This patent infringement suit involves two patents claiming chemical catalysts. We affirm the District Court's finding of noninfringement as to one patent, but reverse as to the other.

Catalysts are chemical substances that affect the rate or course of a given chemical reaction in some manner without becoming a significant part of the resulting product. In this case, both catalysts find wide application in the petrochemical field and are used to cause small molecules of gases or liquids to react together to form solid plastics and synthetic rubbers from which commercially useful articles and objects may be fabricated for use in everyday living.

The nature of these patents and the processes involved in their application pose especially difficult decisions for nonchemist judges. Although patent cases often present complex and technical questions, suits involving chemical patents present a complexity of a higher order. Unlike mechanical patents, the workings of which may at least be visualized and conceptualized by the lay mind, chemical patents take us into a realm of symbolic analysis. The reactions occurring in these processes cannot be appreciated by the independent senses: they cannot be seen, felt, or heard. Hence, chemists must describe them in terms of symbols, including letters and numbers.

Because the scientific underpinnings of chemical patents are so complex, trial courts must depend heavily upon expert witnesses for both explanation and evaluation of the patent disclosures and the accused infringing operations. Similarly, appellate courts must rely heavily upon the findings and credibility choices made by trial courts. Otherwise, we would be thrust into the position of reconsidering credibility and re-resolving conflicts in expert testimony, a position that mistakes this Court's function. American Seating Co. v. Southeastern Metals Co., 412 F.2d 756 (5th Cir. 1969); compare Rule 52(a), F.R.Civ.P. Nonetheless, our review of the District Court's findings is not limited to an examination of the expert testimony. Appellant raises a serious legal contention about the proper construction of a patent on a chemical catalyst, posing questions of law that go beyond the scope of expert testimony and this contention must be resolved before we can turn to the construction of the specific patents in suit.

Before submerging ourselves in an examination of the various contentions put forth by the parties, a brief review of the relevant principles of organic chemistry will be helpful. Both patents here in suit are directed to polymerization catalysts used in the chemical reaction processes that produce synthetic polymers.

A catalyst is defined as a substance which affects the rate or course of a given chemical reaction in some manner without becoming a significant part of the reaction product. Normally, catalysts are used in relatively small amounts as compared with the reactants.

Synthetic polymers are produced by causing hydrocarbon molecules, called monomers, which are generally in liquid or gaseous form, to link together in long chains. These long chains are called polymers. The linking of the monomers is termed a polymerization reaction, and the polymerization catalyst is what causes the monomers to link together to form polymers.

Ethylene, propylene, and butadiene, related gases generally obtained from the petroleum industry, are the monomers involved in this suit. These monomers have at least one double bond between two carbon atoms in their small chains, called an olefinic bond, and are known as olefins. Ethylene has two carbon atoms, propylene three, and butadiene four.1 When these monomers are polymerized, they yield synthetic polymers of primary concern in this suit, polyetheylene, polypropylene, and polybutadiene, the solid material of commercially useful plastic and synthetic rubbers. Prior to the invention of the catalysts covered by the patents in suit, the polymerization of these olefins was difficult. Harsh reaction conditions of extremely high pressure and elevated temperatures were required, and the olefins combined in disorderly arrays with many branched sidechains. "Unlike these inferior polymers, though, the polymers formed with Ziegler's catalysts have a higher density, greater rigidity and strength, and higher flow and melt temperatures."

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

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

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

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

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

The chemical catalyst patents in suit were developed by Karl Ziegler, the plaintiff, a citizen and resident of West Germany. In 1953, Ziegler and three coworkers at the Max-Planck-Institute in Mulheim, West Germany, discovered that several combinations of an organo aluminum compound and a compound of a metal of Groups IV-B, V-B, or VI-B of the Periodic System of Elements, produced polymerization catalysts that polymerized olefins much more effectively than was previously possible. The catalysts caused monomers to combine in a linear fashion, forming straight and not branched chains, without requiring the formation process to include high pressures or excessive temperatures. These various polymerization catalysts were recognized as a great scientific achievement and were named, and are universally known as, "Ziegler catalysts." For this discovery, Ziegler, and his co-workers were awarded the Nobel Prize. This suit asks us to define the breadth of two of the many United States patents protecting Ziegler's research.

