OPINION
RADER, Judge.Plaintiff, Deuterium Corporation, seeks compensation from the United States for patent infringement. Plaintiff’s patent, United States Patent No. 4,123,506 (the ’506 patent), covers a process for removing hydrogen sulfide from geothermal steam. Plaintiff contends the United States used the process in a heat exchanger at The Geysers geothermal power plant in California.
After oral argument on February 20, 1990, this court grants defendant’s cross-motion for summary judgment.
FACTS
On March 9, 1989, this court issued an opinion setting forth the facts of this case and defining claim 57 of the ’506 patent. Deuterium Cory. v. United States, 16 CI.Ct. 454 (1989) (Deuterium I). This court incorporates the facts set forth in that opinion into this opinion as well. See id. at 455-58. In connection with the pending motions, this court finds additional facts.
In 1975, EIC Laboratories, Inc. (EIC) entered a contract with the predecessor to the Department of Energy (DOE) to test and evaluate a system for removing hydrogen sulfide from geothermal steam. At that juncture, EIC entered a separate contract with Pacific Gas and Electric Company (PG & E) to conduct field tests at The Geysers. Before this field testing occurred, DOE’s contract with EIC ended.
On June 14, 1978, about six months after the end of the EIC contract, DOE entered a contract with PG & E to jointly fund a test of EIC’s steam cleansing process. This contract obligated PG & E to design, operate, test, decommission, and evaluate a facility that removes hydrogen sulfide from 100,000 pounds of steam per hour. PG & E entered a separate contract with EIC in August 1978 to accomplish this objective.
EIC conducted its tests in 1979. EIC planned a four-month test as the centerpiece of the demonstration. During this period, EIC planned to tap the main steam line, treat 100,000 pounds of steam per hour, and return the cleansed steam to the main line upstream of PG & E’s turbines. In fact, this test ended after 120 hours when a pipe elbow ruptured on November 7, 1979.
The heat exchanger HX103 was a part of EIC’s pilot plant at The Geysers. During its tests, EIC modified HX103, originally designed as a cooling component, to preheat the reactant used to clean steam.1 At all times, HX103 was an internal component of EIC’s facility.
The '506 patent contemplated preheating of the reactant. The ’506 patent passed contaminated geothermal steam through a liquid reactant upstream of the turbines. By natural reaction, the reactant purged hydrogen sulfide from the raw steam. Step (a) notes that the reactant must be capable of reacting “at the temperature of ... [the] flow of geothermal steam____” To reach the high temperatures of steam from the wells, the process foresaw the need for heating of the reactant. Preheating the reactant prevented dissipation of the steam’s energy during removal of hydrogen sulfide. The clean steam retained its energy for turning the turbines and generating electricity, or, in the words of *626the patent, “for utilization.”2 Thus, preheating the reactant in HX103 played an important role in the '506 patent’s cleaning process.
Plaintiff contends that, for most of HX103’s operation, clean steam, and only clean steam, entered HX103.3 This clean steam heated the reactant. During the heating, steam condensed in HX103. The liquid condensate discharged from HX103 onto the ground within the EIC facility. Moreover, residual noncondensable gases discharged from HX103 into the air.
At oral argument on February 20, 1990, defendant’s counsel showed that the record contained no evidence that discharges from the HX103 were substantially free of hydrogen sulfide. Plaintiff did not show where the record contained such evidence. During the same argument, defendant’s counsel argued that even discharges during the periods when clean steam operated HX103 were not substantially free of hydrogen sulfide. Defendant’s counsel contended that discharges during these periods may have carried residues of hydrogen sulfide left in HX103 during operation with raw steam. The liquid discharge from HX103 combined with other contaminated fluids prior to discharge from EIC’s facility.4
During the 120-hour test, the PG & E turbines used about 1,000,000 pounds of steam per hour. The EIC facility cleansed about 100,000 pounds of steam per hour. The maximum flow rate of steam through HX103 was 200 pounds per hour.
In the pending motions, plaintiff contends that the doctrine of estoppel bars defendant from denying use of the ’506 patent in the heat exchanger. Plaintiff asserts that defendant in fact operated the heat exchanger with cleansed steam in violation of the '506 patent. Plaintiff also argues that defendant may not escape liability by invoking the experimental use exceptions.
Defendant responds that it has never admitted that the heat exchanger use infringed the ’506 patent. Rather defendant *627cross-moves for summary judgment maintaining that HX103 did not use the patented process. Moreover, defendant alternatively argues that operation of the heat exchanger fits within the experimental use exception to patent liability.
These contentions frame the three issues for court resolution:
(1) Did defendant admit that the heat exchanger infringed the ’506 patent?
(2) Did HX103 in fact use the patented process?
