holding that good faith must be taken into account in determining whether nondisclosure constitutes conduct culpable enough to require a finding of intent to deceiveSummary of this case from ATT CORP. v. MICROSOFT CORPORATION
February 13, 1991. Rehearing Denied March 15, 1991.
Donald R. Dunner, Finnegan, Henderson, Farabow, Garrett Dunner, Washington, D.C., argued for plaintiff-appellant. With him on the brief were Allen M. Sokal and Richard L. Stroup.
Ronald L. Palmer, Baker Botts, Dallas, Tex., argued for defendant-appellee. With him on the brief were Mitchell D. Lukin, Sylvia M. Egner, Baker Botts, Houston, Tex. and Dana M. Raymond and Richard G. Berkley, Brumbaugh, Graves, Donohue Raymond, New York City, of counsel.
Appeal from the United States District Court for the Southern District of Texas.
Before NEWMAN, and RADER, Circuit Judges, and SKELTON, Senior Circuit Judge.
Halliburton Company (Halliburton) sued Schlumberger Technology Corporation (Schlumberger) for infringement of U.S. Patent Nos. 4,388,529 ('529 patent), 4,409,481 ('481 patent), and 4,424,444 ('444 patent). These patents embrace tools and techniques for neutron logging of oil wells. Specifically, the patents disclose use of high energy neutrons to measure the properties of underground geological formations.
The United States District Court for the Southern District of Texas determined that Halliburton did not disclose important prior art to the Patent and Trademark Office (PTO) during the application process. Hence, the district court found Halliburton had engaged in inequitable conduct. The district court refused to enforce Halliburton's patents and awarded Schlumberger $3,375,000.00 in attorney fees and expenses. See Halliburton Co. v. Schlumberger Technology Corp., 722 F. Supp. 324, 12 USPQ2d 1765 (S.D.Tex. 1989). Halliburton appeals from this judgment and order. This court reverses.
Oil producers use well logging to detect oil in geological formations. This process involves lowering a complex instrument into a well borehole. As the logging tool moves down the borehole, it measures, or logs, the properties of the surrounding geological formation.
In neutron well logging, the instrument emits high energy neutrons (14 million electron volts) as it descends into the borehole. The neutrons collide with the nuclei of atoms in the borehole and the surrounding earth. These collisions generate high-energy inelastic gamma rays. These collisions also quickly drain energy from the neutrons. As the neutrons slow down, or moderate, they become thermal neutrons (0.025 electron volts). The nuclei of surrounding atoms capture these thermal neutrons. When this capture or neutron decay occurs, the capturing atoms emit capture gamma rays.
The logging tool detects these capture gamma rays. A count of capture gamma rays reveals the neutron population and the rate of neutron decay. Different materials capture neutrons at different rates. Thus, measuring the decay of thermal neutrons tells much about the composition of formations around the borehole. Oil, for instance, captures thermal neutrons at a different rate than saltwater. Figure 1 of the '481 patent depicts the logging tool:
During decay, the thermal neutron population changes in proportion to its size. Therefore, exponential equations express these neutron decay rates. These equations express scientific principles governing exponential growth and decay. The equations describe the results of any process in which a population — in this case thermal neutrons — changes at a rate proportional to its size. After measurement of the capture gamma rays by the logging instrument, these equations express the neutron decay rate.
These equations represent the derivative of the population size (the rate at which it changes) with some multiple, k, of the population size itself. If the population is decreasing, as in the case of neutron decay, the following equation results:
P = — k(dP/dt)
In this formula, P is the population size. dP/dt is the rate of population change. k is a proportionality constant. The solution to the differential equation is:
P(t) = Ae(-kt)
Decay of radioactive material is a well-known example of exponential decay.