Ziegler brought this patent infringement suit against Phillips Petroleum Company, charging infringement of United States Letters Patent Nos. 3,113,115 of December 3, 1963, and 3,257,332 of June 21, 1966. Phillips counterclaimed for a declaratory judgment of invalidity with respect to both patents. After a trial requiring twelve days of extensive oral testimony and 140 exhibits, the District Court denied both Ziegler's claim of infringement and Phillips' counterclaim by adjudging the patents valid but not infringed. Both parties appeal from the District Court judgment.

Patent No. 3,257,332 (the '332 patent), entitled "Polymerization of Ethylene," was issued June 21, 1966, on an application filed November 15, 1954. The fourteen claims in the '332 patent are set out in the margin.2

Although the complaint asserts nine of these claims against the alleged infringing process, it is only necessary to deal with claim Number 1 since it describes in general terms the compounds which are more specifically named in the other claims. An infringement of any of the claims asserted would necessarily infringe claim Number 1.

Claim Number 1 states that the '332 patent is directed toward a

"[p]olymerization catalyst, comprising the product formed by mixing an effective amount of an aluminum trialkyl with a compound of a metal selected from the group consisting of salts, freshly precipitated oxides and hydroxides of metals of Groups IV-B, V-B and VI-B of the Periodic System, including thorium and uranium."In this patent, the organo-aluminum compound part of the catalyst is an aluminum trialkyl.

The other patent, No. 3,113,115, entitled "Polymerization Catalyst," (the '115 patent), was issued December 3, 1963, on an application filed October 29, 1958. It is directed at catalysts in which the organoaluminum compound is dialkyl aluminum instead of trialkyl aluminum as in the '332 patent. Of the eighteen claims in the '115 patent, nine are in suit, and these claims are set out in the margin.3

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

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

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

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

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

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLEIII. The Alleged Infringing Phillips Processes

This case was commenced on September 6, 1967, when Ziegler filed suit on the '332 patent. Ziegler charged specifically that the catalyst employed in Phillips' polybutadiene polymerization operation at Borger, Texas, infringed the '332 patent.

On August 12, 1969, Ziegler amended his original complaint to include the '115 patent. Ziegler charged that the catalyst employed in Phillips' polypropylene operation at its Adams Terminal and Monument plants in Texas infringed the '115 patent.

In the polymerization of butadiene at the Borger plant, Phillips mixed and reacted three components, aluminum triethyl, titanium tetrachloride, and iodine, in the presence of butadiene to produce polybutadiene. The evidence showed that a stream of butadiene and toluene and a stream of aluminum triethyl and toluene are first combined and then introduced into a mix pot. Then separate streams of titanium tetrachloride diluted in toluene and free iodine diluted in toluene are introduced into the mix pot. These materials are then mixed together in the mix pot and the mixture is passed into a reactor where the butadiene polymerizes into high cis 1,4-polybutadiene rubber.

In the polymerization of propylene at the Adams Terminal and Monument plants, Phillips mixed and reacted three components, diethyl aluminum chloride, titanium trichloride, and aluminum trichloride, in the presence of propylene to produce polypropylene. The evidence showed that Phillips' operation involved mixing a stream of propylene with diethyl aluminum chloride in a solvent pentane and then introducing this mixture into a loop-type reactor. A complex of titanium trichloride and aluminum trichloride dispersed in pentane solvent is first mixed with propylene and then introduced into the reactor. A solid polymer, polypropylene, is recovered from the reactor.

Thus, this case involves two separate and distinct polymerization operations of Phillips', with the catalyst claims of the '332 patent being asserted against a polybutadiene process and the different catalyst claims of the '115 patent being asserted against a polypropylene process. Therefore, after setting out the law governing patent infringement and construction, we consider each patent and the corresponding claim of infringement separately.