(3) Did the EIC demonstration facility as a whole fall within the experimental use exception to patent liability?
DISCUSSION
Summary Judgment Standards
In its earlier opinion, this court set forth the applicable standards for resolving summary judgment motions. Deuterium I, 16 Cl.Ct. at 458-59. This opinion follows those standards.
Estoppel
Plaintiff urges this court to estop defendant from denying use of the HX103 and use of the ’506 patent in the heat exchanger. According to plaintiff, “the heat exchanger use issue was already argued by the parties and defendant did not deny such use but rested its case on other matters upon which the Court has ruled against defendant____” Plaintiff’s Brief, filed June 27, 1989 (Pl.Br.), at 8.5
During the long and tortured history of this case, see Deuterium 1,16 Cl.Ct. at 456, both parties have made many different arguments. In 1985, the court and the parties attempted to reduce the allegations in the case to a few manageable issues. As part of this focussing process, defendant elected to argue first that release of EIC necessarily released defendant from liability for use of HX103. Defendant did not abandon other arguments against liability, but relied first upon the implications of EIC’s release:
Defendant believes that there are other reasons why the accused internal use of treated steam [in HX103] did not infringe the claims of the ’506 patent. However, these reasons are not discussed herein but defendant reserves the right to later raise any other non-infringement defense it may have.
Defendant’s Brief, filed Sept. 6, 1985 (Def. Br.), at 21. Defendant reiterated this position on other occasions. For instance, in this judge’s first status conference on this case, defendant’s counsel stated:
The Court: Would you stipulate ... with regard to the heat exchanger, that all the steps of the process were fulfilled?
[Defendant’s Counsel]: No. No, Your Honor. They were not. I know that much, they were not.
Transcript of Proceedings, No. 425-82C, Nov. 9, 1988, at 36-37. Defendant’s counsel further clarified that any admission about use of the patented process in HX103 applied only “for the purpose of this motion.” Id.
In sum, defendant argued first that release of EIC also released DOE from any liability for use of the patented process in the heat exchanger. On the basis of the facts before the court in the motions resolved on March 9,1989, this court rejected *628that argument.6 The estoppel doctrine, however, does not bar defendant from now asserting that the heat exchanger use did not infringe the patent. Defendant made no admissions precluding its opportunity to deny infringement. Upon review of the record, however, this court determines that defendant consistently has denied use of the ’506 patent in the heat exchanger.
Infringement Analysis
This court set forth the standards for infringement in Deuterium I, 16 Ct.Cl. at 459-61. Those standards apply to this motion as well. In sum, the inquiry for patent infringement is a two-step analysis. The first step, claim interpretation, is a question of law. At this stage, the court defines the bounds of the claim. The second step, determination of ultimate infringement, is a question of fact. At this later stage, the court determines whether the accused process is the same as the claim.
This court already has interpreted claim 57. In particular, this court presented a detailed legal reading of steps (d) and (f) of claim 57. Id. at 465-68. These same two steps remain at issue in this motion. Plaintiff contends that the heat exchanger used both steps. Defendant charges that the heat exchanger used neither step.
Step (d) of claim 57 requires delivery of cleansed steam “at superatmospheric pressure to said system for said utilization.” The term “said utilization” means use of the delivered steam as an energy source to generate electricity. See supra note 2. In connection with these motions, this court must determine whether the heat exchanger is a “system” that used cleansed steam as an energy source.
In Deuterium I, this court interpreted the term “said system.” This court determined that “said system” meant power stations, like The Geysers, which use geothermal steam to produce electricity. Relying upon claim language and the specification, this court concluded:
[E]lement (d) requires that steam treated according to the preceding steps (a) through (c) be delivered to the system for generation of electricity.
Deuterium I, 16 Cl.Ct. at 463. Moreover, the prosecution history revealed that the applicant narrowed his claim from a generic process for cleansing gases to a specific process for cleansing geothermal steam prior to its use as an energy source. Id. at 465-66. In light of this history, this court stated:
[T]he inventor indicated that Power Plant No. 7 (or similar geothermal power plants) — complete with main line pipes, turbines, heating units, and condensers— would fit within the definition of utilization system in claim 57.
The term [“said system”] as used throughout claim 57 covers the entire system to utilize steam at super-atmospheric pressure for the generation of electricity.
Id. at 464, 467.
Plaintiff moved this court under RUSCC 59 to reconsider the meaning of “said system.”7 This court undertook that recon*629sideration. Relying on the language of the preamble to claim 57 and other portions of the record, this court stated:
Thus, the cleansing process differs from the system which utilizes steam and produces discharges.
This court cannot properly construe the language of step (d) to mean “said system” is a steam cleansing process or an emissions control process.