Early in its use, however, neutron decay logging presented a major problem. The well casing itself, made up of metal and cement, absorbs neutrons and emits gamma rays. An accurate logging technique, therefore, must distinguish between gamma ray feedback from the well casing and feedback from the surrounding geological formations. In addition, water, oil, gas or air in the borehole may capture thermal neutrons and further distort the logging measurements. Early thermal neutron decay tools could not account for decay occurring within the borehole itself. To compensate for this limitation, these early tools assumed that decay within the borehole occurred faster than in the formation.
Under this assumption, the early logging methods delayed gamma ray measurement for an instant after a high-energy burst to allow borehole nuclei to capture thermal neutrons. These early methods arbitrarily attributed all measurements after the delay to the formation. This technique became known as "timing out the bore-hole." Because borehole conditions vary, this technique was fraught with uncertainty.
The claims at issue cover techniques and tools to measure capture gamma rays. Capture gamma rays represent the thermal neutron population after the inelastic gamma rays associated with fast neutrons have all dissipated. The patents describe pulsemoderation neutron systems. These systems generate short bursts or pulses of fast neutrons about once per millisecond (a thousand times per second). During and after each short burst of neutrons, the fast neutrons rapidly moderate into thermal neutrons. The thermal neutron population gradually declines as the surrounding atoms capture neutrons and emit capture gamma rays.
Claim 1 of the '444 patent is representative of the technology involved in each of the Halliburton patents:
A method for simultaneously measuring the thermal neutron decay time of materials in and about a well borehole, comprising the steps of:
generating, in a well borehole, a relatively short duration discrete burst of fast neutrons which are rapidly moderated by interaction with nuclei of materials in the borehole and surrounding earth formations and slowed down to thermal energy, creating a thermal neutron population in the borehole and surrounding earth formations;
detecting, in the borehole, radiations representative of the thermal neutron population in the borehole and surrounding earth formations, in at least four time intervals subsequent to said burst of fast neutrons and generating at least four count signals representative of said thermal neutron population during said at least four time intervals; and
combining said at least four count signals according to a predetermined relationship to simultaneously separate the borehole and formation decay components and to derive at least two measurement signals representative of the thermal neutron decay time of the borehole medium and the earth formation medium in the vicinity of the borehole.
The Halliburton inventions eliminate the imprecise "timing out" technique. Instead, the inventions measure capture gamma rays at four distinct time intervals. Halliburton's method then uses the two-exponential equation to separate the borehole readings from the readings in the surrounding formation.
In essence, the Halliburton inventions assume the two-exponential equation represents mathematically the total neutron population at any particular time. This equation has four unknowns. After making four gamma ray measurements at different times, the Halliburton method solves the equations for the four unknowns. Two of these values provide the neutron decay feedback for the geological formations around the borehole. The other two values provide the neutron decay information for the borehole itself. The calculated value for each of the unknowns ensures that the resulting two-exponential equation matches the neutron populations at the four different time gates.
The equation in footnote 1, supra, expresses radioactive decay for a single irradiated material. In neutron logging, the borehole and formation are irradiated, thus requiring two terms in the equation:
P(t)=(Ae (-kt)) formation + (Be (-lt)) borehole
A and B are the thermal neutron population in the formation and borehole at an initial time t. k and l are the lifetime of the thermal neutrons in the formation and borehole. A, B, k and l are the four unknowns.
By taking four measurements to obtain four data points and then using a computer program to derive the unknowns from the data points, the patented inventions omit the "timing out" technique of the prior art. The inventions thus facilitate a more reliable measurement of thermal neutron decay in the geological formation. They also reduce the demand for continuous neutron bursts and take capture gamma radiation readings sooner. These measurements therefore encompass practically the entire thermal neutron decay curve. Finally, the patents teach mathematical processing of those measurements to determine the thermal neutron characteristics of the borehole and the formation.
During the patent application process, appellant did not disclose to the PTO any prior art reference. However, Examiner Willis cited six patents directed to pulsemoderation systems. At trial, the district court determined that appellant should have disclosed seven other patents of which it was aware. The district court determined that this failure to disclose constituted inequitable conduct which made the patents unenforceable.