At the outset, we must determine the proper scope of appellate review. The question of infringement of a patent is a question of fact. Coupe v. Royer, 155 U.S. 565, 15 S. Ct. 199, 39 L. Ed. 263 (1895); Winans v. Denmead, 56 U.S. (15 How.) 330, 14 L. Ed. 717 (1853); Marvin Glass & Associates v. Sears, Roebuck & Co., 448 F.2d 60 (5th Cir. 1971); U. S. Industries, Inc. v. Otis Engineering Corp., 254 F.2d 198 (5th Cir. 1958). Hence, the "clearly erroneous" doctrine of Rule 52(a), F.R. Civ.P., is clearly applicable to the District Court's findings on the infringement question. American Seating Co. v. Southeastern Metals Co., supra; Phillips Petroleum Co. v. Sid Richardson Carbon & Gas Co., 416 F.2d 10 (5th Cir. 1969). But infringement ceases to become a pure question of fact, and therefore the force of Rule 52(a) diminishes, when the question of infringement requires the District Court first to construe the patent. The construction of a patent is a matter of law, Coupe v. Royer, supra; Cold Metal Process Co. v. E. W. Bliss Co., 285 F.2d 231 (6th Cir. 1960), cert. denied, 366 U.S. 911, 181 S. Ct. 1085, 6 L. Ed. 2d 235 (1961); 4 Walker on Patents Sec. 230 (Deller's ed. 1964), and appellate courts are not bound by the strictures of Rule 52(a) when the District Court erred in construing the patent in question. Harrington Mfg. Co., Inc. v. White, 475 F.2d 788 (5th Cir. 1973).V. The Applicable Law of Infringement

Infringement exists only where the accused device and the teachings of the patent in suit are "substantially identical in structure, mode of operation, and results accomplished." American Seating Co. v. Southeastern Metals Co., supra; Stewart-Warner Corp. v. Lone Star Gas Co., 195 F.2d 645 (5th Cir. 1952); see Hobbs v. United States, Atomic Energy Commission, 451 F.2d 849 (5th Cir. 1971). In applying this general test of infringement, courts follow a two-step analysis. First, does the accused device, process, or composition of matter literally infringe the patent in question? If not, then second, does the "doctrine of equivalents" nevertheless lead to a finding of infringement?

When we seek to determine what the patented invention is, we must first turn to the claims of the patent because the scope of every patent is limited to the invention described in its claims, read in the light of the specifications. Motion Picture Patents Co. v. Universal Film Mfg. Co., 243 U.S. 502, 37 S. Ct. 416, 61 L. Ed. 871 (1971); Texsteam Corp. v. Blanchard, 352 F.2d 983 (5th Cir. 1965), cert. denied, 387 U.S. 936, 87 S. Ct. 2064, 18 L. Ed. 2d 1000 (1967). As the Supreme Court stated in Graver Tank & Mfg. Co., Inc. v. Linde Air Products, Inc., 339 U.S. 605, 607, 70 S. Ct. 854, 855, 94 L. Ed. 1097 (1950): "In determining whether an accused device or composition infringes a valid patent, resort must be had in the first instance to the words of the claim. If accused matter falls clearly within the claim, infringement is made out and that is the end of it." See Williams Bit & Tool Co. v. Christensen Diamond Products Co., 399 F.2d 628 (5th Cir. 1968).

If patents were interpreted only by the literal scope of their claims, however, minor deviations in the structure of almost any invention could be devised to elude the reach of the patent's protection. Thus, experience with patent cases demonstrates that seldom may the question of infringement be determined on the literal words of the claim. U. S. Industries, Inc. v. Otis Engineering Corp., 277 F.2d 282 (5th Cir. 1960). In recognition of the fact that a patent would be virtually worthless if it did not protect against devices which incorporate unimportant variations of the patented device, courts developed the doctrine of equivalents to protect the patentee from devices that differ merely in name, form, or shape from the patented invention, Machine Co. v. Murphy, 97 U.S. 120, 125, 24 L. Ed. 935 (1877), but perform substantially the same function in substantially the same way to obtain the same result. Graver, supra; Sanitary Refrigerator Co. v. Winters, 280 U.S. 30, 50 S. Ct. 9, 74 L. Ed. 147 (1929).