Deuterium Corp. v. United States, No. 425-82C, Memorandum Opinion filed July 14, 1989, at 3. These statements summarize this court’s interpretation of step (d).
With this claim interpretation already done, this court must now determine whether the heat exchanger actually used the ’506 patent. Based on this court’s interpretation of step (d), HX103 did not infringe the '506 patent. The heat exchanger was part of the cleansing system, not part of the utilization system.
The primary teaching of the '506 patent was cleansing of geothermal steam upstream of the turbines. Deuterium I, 16 Cl.Ct. at 461. By removing impurities upstream of the turbines, the patent both protected metallic machinery against corrosion from sulfuric acid and protected the environment against pollution. Id. To ramain economically feasible, however, the cleansing process could not rob the steam of its power generation potential.8 Thus, to achieve its cleansing objectives while preserving the steam's power, step (a) of claim 57 states that the reaction must be capable of occurring at the high temperatures of raw steam. The cleansing reaction had to occur at high temperatures to prevent dissipation of the steam’s energy.
In sum then, the '506 patent presupposed preheating of the reactant. To reach the extreme temperatures of raw steam from the wells, the reactant had to undergo preheating. Thus, HX103 was an important part of the '506 patent’s cleansing process.
As part of cleansing process, HX103 cannot also be part of a utilization system for the generation of energy. First, HX103 is not a power plant that generates electricity. The ’506 patent requires delivery of cleansed steam to “said system.” By use of the term “system,” the ’506 patent suggests a group of components interacting to perform a single function.9 The Geysers power plant — with its turbines, power lines, steam wells, steam conveyance pipes, and more — fits this description well. The single device that preheats chemical reactant is far less susceptible to description as a “system.”10
*630Plaintiff nevertheless suggests that HX103’s use of clean steam for heat satisfies the ’506 patent’s terms.11 To the contrary, HX103 used clean steam to prepare reactant to clean more steam. HX103 used clean steam for cleansing steam, not for producing energy. HX103 did not use the clean steam as a source of energy. Thus, the heat exchanger did not fit the meaning of “said system” in steps (d) and (f) of claim 57.
The heat exchanger also does not satisfy the terms “said utilization” as required by step (d). HX103 was part of the ’506 patent. It performed an essential part of the cleansing process. HX103 cannot be both a system which uses the product of the patented process and a part of the process which produces the product. HX103 heats reactant to clean steam, which, in turn, returns to HX103 to heat more reactant to clean more steam. This endless cleaning process does not ever provide for “utilization” of the clean steam. HX103 never uses clean steam as an energy source or otherwise, except to clean more steam. Until delivery of clean steam for use as an energy source, step (d) is absent.
The language of the ’506 patent distinguishes the utilization system from the cleansing process. The preamble of claim 57 states:
57. A process controlling emissions of hydrogen sulfide from a system which utilizes a flow of geothermal steam at superatmospheric pressure and produces discharges of liquid water and of noncondensable gases____
The ’506 patent, Col. 18, lines 24-27. The cleansing process differs from the system that utilizes steam and produces discharges. The heat exchanger — an essential part of the process — is not an independent utilization system within the meaning of the patent. Therefore, the heat exchanger did not use step (d) of claim 57.
Furthermore, step (f) is absent from the operation of HX103. Step (f) covers dis*631charges from “said system.” HX103 does not satisfy the term “said system.” HX103 is part of the cleansing process, not a separate system for utilization of cleansed steam. In the absence of a “system,” none of the discharges from HX103 qualify to satisfy step (f).
Moreover, as defendant’s counsel pointed out during the February 20,1990 oral argument, plaintiff showed no evidence within the record that the discharges from HX103 were substantially free of hydrogen sulfide. Plaintiff urged this court to assume that clean steam would only produce clean discharges. Defendant raised, however, the possibility that discharges during periods HX103 operated with clean steam could carry residues from the periods HX103 operated with raw steam. Plaintiff produced no evidence to show that the discharges were in fact substantially free of hydrogen sulfide. Thus, plaintiff did not show that HX103 used step (f).
Because each step of a claim is “material and essential,” Pennwalt Corp. v. Du-rand-Wayland, Inc., 833 F.2d 931, 935 (Fed.Cir.1987), the absence of a single claimed element in the accused process precludes a finding of infringement. Teledyne McCormick Selph v. United States, 214 Ct.Cl. 672, 558 F.2d 1000 (1977). In the case at bar, steps (d) and (f) of claim 57 are absent from the accused HX103 process. Therefore, the heat exchanger did not infringe the ’506 patent.
