasons, the ITC provides only nonmonetary relief). Simply to convince the ITC to initiate a section 337 action, a domestic industry had to make difficult showings. First, the commission required the domestic industry to manufacture or have manufacture-related activity under the patented process. Loral had very limited manufacturing capacity from 1985-89. In fact, the record shows that Loral solicited Sony to produce CCDs for Loral’s defense contracts because Loral lacked the capacity to meet this demand on its own. Loral’s inability to support the growing camcorder market in this era would have likely been a significant hurdle to initiation of a Section 337 action after 1985.

Another requirement for initiation of a 337 action was proof of economic injury to a domestic industry. Once again, Loral’s very limited manufacturing capacity militated against a showing on this factor. Loral was not even in competition with Sony in the 1985-89 era. Rather Loral sought to become a customer of Sony. Thus, Loral would have faced a very difficult, if not impossible, burden to show injury to a domestic industry.

Moreover, even if Loral had undertaken the expense and difficulty of overcoming the manufacturing-capacity and injury-to-domestic-industry hurdles, these showings only were relevant to instituting a section 337 action. Winning the action would require additional expense and legal effort. At the conclusion of what might be an expensive twelve- to eighteen-month effort, Loral still would gain no money damages from its Section 337 case. Moreover a 337 proceeding has no res judicata effect. [Transcript of Trial, April 16, 1996, at 504]. In determining a laches defense, a court must consider and weigh any justification offered by the plaintiff as justification for delay in bringing its suit. Aukerman, 960 F.2d at 1033.

In sum, Loral acted reasonably in its decision to forego instituting an expensive section 337 action. Because Fairchild was not in competition with Sony’s color camcorder business, it faced unlikely prospects of success in a Section 337 action, which might have yielded only limited remedies at best. Moreover, because of these limitations, Sony felt no reluctance to step up its CCD production and importation. The record shows that Sony felt no deterrent threat of a Section 337 action. Therefore, this court determines that Fairchild’s decision to forego its pursuit of remedies for process patent infringement under Section 337 was neither unreasonable nor inexcusable.

Within two years of the effective date of the 1988 Act, Loral initiated this suit. Under the equitable circumstances of this case, that period was not an unreasonable delay. For instance, Loral bought Fairchild in 1989. Loral took only a reasonable time to ascertain which Sony products infringed Loral’s newly acquired patents. With the wide variety of potential products and the difficulty of ascertaining infringing activity, two years was a reasonable period for Loral’s investigation.

The circumstances of this case show no unreasonable delay in Loral’s initiation of its patent infringement action. From 1985 until 1989, Loral had only the limited Section 337 action available. Loral’s failure to initiate a Section 337 action was not an unreasonable nor unexeusable delay. After 1989, Loral proceeded in a reasonable fashion to initiate its suit. Thus, the circumstances of Loral’s conduct do not warrant invocation of the equitable defense of laches.

The second factor, material prejudice resulting from patentee’s delay, “may be either economic or evidentiary.” Aukerman, 960 F.2d at 1033. The change in economic position must occur during the period of delay. Id. “It is not enough that the alleged in-fringer changed his position—i.e., invested in production of the allegedly infringing device. The change must be because and as a result of the delay.” Hemstreet, 972 F.2d at 1294.

In this case, this court has found no unreasonable delay. However, even if this court had determined that Loral unreasonably delayed by foregoing a Section 337 remedy, that period of delay would cover only 1985-89. During that period, Sony suffered little, if any, economic damage. Sony bought Fair-child’s Japanese semiconductor facility in 1987. [STX 2123A]. Not until 1989 did Sony increase its investment in CCD production by expanding the entire Kokobu production line and by starting production at its Nagasaki plant. [STX 710]. By that time, Loral had undertaken its infringement investigation under the 1988 Act. During any alleged delay period, Sony suffered little, if any, economic prejudice.