To establish equivalency for the purpose of showing infringement of a patent claim, the patentee has the burden of proving a real identity of means, operation, and result. Foster Cathead Co. v. Hasha, 382 F.2d 761 (5th Cir. 1967), cert. denied, 390 U.S. 906, 88 S. Ct. 819, 19 L. Ed. 2d 872 (1968); Texsteam Corp. v. Blanchard, supra; Stewart-Warner Corp. v. Lone Star Gas Co., supra. But while substantial identity of function, operation, and result are necessary, the doctrine

"does not require complete identity for every purpose and in every respect. In determining equivalents, things equal to the same thing may not be equal to each other and, by the same token, things for most purposes different may sometimes be equivalents."

Graver, supra, 339 U.S. at 609, 70 S. Ct. at 857. Hence, minor variations between the accused structure and the precise language of the claims in question do not prevent a holding of infringement. Graver, supra; Samuelson v. Bethlehem Steel Co., 323 F.2d 944 (5th Cir. 1963), cert. denied, 376 U.S. 938, 84 S. Ct. 793, 11 L. Ed. 2d 659 (1964); Up-Right, Inc. v. Safway Products, Inc., 315 F.2d 23 (5th Cir.), cert. denied sub nom. Beatty Safway Scaffold Co. v. Up-Right, Inc., 372 U.S. 934, 83 S. Ct. 881, 9 L. Ed. 2d 766 (1963).

In its early development, the doctrine of equivalents generally was applied in cases involving the equivalence of devices having mechanical components. Today, however, the same principles are applied to compositions of matter where there is equivalence between chemical ingredients. See Graver, supra; Chemical Cleaning, Inc. v. Dow Chemical Co., 379 F.2d 294 (5th Cir. 1967), cert. denied, 389 U.S. 1040, 88 S. Ct. 777, 19 L. Ed. 2d 829 (1968). See also 4 Walker on Patents, supra, Sec. 250. In Graver the Supreme Court spoke to the problem of determining what constitutes equivalency.

"Consideration must be given to the purpose for which an ingredient is used in a patent, the qualities it has when combined with the other ingredients, and the function which it is intended to perform."

Graver, supra, 339 U.S. at 609, 70 S. Ct. at 857.

A necessary corollary to the general doctrine of equivalents is that patent claims are construed in the light of the description and to cover the real invention found in the specification and drawings. Kinnear-Weed Corp. v. Humble Oil & Refining Co., 259 F.2d 398 (5th Cir. 1958), cert. denied, 361 U.S. 903, 80 S. Ct. 210, 4 L. Ed. 2d 158 (1959). Thus a claim may be given its true construction and meaning by reference to the accompanying specifications and drawings which, although they cannot enlarge it, may give the claim such limitation and definition as is necessary to make its abstract words descriptive of a specific invention. See Laitram Corp. v. Deepsouth Packing Co., 443 F.2d 928 (5th Cir. 1971). It is well settled, however, that a patent is not to be limited to the preferred embodiments shown in the specification. See Continental Paper Bag Co. v. Eastern Paper Bag Co., 210 U.S. 405, 28 S. Ct. 748, 52 L. Ed. 1122 (1908). And the general rule governing the relative weight to be accorded examples in construing a patent is set out in Sterner Lighting, Inc. v. Allied Electrical Supply, Inc., 431 F.2d 539 (5th Cir. 1970), cert. denied, 401 U.S. 909, 91 S. Ct. 869, 27 L. Ed. 2d 807 (1971): "Usually a patent will not be so strictly construed that it is held to the strict device described by the claims or shown through the drawing." 431 F.2d at 544. More specifically, "[t]he reference to a specific aspect of the invention does not limit the broader statement of the 'principal object' and the claims. . . . It is well settled that the claims delineate the scope of protection afforded by a patent, not the specific embodiments shown in patent drawings." Edward Valves, Inc. v. Cameron Iron Works, Inc., 286 F.2d 933, 942 (5th Cir.), cert. denied, 368 U.S. 833, 82 S. Ct. 55, 7 L. Ed. 2d 34 (1961).