The doctrine of inequitable conduct requires a trial court to undertake a two-step analysis. The trial court must discern whether the withheld references satisfy a threshold level of materiality. The court must also determine whether the applicant's conduct satisfies a threshold showing of intent to mislead. See J.P. Stevens Co. v. Lex Tex, Ltd., 747 F.2d 1553, 1559-60, 223 USPQ 1089, 1092 (Fed. Cir. 1984), cert. denied, 474 U.S. 822, 106 S.Ct. 73, 88 L.Ed.2d 60 (1985). These initial determinations are questions of fact, which this court reviews under a clearly erroneous standard. Next, assuming satisfaction of the thresholds, the trial court must balance materiality and intent. Under Sea Indus., Inc. v. Dacor Corp., 833 F.2d 1551, 1559, 4 USPQ2d 1772, 1777 (Fed. Cir. 1987). The more material the omission, the less culpable the intent required, and vice versa. N.V. Akzo v. E.I. Dupont De Nemours, 810 F.2d 1148, 1153, 1 USPQ2d 1704, 1708 (Fed. Cir. 1987). The trial court has discretion to determine inequitable conduct. This court reviews that discretionary determination under an abuse of discretion standard. Kingsdown Medical Consultants, Ltd. v. Hollister Inc., 863 F.2d 867, 876, 9 USPQ2d 1384, 1392 (Fed. Cir. 1988), cert. denied, 490 U.S. 1067, 109 S.Ct. 2068, 104 L.Ed.2d 633 (1989).
Kingsdown Medical Consultants, Ltd. v. Hollister Inc., 863 F.2d 867, 9 USPQ2d 1384 (Fed. Cir. 1988), did not remove from the patent law the requirement to show threshold levels of materiality and intent prior to engaging in the balancing test enunciated by J.P. Stevens Co. v. Lex Tex, Ltd., 747 F.2d 1553, 223 USPQ 1089 (Fed. Cir. 1984), cert. denied, 474 U.S. 822, 106 S.Ct. 73, 88 L.Ed.2d 60 (1985). Kingsdown did discard the idea that a finding of gross negligence itself satisfies the intent threshold.
PTO Rule 56 is the appropriate starting point in determining the threshold level of materiality. This broad standard most closely delineates how an applicant ought to conduct business with the PTO. American Hoist Derrick Co. v. Sowa Sons, Inc., 725 F.2d 1350, 1363, 220 USPQ 763, 773 (Fed. Cir. 1984). Information is material if there is "substantial likelihood that a reasonable examiner would consider it important in deciding whether to allow the application to issue as a patent." 37 C.F.R. § 1.56 (1989).
However, a patentee has no obligation to disclose an otherwise material reference if the reference is cumulative or less material than those already before the examiner. See Specialty Composites v. Cabot Corp., 845 F.2d 981, 992, 6 USPQ2d 1601, 1609 (Fed. Cir. 1988). Therefore, if the references found by Examiner Willis more closely related to the Halliburton applications than the uncited art, Halliburton did not engage in inequitable conduct. With these standards in mind, this court reviews the district court's materiality findings. The non-disclosed prior art was U.S. Patent No. 4,326,129 to Neufeld, four Texaco patents (U.S. Patent Nos. 3,930,152, 3,940,611, 3,940,613, and 3,979,300) and two so-called skip-a-beat patents (U.S. Patent Nos. 3,733,486 and 4,002,904) also to Texaco. Examiner Willis cited six references. The most pertinent of these were U.S. Patent Nos. 4,122,338 and 4,152,590, both to Smith.
The Cited References
Halliburton's method of measuring directly the neutron decay for the borehole and formation is a direct extension of the methods in the cited references. Both of the Smith patents assume the existence of the mathematical relationship expressed by the two-exponential equation. The Smith patents, however, measure the neutron population at only two time gates. Thus, the Smith methods can calculate only two of the four unknowns in the two-exponential equation. They must rely on other information to estimate the remaining two unknowns.