Moreover, even if Sony has suffered some evidentiary prejudice, Aukerman gives the trial court wide discretion in applying laches. As the Aukerman court explained, a court need not find laches in every case where a defendant proves undue delay and prejudice. 960 F.2d at 1036. That court went on to explain: “[L]aches is not established by undue delay and prejudice. Those factors merely lay the foundation for the trial court’s exercise of discretion.” Id. This court may, and does, deny this defense because evidence of other factors would make it inequitable to recognize the defense. See id. at 1036.

Sony’s Defense of Equitable Estoppel

In addition to laches, Sony contends equitable estoppel bars Loral’s suit. The equitable defense of estoppel may entirely bar an infringement suit. Equitable estoppel requires proof of three elements: (1) the- patent holder—through misleading conduct, statements, or inaction—leads the alleged infringer to reasonably infer that it does not intend to enforce the patent against the alleged infringer; (2) the alleged infringer relies on the misleading conduct; and (3) the reasonable reliance materially prejudices the alleged infringer. ABB Robotics, Inc. v. GMFanuc Robotics Corp., 52 F.3d 1062, 1063 (Fed.Cir.), cert. denied, — U.S. -, 116 S.Ct. 306, 133 L.Ed.2d 211 (1995). 'Unlike laches, this defense involves no presumption. Therefore, a party relying on an estoppel defense must prove each element. Aukerman, 960 F.2d at 1043.

The alleged estoppel in this case arose from a series of negotiations over licenses for patents involving semiconductor inventions. In 1979, Sony entered a licensing agreement with Loral’s predecessor that covered many semiconductor inventions, including the ’674 patent. This license lapsed in July 1985. Accordingly, Sony and Loral’s predecessor entered negotiations over the extension of the agreement. In September 1985, Sony sent a letter to Fairchild noting the lapse of their patent license agreement and requesting an extension. Fairchild responded: “We are willing to extend the 1979 License Agreement and await your specific proposal.”

This response alone, however, did not indicate acceptance of Sony’s extension proposal. To the contrary, Fairchild simply indicated willingness to consider alternatives for extension. The parties had reached no firm agreement to continue their licensing relationship, let alone the specifies of such a contractual agreement. Fairchild’s terse response could not mislead Sony to believe Fairchild had accepted an extension proposal. To this point, Fairchild had misled no one. Moreover Sony had no reasonable basis to esteem itself exempt from legal action for patent infringement.

Nonetheless Sony responded with optimism: “Thank you for your letter dated October 4, 1985 accepting our proposal of the extension of the subject of the license agreement.” Despite the optimistic tone of this letter, it did not show reasonable reliance on misleading Fairchild conduct. It simply offered a optimistic prospect of a continuing licensing relationship. Fairchild did not respond to Sony’s optimistic response. At no time did Fairchild confirm its intention to accept a formal Sony licensing extension proposal.

Mr. Taira was the Sony representative conducting the license extension negotiations with Fairchild. Mr. Taira died a year after the license expired and before Loral initiated suit. To determine the equitable estoppel issue, however, this court need-not determine Mr. Taira’s state of mind. The documentary evidence alone reveals Fairchild’s willingness to hear proposals, not its assent to any formal extension of the licensing agreement.

In July 1986, after Mr. Taira’s death, Sony again wrote Fairchild noting Fairchild’s failure to respond to Mr. Taira’s letter. Fair-child responded: ‘We will shortly advise you of our position as to extending the Agreement.” Fairchild did not communicate with Sony any further about the extension proposal. In other words, Fairchild’s last communication with Sony reiterated that the parties had reached no licensing agreement. Indeed, Fairchild expressly asserted that it had not agreed to any extension of the 1979 agreement.

In weighing the first element of equitable estoppel, this court must examine whether Fairchild’s silence or inaction amounts to misleading conduct. To establish the first element of equitable estoppel, Sony must show that Fairchild’s statements and conduct supported an “inference that [Fairchild] did not intend” to bring an infringement claim against Sony. Meyers v. Asics Corp., 974 F.2d 1304, 1308 (Fed.Cir.1992) (quoting Aukerman, 960 F.2d at 1042). At most Sony can show silence in the face of Sony’s offers to extend the licensing agreement.