Experimental Use
The basic law of patents establishes that unauthorized use of a patented product or process constitutes infringement. Aro Mfg. Co. v. Convertible Top Replacement Co., 377 U.S. 476, 84 S.Ct. 1526, 12 L.Ed.2d 457 (1964). However, the Court of Claims has acknowledged that an experimental use does not infringe. Chesterfield v. United States, 141 Ct.Cl. 838, 846, 159 F.Supp. 371 (1958); see also, Finney v. United States, 202 Ct.Cl. 1130 (1973) (Order).
Two subsequent Court of Claims opinions limited the application of the experimental use exception.12 The first case, Douglas v. United States, 181 USPQ 170 (Ct.Cl. Trial Division 1974), affd, 206 Ct.Cl. 96, 510 F.2d 364 (1975), described the experimental use doctrine as “an expression of the maxim de minimis non curat lex.” Id. at 177. This court questions whether any infringing use can be de minimis. Damages for an extremely small infringing use may be de minimis, but infringement is not a question of degree. Indeed, the Court of Claims trial judge proceeded to apply a purpose test for application of the doctrine:
At no time were the accused devices used for amusement, to satisfy idle curiosity, or for philosophical inquiry; to the con*trary, each use was in keeping with the legitimate business of the using agency and served a valuable governmental and public purpose.
Douglas, 181 USPQ at 177.13
Similarly, in Pitcairn v. United States, 212 Ct.Cl. 168, 547 F.2d 1106 (1976), cert. denied, 434 U.S. 1051, 98 S.Ct. 903, 54 L.Ed.2d 804 (1978), the Court of Claims held that seven models of helicopters infringed 59 patents. Defendant asked the court to exclude from compensation those aircraft used for tests, demonstrations, and experiments. The court refused this request and adopted the purpose test found earlier in Douglas:
Use for such purposes is use by or for the Government and is compensable. Obviously every new helicopter must he tested for lifting ability, for the effect of vibration on installed equipment, flight speed and range, engine efficiency, and *632numerous other factors. Tests, demonstrations, and experiments of such nature are intended uses of the infringing aircraft manufactured for the defendant and are in keeping with the legitimate business of the using agency. Experimental use is not a defense in the present litigation.
Pitcairn, 547 F.2d at 1125-26.
The United States Court of Appeals for the Federal Circuit followed Pitcairn in holding that the experimental use doctrine did not cover “limited use of a patented drug for testing and investigation strictly related to FDA drug approval requirements during the last 6 months of the term of the patent____” Roche Prods, v. Bolar Pharmaceutical Co., 733 F.2d 858, 861, cert. denied, 469 U.S. 856,105 S.Ct. 183, 83 L.Ed.2d 117 (1984).14 The Federal Circuit applied the purpose test and determined that Bolar’s use was “solely for business reasons and not for amusement, to satisfy idle curiosity, or for strictly philosophical inquiry.” Roche, 733 F.2d at 863. The Federal Circuit declined to apply the experimental exception to unauthorized uses with “definite, cognizable, and not insubstantial commercial purposes.” Id.
Thus this court must determine whether DOE’s alleged use of the ’506 patent was experimental or “in keeping with the legitimate business of the using agency.” Pitcairn, 547 F.2d at 1125-26. The task of weighing this standard is complex because a part of the “legitimate business” of DOE is to conduct energy experiments and demonstration projects.
In oral argument on February 20, 1990, defendant urged this court to adopt the objective standards for determining an experimental use which the Federal Circuit employs in connection with a different experimental use exception. In a related branch of patent law, the Federal Circuit has provided several objective factors for ascertaining experimental uses. Section 102(b) of Title 35 bars applicants from receiving a patent if their invention was “in public use or on sale in this country, more than one year prior to” the application. 35 U.S.C. § 102(b) (1982). A public use or sale for experimental rather than commercial purposes, however, is an exception to the bar. In re Brigance, 792 F.2d 1103 (Fed.Cir.1986). In articulating the experimental use exception to the public use or sale bar, the Federal Circuit has enunciated several objective factors showing experimental use. Brigance, 792 F.2d at 1108; Hycor Corp. v. Schlueter Co., 740 F.2d 1529,1535 (Fed.Cir.1984); TP Laboratories v. Professional Positioners, Inc., 724 F.2d 965, 971-72 (Fed.Cir.), cert. denied, 469 U.S. 826, 105 S.Ct. 108, 83 L.Ed.2d 51 (1984).
This court, however, declines to apply these factors because they evolved for a very different purpose. These factors provide broad protection for an inventor who has yet to secure a patent. The experimental use exception to a public use or sale protects an inventor’s intellectual property for the period of development and testing prior to patent application. The experimental use exception to infringing uses, on the other hand, narrows the scope of intellectual property protection. This separate experimental use exception protects an individual making unauthorized use of a patent (a potential infringer) during tests seeking advancement or commercialization of the patented teaching. Thus, the broader objective standards developed to protect an inventor during experimentation prior to patent application do not apply to experiments by a potential infringer.