The nature of the invention determines the breadth of its protection under the doctrine of equivalents. Because the doctrine rests on equitable considerations, the principle has developed that an inventor, with respect to his patent, is entitled to a range of equivalents commensurate with the scope of his invention. American Seating Co. v. Southeastern Metals Co., supra; Rothstein v. Atlanta Paper Co., 321 F.2d 90 (5th Cir. 1963); Up-Right, Inc. v. Safway Products, Inc., supra; Industrial Instrument Corp. v. Foxboro Co., 307 F.2d 783 (5th Cir. 1962); U. S. Industries, Inc. v. Otis Engineering Corp., supra; Cameron Iron Works, Inc. v. Stekoll, 242 F.2d 17 (5th Cir. 1957); Southern Saw Service, Inc. v. Pittsburgh-Erie Saw Corp., 239 F.2d 339 (5th Cir. 1956), cert. denied, 353 U.S. 964, 77 S. Ct. 1047, 1 L. Ed. 2d 914 (1957); Big "G" Distributing Co. v. Air Cleaner Service Co., 179 F.2d 122 (5th Cir. 1950); Hughes v. Magnolia Petroleum Co., 88 F.2d 817 (5th Cir. 1937). Thus, "[i]f the invention is broad and primary in its character, the range of equivalents will be correspondingly broad, under the liberal construction, which the courts give to such inventions." Continental Paper Bag Co. v. Eastern Paper Bag Co., supra, 210 U.S. at 414, 28 S. Ct. at 749, citing Miller v. Eagle Mfg. Co., 151 U.S. 186, 207, 14 S. Ct. 310, 38 L. Ed. 121 (1894). See also Hunt Tool Co. v. Lawrence, 242 F.2d 347 (5th Cir.), cert. denied, 354 U.S. 910, 77 S. Ct. 1296, 1 L. Ed. 2d 1428 (1957); Southern Saw Service, Inc. v. Pittsburgh-Erie Saw Corp., supra.

Under the doctrine of equivalents, the broadest protection is reserved for "pioneer" or "generic" patents. A pioneer patent is "a patent covering a function never before performed, a wholly novel device, or one of such novelty and importance as to mark a distinct step in the progess of the art, as distinguished from a mere improvement or perfection of what had gone before." Westinghouse v. Boyden Power Brake Co., 170 U.S. 537, 561-562, 18 S. Ct. 707, 718, 42 L. Ed. 1136 (1898). See Southern Saw Service, Inc. v. Pittsburgh-Erie Saw Corp., supra. Where the court is confronted with a pioneer patent, "liberality becomes the keynote of construction requiring the court to give the patentee a wide breadth of protection in construing the patent claims and specifications . . . ." Corning Glass Works v. Anchor Hocking Glass Corp., 374 F.2d 473, 476 (3d Cir.), cert. denied, 389 U.S. 826, 88 S. Ct. 65, 19 L. Ed. 2d 80 (1967). See also Hildreth v. Mastoras, 257 U.S. 27, 36, 42 S. Ct. 20, 24, 66 L. Ed. 112 (1921) ("As [plaintiff's] patent is a generic patent, the doctrine of broad equivalents properly applies here").

An invention is construed not only in the light of its claims but also with reference to its file wrapper or prosecution history in the Patent Office. Graham v. John Deere Co. of Kansas City, 383 U.S. 1, 86 S. Ct. 684, 15 L. Ed. 2d 545 (1966); Crawford v. Heysinger, 123 U.S. 589, 8 S. Ct. 399, 31 L. Ed. 269 (1887); Hogg v. Emerson, 52 U.S. (11 How.) 587, 13 L. Ed. 824 (1850). The file wrapper contains the complete Patent Office file of papers relating to the patent. Secret and confidential while the patent application is pending, it becomes public information once the patent is issued.