The Federal Circuit has discussed examples in which silence combined with other facts does or does not mislead an alleged infringer to reasonably infer that the paten-tee had foregone pursuit of its patent rights against an alleged infringer. In Meyers v. Brooks Shoe, Inc., the Federal Circuit stated: “[W]e do not believe that a suggestion of infringement coupled with an offer to license followed by silence would suffice to establish equitable estoppel.” 912 F.2d 1459, 1464 (Fed.Cir.1990); see MCV, Inc. v. King-See-ley Thermos Co., 870 F.2d 1568, 1572 (Fed.Cir.1989) (patentee expressly declined to pursue an infringement claim), Jamesbwry, 839 F.2d at 1555 (plaintiff had eight year silence following a letter informing alleged infringer he was infringing plaintiffs patent, and another letter informing him that the matter had been referred to his attorneys); cf. Hottel Corp. v. Seaman Corp., 833 F.2d 1570, 1574-75 (Fed.Cir.1987) (plaintiff’s period of silence did not follow any communication indicating it would take immediate action).

As these Federal Circuit decisions show, silence alone does not amount to misleading conduct. As the Federal Circuit stated in Aukerman, inaction — “combined with other facts respecting the relationship or contacts between the parties” — may allow defendant to make the inference that the claim against him has been abandoned. 960 F.2d at 1042; accord ABB Robotics, 52 F.3d at 1064. Thus, silence alone — even silence over an extended period of time — does not amount to misleading conduct without notice of infringement or failed licensing negotiations. Aukerman, 960 F.2d at 1042 (citing Continental Coatings Corp. v. Metco, Inc., 464 F.2d 1375, 1379-80 (7th Cir.1972) (“In each such case the infringement notice was either withdrawn or followed by such a long period of inactivity as to justify an inference of abandonment.”); A.C. Aukerman Co. v. Miller Formless Co., 693 F.2d 697, 701 (7th Cir.1982) (“With a delay of this length [over six years], there has to be something significantly more than continued jousting, at least where there is no real possibility of success on the part of patentee in obtaining the solid type of license it wants.”); Jensen v. Western Irrigation & Mfg., Inc., 650 F.2d 165, 169 (9th Cir.1980) (threats of enforcement followed by long period of silence)). In other words, the active conduct of patentee, beyond mere silence, must also support an inference that patentee did not intend to bring an infringement action against the alleged in-fringer.

Loral, in this ease, has not engaged in conduct which would mislead Sony to infer reasonably that Loral had abandoned any intent to initiate an infringement action. The record shows only an ongoing exchange of proposals to propose an extension of licensing negotiations. See Hemstreet, 972 F.2d at 1295. Thus, the character of the communications between Fairchild and Sony did not amount to misleading conduct. Sony thus had no basis for a reasonable inference that Fairchild had abandoned intentions to use legal remedies.

Moreover the period of silence which constitutes Sony’s primary argument for misleading was not so prolonged as to create a misleading impression. In this case, Loral’s silence after ambiguous communications lasted about five years. Generally, the longer the delay the stronger the inference becomes. See Aukerman, 960 F.2d at 1044. However, because equitable estoppel is a defense in equity each case varies on its facts as to what constitutes a misleading inference. Therefore, the length of periods of silence, combined with other facts, varies in determining an inference of misleading silence. Studiengesellschaft Kohle, m.b.H. v. Dart Indus., Inc., 726 F.2d 724, 728-29 (Fed.Cir.1984) (finding that five-year silence alone did not give rise to estoppel); Hemstreet, 972 F.2d at 1295 (noting that negotiations plus six years of silence did not justify a finding of equitable estoppel); Advanced Hydraulics, Inc. v. Otis Elevator Co., 525 F.2d 477, 481 (7th Cir.), cert. denied, 428 U.S. 869, 96 S.Ct. 132, 46 L.Ed.2d 99 (1975) (explaining that notice of infringement plus five-year silence required finding of estoppel). Therefore, this court finds that Fairchild’s conduct did not rise to the level of misleading conduct.