The Smith '590 patent relies on the timing-out technique. The '590 patent thus assumes that the value of the exponential component due to borehole decay will eventually reach zero. This assumption reduces the two-exponential equation to a single exponential equation with only two unknowns. Thus, by measuring at two separate time gates, the Smith patent obtains two data points from which it computes the two unknowns due to the formation.
The Smith '338 patent uses the salinity of the borehole water to estimate the two unknowns attributable to the borehole. The '338 patent bases its estimation from prior knowledge of how salt water responds to neutron irradiation. Having estimated the two unknowns, the two-exponential equation becomes a single exponential equation with two unknowns. Measurement at two time gates permits calculation of the unknowns based on two data points.
The district court erroneously dismissed the cited references as less material because they relied on the timing-out technique. The Smith patents, however, like the Halliburton patents, assume the two-exponential equation and describe a method to calculate the unknowns related to the formation component of the decay. In the art of neutron well logging, the Smith patents were highly material to Halliburton's applications. The most pertinent prior art in the field was thus before the examiner. Nonetheless, Schlumberger asserts that Halliburton remained obliged to disclose the uncited references to the examiner.
The Neufeld Patent
The district court emphasized appellant's failure to disclose the Neufeld patent. The Neufeld patent recognizes that the exponential equation derived from both the borehole and the formation (the two-exponential equation) expresses total thermal neutron decay. Neufeld, however, reaches these results in a very different way than appellant's claimed inventions and the cited references.
Neufeld derives its neutron decay results mathematically. In practice, the Neufeld device discharges random bursts of neutrons into the borehole. A detector measures the entire gamma ray feedback. By comparing the detector output with the neutron burst, Neufeld then computes a formation and borehole decay rate from the continuous feedback.
The patent claims at issue, on the other hand, measure directly the neutron decay results. The inventions emit pulses at regular intervals. Between each pulse, the claimed inventions measure the decay rate directly.
The district court found that both appellant's patent claims and the Neufeld patent recognize exponential relationships in deriving the decay rate. Based on this similarity, the district court concluded that Neufeld was highly material to examination of appellant's patents. The district court, however, did not appreciate the significance of the differences between Neufeld and the claims at issue which make Neufeld less material than the cited references.
When weighing whether uncited prior art is more material than that before the examiner, a trial court considers similarities and differences between prior art and the claims of the patent. In making this determination, the trial court must consider portions of prior art references which teach away from the claimed invention. See Bausch Lomb, Inc. v. Barnes-Hind/Hydrocurve, Inc., 796 F.2d 443, 448-49, 230 USPQ 416, 420 (Fed. Cir. 1986), cert. denied, 484 U.S. 823, 108 S.Ct. 85, 98 L.Ed.2d 47 (1987).
Both Neufeld and appellant's claimed inventions recognize the two-exponential equation. This equation simply expresses mathematically an axiom of exponential decay. Thus, the two-exponential equation is not a critical component of either Neufeld or appellant's claimed inventions, but rather a scientific truism acknowledged by both. A reasonable patent examiner would not consider Neufeld important merely because both Neufeld and appellant's patents acknowledge this scientific equation.
The district court nonetheless found that the "principal invention of the patents-in-suit is their method for measuring and simultaneously decomposing the entire neutron decay curve . . . by a two-exponential relationship." Halliburton, 722 F. Supp. at 333, 12 USPQ2d at 1772. This finding is clearly erroneous. Appellant claims a method of measuring directly the formation and borehole components of the decay curve between regular neutron pulses in four time gates. By misconstruing the crux of appellant's claimed inventions, the district court also missed the significance of the differences from the Neufeld reference.