*633The standard for the experimental use exception to infringing uses grows from the judicial origins of the exception. The Federal Circuit quoted extensively from Professor W. Robinson’s famous treatise to show the origins of this doctrine:
The interest of the patentee is represented by the emoluments which he does or might receive from the practice of the invention by himself or others.... Hence acts of infringement must attack the right of the patentee to these emoluments____ An unauthorized sale of the invention is always such an act. But the manufacture or the use of the invention may be intended only for other purposes, and produce no pecuniary result. Thus where it is made or used as an experiment, whether for the gratification of scientific tastes, or for curiosity, or for amusement, the interests of the patentee are not antagonized, the sole effect being of an intellectual character in the promotion of the employer’s knowledge or the relaxation afforded to his mind. But if the products of the experiment are sold, or used for the convenience of the experimenter, or if the experiments are conducted with a view to the adaptation of the invention to the experimenter’s business, the acts of making or of use are violations of the rights of the inventor and infringements of his patent.
Roche, 733 F.2d at 862 (quoting W. Robinson, The Law of Patents for Useful Inventions § 898 (1890)) (emphasis in original). This explanation distinguishes experimente of a purely intellectual character from experiments designed to adapt the invention to pecuniary and business uses. Moreover, the Federal Circuit advises “narrow” application of this experimental use exception. Roche, 733 F.2d at 863.
The weight of the objective evidence in the case at bar shows that DOE’s participation was not strictly intellectual experimentation, but development of technology and processes for commercial applications. At the EIC pilot plant, DOE engaged in a demonstration project, not mere experimentation. Moreover the objective of the demonstration project was to develop an economically feasible commercial application of the ’506 patent.
On June 14, 1978, PG & E and DOE entered a cooperative agreement to develop a demonstration project for the steam cleaning process:
WHEREAS, PG and E ... has submitted an unsolicited proposal to DOE requesting assistance in demonstrating, on a scale plant basis, the specific technology involved in the EIC Corporation’s copper sulfate process for upstream removal of H2S and other trace contaminants ...
NOW, THEREFORE, the parties hereby agree to cooperate in realizing the development and construction of the aforementioned demonstration project according to the terms and conditions which follow:
ARTICLE II
GENERAL SCOPE OF EFFORT AND PLACE OF PERFORMANCE
PG & E shall ... cause to be designed, fabricated, made operational, tested, decommissioned and technically evaluated, a 100,000 lb./hr. scale demonstration facility which utilizes EIC Corporation’s copper sulfate upstream hydrogen sulfide removal process____
See Plaintiff’s Brief, No. 425-82C, filed Oct. 7, 1985, Appendix, at 16, 18-19 (Cooperative Agreement, No. ET-78-A-03-1744) (emphasis added). Thus, the contract between PG & E and DOE acknowledged that the pilot plant would be a demonstration facility. Indeed, the facility operated for at least 120 hours during which The Geysers power plant generated 55 megawatts of electric power for commercial consumption.
On June 15, 1977, PG & E submitted a proposal to DOE which later became part of the cooperative agreement between the two. In this document, PG & E described the purpose of the demonstration facility:
Because of the inherent advantages of an upstream system, and the disadvan*634tages of the downstream processes that have been developed to date, PG & E and Union believe it is in the best interest of ERDA [Energy Research & Development Administration] and in the national interest to complete the development and commercialization of EIC’s upstream H2S, removal system.
Plaintiff’s Brief, No. 425-82C, filed June 27, 1989, Exhibit A (Pl.Ex. A), at 4 (Proposal to ERDA). Thus, the demonstration facility anticipated development and commercialization of the steam cleaning process.
The EIC pilot plant has none of the hallmarks of an experiment conducted for curiosity, amusement, or intellectual stimulation. Rather, on the continuum from onetime laboratory tests to lengthy laboratory experimentation, to small field tests, to scale demonstration projects, to large-scale pilot projects, to commercial exploitation, the EIC scale plant is just short of profitable exploitation. Indeed, for a short period, steam processed at the EIC facility helped generate commercial electricity.
The record shows that EIC had tested its steam cleaning process extensively in laboratories before proposing a pilot project. PG & E’s proposal to ERDA stated:
All steps of this process were initially tested in the laboratory by EIC under contract with ERDA (No. E(ll-l)-2730). The results of these very successful tests are summarized in report No. C002730-2 included herein____
PLEx. A, at 8. The EIC process had undergone extensive testing. Any experimentation motivated by curiosity, amusement, or general intellectual inquiry took place long before creation of the detailed and expensive proposal for a pilot project. EIC’s pilot demonstration facility was far down the path toward commercial exploitation.