Fairchild’s failure to sue Sony for infringement of its foreign patents also does not justify any reasonable reliance by Sony. Fairchild owned patents in Britain, France, and Hong Kong, based on the ’674 patent. These foreign patents contained apparatus claims. Failure or refusal to sue on foreign patents has no relevance to the right to sue on United States patents, which are limited to acts of infringement in the United States. This case simply does not feature misleading conduct.

Under the second element of equitable es-toppel, the accused infringer must show its substantial reliance on the patentee’s misleading conduct. Reliance and prejudice or harm are not the same. To demonstrate reliance, “the infringer must have had a relationship or communication with the plaintiff which lulls the infringer into a sense of secu-rity_” in progressing with its business. In other words, the harm must be in direct reliance on patentee’s conduct. See Aukerman, 960 F.2d at 1043. Although the evidence indicates that Loral was aware that Sony was probably infringing the ’674 patent. Loral had no effective remedy until passage of the 1988 Process Amendments Act. Fair-child’s failure to warn Sony earlier of possible infringement does not rise to the level of misleading conduct. Therefore, Sony cannot show reasonable reliance on this conduct.

Lastly, the accused infringer must show that its reliance caused it material prejudice. As with laches,- prejudice may take the form of either a change in economic position or a loss of evidence. Id. (citing Advanced Hydraulics, 525 F.2d at 481-82). However, without misleading conduct and reasonable reliance, this court finds that Sony did not show any material prejudice from misleading conduct. The Federal Circuit in Aukerman instructed the trial courts to take into consideration, even where the. three elements of equitable estoppel are present, other evidence and facts affecting the equities of the parties in exercising its discretion to allow the equitable estoppel defense. Id. at 1043. Because Fairchild had no viable cause of action until passage of the 1988 Process Amendments Act, this court denies Sony’s request to bar this suit under equitable es-toppel.

CONCLUSION

The court grants Sony’s motion for judgment as a matter of law as to infringement of the asserted claims of the ’485 and ’674 patents. With that ruling, the court also grants Sony’s motion for judgment as a matter of law as to inducement to infringe of claim 1 of the ’674 patents. The court denies Sony’s motion on all other grounds. Additionally, the court finds prosecution history estoppel precludes infringement of the asserted claims of both the ’485 and ’674 patents under the doctrine of equivalents. Finally, the court finds that neither laches nor equitable estop-pel bar recovery in this suit. In the alternative, the court grants Sony’s motion for a new trial on infringement of the asserted claims of the ’485 and ’674 patents under the doctrine of equivalents. The court will enter an order in accordance with this Opinion.

ORDER GRANTING SONY’S MOTION FOR JUDGMENT AS A MATTER OF LAW, FOR A NEW TRIAL, AND ON PROSECUTION HISTORY ESTOP-PEL

For the reasons set forth in the foregoing Opinion of this date, IT IS HEREBY ORDERED as follows:'

1. Sony’s motion for judgment as a matter of law is granted.

2. In the alternative, Sony’s motion for a new trial on the issue of infringement of the ’674 and ’485 patents is granted.

3. Sony’s motion on prosecution history estoppel is granted.

4. Within five (5) days from the date of this order, the parties shall contact the court to schedule a telephone conference to discuss further proceedings in this matter. The court intends to issue an order certifying this case for immediate appeal pursuant to Fed. R.Civ.P. 54(b).

2 . At trial, Sony showed use of trapezoidal electrodes and angled doping to prevent the barrier regions from spreading under the gates.