The district court itself noted these substantial differences between Neufeld's and appellant's subject matter. For instance, Neufeld depends on continuous random energy pulses and achieves a 50% radiation use rate. Appellant uses short discrete neutron bursts at regular intervals having a radiation use rate of 1.5%. Neufeld continuously detects gamma rays between random pulses. Appellant detects gamma rays in four distinct time gates after each pulse. Neufeld performs a correlation calculation to derive a decay curve from both inelastic and capture gamma ray data. Appellant directly measures separate borehole and formation components of the decay curve using only capture gamma rays. Halliburton, 722 F. Supp. at 334, 12 USPQ2d at 1772-73.
These significant differences distinguish the Neufeld disclosure and appellant's claimed inventions. Comparing these differences with the similarities leads to the conclusion that Neufeld is less material than the cited references. In sum, while in the same field of endeavor, Neufeld was far less material to appellant's applications than the cited references. Neufeld disclosed a technique that emitted neutrons and measured capture gamma rays very differently from that claimed by appellant's patents. Therefore, no disclosure obligation attached to Neufeld.
The Texaco Patents
The district court also based its inequitable conduct finding on appellant's failure to disclose the four Texaco patents. The Texaco patents also recognize the two-exponential equation. They also disclose a way to get both the borehole and formation thermal neutron decay times. However, the Texaco patents, like Neufeld, use a method different from appellant's claimed inventions to get these results. Therefore, the Texaco patents are also less material than the cited references.
The Texaco disclosures show the continuous emission of fast neutrons. These continuous emissions vary harmonically as a function of time. Texaco also modulates these emissions at three different frequencies. Each frequency generates both inelastic and thermal gamma rays. These rays, in turn, generate a phase coherent thermal neutron population. A detector continuously measures the relative phase angle of these neutron populations. The tangents of these phase angles show the neutron decay time for both the borehole and formation.
Texaco's method is unlike that disclosed in appellant's patents or in the cited references. Appellant's claimed inventions measure different feedback data in a different way. The Texaco patents therefore are not more material to appellant's patents than the cited references.
In determining that the Texaco patents were material, the district court noted that appellant's patent attorney, Mr. Beard, used the '611 patent as a template in drafting the '444 application. However, because the Texaco patents are less material than the cited references, Mr. Beard's alleged use of the '611 patent as a template for the background portions of the '444 specification does not raise its level of materiality.
The Skip-a-Beat Patents
The skip-a-beat patents describe a method for eliminating inaccurate gamma radiation readings originating from the neutron pulse generator, foreign material, or the detector crystal. Several claims of appellant's patents include a step which uses this background correction method, along with many other limitations on which appellant based patentability. The inclusion in claims of a step that is known to the art does not require the applicant to disclose, under pain of inequitable conduct, references that describe each known step. The inventors did not represent that they invented this step, which is included in only some of the claims. See generally Stevenson v. International Trade Comm'n, 612 F.2d 546, 555, 67 CCPA 109, 204 USPQ 276, 284 (1979). In any event, because the skip-a-beat patents are less material than the cited references, no disclosure obligation attached.
Inequitable conduct requires an intent to act inequitably. FMC Corp. v. Manitowoc Co., 835 F.2d 1411, 1415, 5 USPQ2d 1112, 1115 (Fed. Cir. 1987). Materiality of an undisclosed reference does not presume an intent to deceive. Allen Organ Co. v. Kimball Int'l, Inc., 839 F.2d 1556, 1567, 5 USPQ2d 1769, 1778 (Fed. Cir. 1988). Further, a mere showing that references having some degree of materiality were not disclosed does not establish inequitable conduct. FMC Corp., 835 F.2d at 1411, 5 USPQ2d at 1115.
This record contains no direct evidence that appellant intended to mislead the PTO by withholding references. Such direct evidence of intent to mislead is often absent. Rohm Haas Co. v. Crystal Chem. Co., 722 F.2d 1556, 1571, 220 USPQ 289, 301 (Fed. Cir. 1983). The district court concluded, however, that appellant was grossly negligent in failing to disclose the withheld material references. From this finding of gross negligence, the district court inferred intent.