DOE responds that its involvement was purely experimental. To support this proposition, defendant produces the affidavits of Gerald Katz and S. Garratt Sharp.15 Both of these knowledgeable authorities characterize the EIC pilot plant as experimental. Although using the term “experimental,” both describe the EIC project in terms leaving little doubt about the extent and purpose of the demonstration plant.
Mr. Katz, DOE’s Project Manager in 1978, described the purpose of the project as “to explore the technical and economic feasibility of controlling and eliminating the hydrogen sulfide emissions____” Def. Br., Appendix, Exhibit 2, at A-35 (Katz Affidavit). Mr. Katz further explained that the EIC plant intended to accomplish “further development work.” Id. at A-36. Although, according to Mr. Katz, “[t]here was never any intent to use the experimental apparatus for any commercial purpose,” id. at A-37, the application of the EIC apparatus is not in question. Rather, Mr. Katz’s affidavit suggests that DOE and EIC conducted this development work to determine economic feasibility. In sum, the project anticipated and advanced toward potential commercial applications. Mr. Katz does not speak in terms of pure scientific and intellectual inquiry.
Similarly, Mr. Sharp, a PG & E chemist, describes the demonstration plant in terms of exploring ways to “reduc[e] atmospheric pollution.” Id., Ex. 3, at A-48. Moreover, he addressed the benefits of extended turbine life and decreased turbine maintenance. Again, this explanation suggests potential commercial benefits and applications. Mr. Sharp does not speak in terms of pure scientific and intellectual inquiry.
The record does not show that DOE engaged in the type of experimentation cognizable as an exception to infringement doctrines. Rather the EIC pilot plant was a demonstration facility. This demonstration facility examined the economic feasibility and benefits of the EIC steam cleansing process. If fully successful, the pilot plant could have paved the way for broader commercial applications. The experimental use exception does not apply to DOE’s cooperative activities with PG & E and EIC.
CONCLUSION
Nothing in the history of this litigation precludes or estops DOE from contending *635that EIC’s HX103 did not use the ’506 patent. Moreover, HX103 did not use steps (d) and (f) of claim 57. Therefore, it did not infringe the ’506 patent. Finally, the experimental use exception does not apply to the EIC pilot program. Because HX103 did not infringe the ’506 patent, this court grants defendant’s motion for partial summary judgment.
No costs.
. EIC’s pilot plant, as initially designed, had three heat exchangers. The first heat exchanger, HX101, functioned to preheat the copper sulfate reactant for the cleansing process. Shortly after EIC began to test its steam cleansing process, however, HX101 clogged. High temperatures caused rapid formation of deposits on HXlOl’s mechanisms. A tenacious scale of copper salts fouled its gaskets and plates. Due to the problem, the entire EIC pilot plant failed. Plaintiffs Statement of Genuine Issues and Plaintiffs Version of Defendant’s Proposed Facts, No. 425-82C, filed Oct. 10, 1989, at ¶ 4(b). Therefore, EIC’s program manager, Dr. Francis C. Brown, modified HX103 to take over the preheating function. Plaintiff suggests that Dr. Brown acted under supervision from PG & E and DOE. As modified, the HX103 allowed EIC to regulate temperatures in the heat exchanger to avoid deposits.
. As granted on October 31, 1978, step (d) of claim 57 read:
(d) delivering said recovered flow of steam at superatmospheric pressure to said system for said utilization as an energy source____
United States Patent No. 4,123,506, Col. 18, lines 53-55 (the '506 patent). On January 13, 1979, the United States Patent and Trademark Office issued a Certificate of Correction. The Certificate deleted the last four words of step (d). Thus, step (d) reads:
(d) delivering said recovered flow of steam at superatmospheric pressure to said system for said utilization....
Defendant’s Motion for Partial Summary Judgment, filed Sept. 6, 1985, Appendix Vol. I, at A-16. The specification, however, clarifies repeatedly that the patent contemplates use of the clean steam "as an energy source.” The '506 patent, Col. 2, lines 16, 27, 30; Col. 3, lines 5, 10-11, 25, 39, 43-44, 48. Moreover, in discussing ways to increase the "efficiency of the steam as a medium for transfer of its energy," id., at Col. 2, lines 39-40, the specification refers directly to "geothermal steam such as is found in wells in the California area known as The Geysers.’ ” Id., at Col. 2, lines 43-44. The specification also claims that the patent reduces "unnecessary expenditures for power and equipment to accommodate and eject the same.” Id., at Col. 3, lines 1-2. The specification emphasizes the patent’s benefits for systems using steam to generate power. Thus, the term "for utilization” in step (d) refers to using clean steam as an energy source to generate electricity. As discussed in this court’s earlier opinions, the utilizing systems of step (d) are power plants like those at The Geysers that use steam to generate electricity-
. Defendant explained:
Both raw (untreated) process steam and treated (scrubbed) steam were available for use in HX103 and both types of process steam were in fact used at various times in HX103.