3 . Loral attempts to argue that Dr. Erb derived the process and structure described in his article from Dr. Kim’s presentation in San Diego. To the contrary, Dr. Kim also cites Dr. Erb's article (which includes Dr. Su as a co-author) as a reference in one of his own later publications. Thus, Dr. Kim apparently credited Dr. Erb with independent work. [SIX 595]. More important, as noted hereafter, Dr. Su’s notebook shows an earlier conception date than Amelio’s or Kim’s. Consequently, no reasonable jury could conclude Dr. Erb derived the process and device described in the December article from Dr. Kim.

4 . As noted, any suggestion that the structure of a CCD does not reveal the process by which that CCD was made conflicts with Loral's arguments that Sony learned the '674 process, at least in part, by cutting open or reverse engineering a Fairchild CCD. Moreover, the depiction of structure in Dr. Su's notebook clearly shows alignment of the barrier regions with the first gate electrodes just as the diagrams in the '674 patent depict alignment with the first insulation layer over the first gate electrodes. Although witnesses agreed a structure could be made different ways, that observation does not refute a diagram’s ability to disclose the process for making the disclosed structure. That is, a diagram may. convey to one of ordinary skill in the art many processes that could be used to make that device. The' skilled artisan may not be able to pinpoint the precise- method used to make that device. Nevertheless, the diagram would still disclose all of the methods understood by the person of skill in the art.

5 . Fairchild abandoned its apparatus application after the rejection based on Erb. [STX 502]. This apparent acquiescence in the examiner’s rejection supports this court’s conclusion that the Erb reference is prior art. If Fairchild had questioned the prior art status of the Erb reference during prosecution, a time much closer to the actual inventive events, it could have “sworn behind” the Erb article under PTO Rule 131, 37 C.F.R. 1.131. Instead, at that time when evidence was more likely obtainable, Fairchild abandoned its apparatus claims based on a rejection in light of Erb.

By the way, the .PTO processed these divisional applications in two separate art units. Art Unit 325 handled the process claims which resulted in the ’674 patent. Art Unit 254 rejected the apparatus claims.

6 . The record suggests that both divisions should have been aware of Sony’s infringement. For instance, in August 1984 Dr. Wen, Joseph Roth-stein, General Manager of Fairchild’s CCD Imaging Division, and Irving Hirschberg, Director of Technology Development of Fairchild Weston, visited Sony. These Loral executives solicited Sony to manufacture CCD area imagers for Fair-child. [STX 2156-2]. During these discussions in 1984, Fairchild and Sony exchanged considerable information about their technology. [STX 2156-4-6; 2156-13-16],

Loral should have also been aware of the structure of Sony's CCDs from technical publications. Hiroyuki Matsumoto and his colleagues at Sony published two papers in 1983 describing Sony’s interline transfer CCD’s. [STX 2148, 2149]. Harvey Balopole testified that Fairchild’s California group follow the literature very carefully. [Balopole Dep. 1/29/93 at 555], Dr. Dyck also testified that he was quite knowledgeable about CCDs and followed technical publications. [Dyck Dep. 12/3/92 at 507], Furthermore, in 1985, Fairchild specifically requested a translation of one of Mr. Matsumoto’s papers. [STX 2157-29], Also found in Fairchild's files was an English translation of Mr. Matsumoto’s first paper. Although this translation was admittedly more coarsely translated than the later article it still stated in reference to the Sony CCD:

A 2-Iayer polysilicon overlap gate electrode structure is used for vertical and horizontal readout CCD registers. Charge transfer is the 2-phase driving n-channel flush type in which substrate concentration varies with ion implantation.

The translation bears a note indicating that it had been in the possession of at least "D. Wen" and "Rudy [Dyck].” [STX 2019]. There were also handwritten notes indicating that Fairchild recognized as referring to the “Sony 16” chip.

7 . The parties did not contest laches with respect to the ’485 patent because this court had already determined on summary judgment that Loral’s failure to require marking precludes pre-suit damages. Loral, 906 F.Supp. at 818.