Gross negligence does not of itself justify an inference of intent to deceive. Kingsdown, 863 F.2d at 876, 9 USPQ2d at 1392. This court has clarified that negligent conduct can support an inference of intent only when, "viewed in light of all the evidence, including evidence indicative of good faith," the conduct is culpable enough "to require a finding of intent to deceive." Id.
Appellant admits awareness of the withheld references. Appellant's counsel, however, did not consider the references material. As discussed above, the withheld references obtain the neutron decay curve in an entirely different way than disclosed by Halliburton's patents. While all acknowledge the two-exponential equation, the equation is not part of the claims in any of the patents. In light of these facts, Mr. Beard's assertion that he did not intend to mislead is objectively reasonable. See Argus Chem. Corp. v. Fibre Glass-Evercoat Co., 759 F.2d 10, 14, 225 USPQ 1100, 1103 (Fed. Cir.), cert. denied, 474 U.S. 903, 106 S.Ct. 231, 88 L.Ed.2d 230 (1985).
The district court cited various actions by Halliburton or its representatives in support of its holding that Halliburton withheld the references with intent to deceive. With respect to Neufeld, the district court held that Beard's involvement with the prosecution of the patents-in-suit imposed an obligation to disclose Neufeld. The court held that Beard's failure to disclose Neufeld constituted gross negligence because "a reasonable person in Beard's position should have known of the materiality of Neufeld and disclosed it to the PTO." Halliburton, 722 F. Supp. at 336, 12 USPQ2d at 1774. The trial court further cited Halliburton's attempt, due to fear of infringement, to purchase Neufeld contemporaneous with the prosecution of the patents-in-suit. Id. at 336-37, 12 USPQ2d at 1774. The district court held that Halliburton's gross negligence justified finding an intent to deceive. Id. at 337, 12 USPQ2d at 1774.
As support for finding intent to deceive with respect to the Texaco patents, the district court cited Halliburton's belief that the Texaco patents were material and that its TMD tool infringed the Texaco patents. The district court further cited Beard's alleged use of the '611 patent as a template in drafting the '444 patent, and the examiner's failure to reference any prior art that disclosed the key features of Halliburton's inventions. Id. at 337-38, 12 USPQ2d at 1775. The district court relied on other similar factors with respect to the skip-a-beat patents to find intentional non-disclosure. Id. at 340, 12 USPQ2d at 1777.
This court need not decide whether the factors relied upon by the district court tend to support a finding of carelessness or gross negligence in prosecuting Halliburton's applications, or whether they simply show Halliburton's appreciation of the narrow technological basis of its claimed inventions. See In re Jerabek, 789 F.2d 886, 891, 229 USPQ 530, 533 (Fed. Cir. 1986). As Kingsdown made abundantly clear, gross negligence alone does not compel the inference of intent to deceive. Gross negligence cannot elevate itself by its figurative bootstraps to an intent to mislead based on the identical factors used to establish gross negligence in the first instance unless all the facts and circumstances indicate sufficient culpability. An applicant's conduct in its entirety must "manifest a sufficiently culpable state of mind to warrant a determination that it was inequitable." Consolidated Aluminum Corp. v. Foseco Int'l Ltd., 910 F.2d 804, 809, 15 USPQ2d 1481, 1484 (Fed. Cir. 1990).
Moreover, as stated above, the cited references were more pertinent to the Halliburton applications than were the withheld references. Halliburton had no obligation to disclose cumulative references. Specialty Composites, 845 F.2d at 992, 6 USPQ2d at 1609; J.P. Stevens, 747 F.2d at 1559-60, 223 USPQ at 1092. Halliburton did not act with culpable intent. Halliburton could not have acted with culpable intent with respect to cumulative references. The district court's finding to the contrary is clearly erroneous.
The record does not support a finding of intent to mislead. Nor does the record support a finding that the withheld references were more material than the cited references. The district court's materiality and intent findings are clearly erroneous. Therefore, no balance of materiality and intent is necessary. This court concludes that appellant did not engage in inequitable conduct.