Defendant’s Brief, No. 425-82C, filed Aug. 7, 1989, (Def.Br.) at 8. At oral argument on February 20, 1990, defendant’s counsel further stated, and plaintiffs counsel did not contest, that clean steam ran through HX103 most of the time it was in operation. On some occasions, however, raw steam entered the HX103.
. Mr. Jerome S. Spevack, the patentee, stated in a 1984 affidavit:
The condensate and residual noncondensable gases resulting [from HX103] ... which were substantially free of hydrogen sulfide, became combined with and were discharged from the system with other fluid discharges.
Def.Br., Appendix, Affidavit of Jerome S. Spevack, Sept. 24, 1984, at A-82. This statement also tends to show that the patentee did not regard HX103 as a separate utilizing system for the '506 patent. Rather, the patentee regarded the HX103 as part of the larger steam cleaning system.
. Plaintiff relies heavily on a statement in defendant’s 1988 proposed findings. In that 1988 submission, defendant stated:
37. Claim 57 of the '506 patent is directed to a process and it does not claim an apparatus. A process claim is infringed only when the process is used and not when an apparatus is constructed. Since it was EIC that operated the pilot plant, it follows that it was EIC that used the accused process.
Defendant’s Proposed Findings, filed June 21, 1988, ¶ 37. Plaintiff apparently contends that the last sentence admits defendant’s use of the EIC process. To the contrary, this sentence states only that EIC used "the accused process.” Moreover this sentence says nothing about defendant’s use of the patented process. Defendant might have included within this sentence an explicit denial of its use of the accused process or the patented process. This additional clarity, however, was not necessary to reserve defendant’s opportunity to deny liability for any possible infringement by the HX103. Moreover, defendant explicitly reserved its arguments on other occasions.
. In reaching this conclusion, this court “resolved all factual questions in favor of the plaintiff____" Deuterium Corp. v. United States, 16 Cl.Ct. 454, 471 (1989) (Deuterium I). In construing the facts in favor of plaintiff, this court assumed that EIC "only performed part of the accused process____” Id. at 472. This court accepted allegations that EIC "performed only certain tasks for PG & E, which, when combined with the rest of PG & E’s activities, amounted to use of the '506 patent.” Id. With respect to HX103, however, Dr. Francis C. Brown — EIC’s project manager — states by affidavit that EIC built, installed, modified, and operated the entire heat exchanger operation. Def.Br., Appendix, Brown Declaration, Aug. 3, 1989, at A-64-65. According to Dr. Brown, EIC alone controlled all aspects of HX103’s operation. Id. at A-62-63. Dr. Brown states that EIC, not PG & E, modified HX103. To the extent that Dr. Brown correctly presents these facts, the release of EIC also released DOE from liability for any alleged use of the '506 patent by HX103. Vintén Ltd. v. United States, 213 CtCl. 759, 566 F.2d 1188 (1977) (Order). Plaintiff contests Dr. Brown’s declaration. In this opinion, this court, on other grounds, disposes of DOE’s liability for HX103. Therefore, this court need not resolve at this time any disputes about Dr. Brown’s Declaration.
. Although moving for reconsideration, plaintiff had earlier agreed with this court’s interpreta*629tion of the terms "said system for utilization.” In arguing for dismissal of a declaratory judgment action brought in a district court, plaintiff stated:
[I]nfringement can — and will — occur only when that [’506] process is used, by a power plant, actually to cleanse its emissions of hydrogen sulfide.
EIC Laboratories, Inc. v. Deuterium Corp., No. 81 Civ. 4232 (WK), slip op. at 3 (SDNY Feb. 11, 1982). Thus, in an earlier proceeding before another court, plaintiff agreed that the ’506 patent contemplated use of the treated steam to generate power. In the litigation before the Claims Court, plaintiff also stated that its patent requires ”use[] as a power plant with a condenser at the end does use it.” Transcript of Proceedings, No. 425-82C, filed Mar. 18, 1985, at 43.
. The '506 patent’s specification clarifies this purpose:
With these objects in mind the invention aims to [extract contaminants] ... with only negligible reduction of the heat content of the steam and with increase in the efficiency of the steam as a medium for transfer of its energy.
The '506 patent, Col. 2, lines 31-41. To preserve the heat content of the steam, the reaction had to occur at high temperatures. Thus, this objective required preheating the reactant.
. According to Webster’s New Collegiate Dictionary (1975), "system” means, among other things, "1: a regularly interacting or interdependent group of items forming a unified whole....” Id. at 1184.
. This court’s reading of the term "system” is further buttressed by claim 57’s inclusion of step (e). Step (e) requires condensation of the used steam. The cooling tower of The Geysers power plant affirmatively implements step (e). HX103, however, has no specific mechanism or means for condensing, but merely acknowledges that condensation occurs as a passive result. Although a process patent need not supply a device to implement each step of the claim, The Geysers power plant seems again to satisfy well claim 57’s definition of "system," while HX103 struggles and fails to satisfy the claim.
. Plaintiff relies on an ambiguous statement in the specification discussing uses of geothermal steam:
Steam, including geothermal steam, has been utilized to provide heat energy and/or mechanical energy by way of heat exchangers and prime movers, e.g., steam engines and turbines, in systems usually constituting or employing means for condensing the steam from gaseous to liquid phase.
The '506 patent, Col. 1, lines 17-22. This language, which refers to both heat exchangers and turbines as uses of steam, appears at the outset of a very general description of the "State of the Art" of generic steam cleaning. This general description of the uses of all steams, "including geothermal steam,” does not refer to uses of geothermal steam cleaned by the '506 patent. As the specification continues, however, it focuses on the use of geothermal steam cleaned by the '506 process as an energy source to generate electricity:
In practice, geothermal steam driven turbines and related equipment for producing power have been constructed with special materials intended to resist the corrosive nature of hydrogen sulfide ... Such hydrogen sulfide pollution abatement practices have been found to be expensive, deal only with environmental aspects of the power plant effluents, create catalyst recovery and waste sulfur materials collection and disposal problems, and fail to take advantage of the ammonia contained in geothermal steam.
Id. at Col. 1, lines 45-48, 65-68; Col. 2, lines 1-3 (emphasis added). Thus, as the specification moves from the general to the more specific, it clarifies the utilization systems under the '506 patent. As mentioned earlier, the specification then moves from discussing turbines and power generation equipment to an even more specific discussion of the contaminants in geothermal steam from wells at The Geysers. The '506 patent at Col. 2, lines 41-66.
In context, plaintiff s citation to the specification’s general reference to "heat exchangers” also fails for another reason. The reference appears immediately under the heading "State of the Art." Moreover, the sentence in question focuses retrospectively on past uses of steam ("Steam, including geothermal steam, has been utilized ...”) (emphasis added), rather than on the future uses of steam cleansed by the '506 process. This context clarifies that the reference is an allusion to the use of heat exchangers as described in the prior geothermal art of the Pauli, et al. patent. The heat exchanger concepts of the Pauli, et al. patent have little or no relation to the HX103. See, United States Patent No. 3,817,038, Col. 2, line 34; Deuterium I, 16 Cl.Ct. at 470.
Thus, the overall picture provided by the claim language, the specification, the prosecution history, and other aids to interpreting the '506 patent narrows the uses for steam cleansed according to the protected process. When step (d) of claim 57 discusses "said system for utilization,” it has the meaning set forth by this court in Deuterium I.
. The United States Court of Appeals for the Federal Circuit refers to the experimental use doctrine as both an “exception" and a "defense” to the basic patent doctrine that unauthorized use constitutes infringement. Roche Prods, v. Bolar Pharmaceutical Co., 733 F.2d 858, 862, cert. denied, 469 U.S. 856, 105 S.Ct. 183, 83 L.Ed.2d 117 (1984).
. The Federal Circuit stated that Douglas lacks precedential value because the Court of Claims did not affirm the part of the trial judge’s opinion dealing with experimental use. Roche, 733 F.2d at 863.
. In Title II of the Drug Price Competition and Patent Term Restoration Act of 1984, Congress changed the law enunciated in the Roche case. 35 U.S.C. § 271(e) (Supp. II 1984). Section 271(e)(1) now permits individuals to "make, use, or sell, a patented invention (other than a new animal drug or veterinary biological product ...) solely for uses reasonably related to the development and submission of information under a Federal law which regulates the manufacture, use, or sale of drugs.” Id. Congress excluded new animal drugs and veterinary products because the rest of the complex patent term restoration law excluded these drugs. Although it changed that narrow application of the doctrine affecting reporting requirements for federal drug laws, Congress did not disturb the Federal Circuit’s enunciation of the parameters of the experimental use exception.
. In 1983, plaintiff moved to strike these two affidavits. Defendant opposed. In an Order, filed Sept. 5, 1989, this court denied plaintiff’s motions.