Under the file wrapper estoppel doctrine, if the file wrapper reveals that the patentee has surrendered claims, or has amended, narrowed, or otherwise limited his claims in response to the objections of the Patent Office, he may not later recapture through the doctrine of equivalents what he has given up. Graham v. John Deere Co. of Kansas City, supra; Dry Hand Mop Co. v. Squeez-Ezy Mop Co., 17 F.2d 465 (5th Cir. 1927). As this Court stated in Williams Bit & Tool Co. v. Christensen Diamond Products Co., supra, 399 F.2d at 633:

"It is settled that when the holder of a patent has voluntarily amended a claim by the inclusion of narrowing language in order to avoid prior art cited by the Patent Office and thus obtain the issuance of the patent, he is estopped from asserting that the amended claim covers what has been eliminated by the amending language."

A patentee's consent to the demands of the patent examiner, made to obtain his patent, operates as a "disclaimer," and he is thereafter bound by it. Schriber-Schroth Co. v. Cleveland Trust Co., 311 U.S. 211, 221-222, 61 S. Ct. 235, 85 L. Ed. 132 (1940); 4 Deller's Walker on Patents Sec. 234 (2d ed. 1965).

The mere surrender or amendment of a claim in the Patent Office does not, however, forever bar a patentee from asserting the doctrine of equivalents. He is estopped only from reclaiming what he surrendered. Borg-Warner Corp. v. Paragon Gear Works, 355 F.2d 400 (1st Cir. 1965). Hence it is necessary to determine what in fact the patentee gave up in order to receive its patent, see Edward Valves, Inc. v. Cameron Iron Works, Inc., supra, and what is the essence of the invention that remains. U. S. Peg-Wood, Shank and Leather Board Co. v. Sturtevant Co., 125 F. 382, 384 (1st Cir. 1903). Moreover, an applicant should not be presumed to have made a disclaimer broader than necessary to yield to the actual challenge to his claim. Hunt Tool Co. v. Lawrence, supra.

An improver cannot appropriate the basic patent of another, and an unlicensed improver is an infringer. Temco Electric Motor Co. v. Apco Mfg. Co., 275 U.S. 319, 48 S. Ct. 170, 82 L. Ed. 298 (1928); Cantrell v. Wallick, 117 U.S. 689, 6 S. Ct. 970, 29 L. Ed. 1017 (1886); Cochrane v. Deener, 94 U.S. 780, 24 L. Ed. 139 (1876); Williams Bit & Tool Co. v. Christensen Diamond Products Co., supra; Reed v. Hughes Tool Co., 261 F. 192 (5th Cir. 1919); Yancey v. Enright, 230 F. 641 (5th Cir. 1916); Bartlett v. Winton, 237 F. Supp. 631 (S.D.Fla. 1964); Continental Gin Co. v. Murray Co. of Texas, 171 F. Supp. 730 (N.D. Ala.), rev'd on other grounds sub nom. Murray Co. of Texas, Inc. v. Continental Gin Co., 264 F.2d 65 (5th Cir.), cert. denied, 360 U.S. 911, 79 S. Ct. 1296, 3 L. Ed. 2d 1261 (1959).

Correlatively, the grant of a patent on an improvement of a patented article does not excuse infringement of the dominant patent. Bryan v. Sid W. Richardson, 254 F.2d 191 (5th Cir. 1958); Bowen-Itco, Inc. v. Houston Engineers, Inc., 192 F. Supp. 223 (S.D. Tex.1961), rev'd on other grounds, 310 F.2d 522 (5th Cir. 1962), cert. denied, 372 U.S. 930, 83 S. Ct. 875, 9 L. Ed. 2d 734 (1963). Therefore, one who appropriates the substance of a patented invention without the consent of the patentee does not escape infringement by improving upon or subtracting from the invention so long as the essential elements are retained.

Infringement, the "Best Mode," and the Countervailing Policy

of Disclosure

Under 35 U.S.C.A. Sec. 271, a patentee's rights are infringed by anyone who "makes, uses or sells" the claimed invention. Moreover, the patentee need not spell out in the patent claims each of the many possible examples of the making and using of his invention. Section 112 of the Patent Code, 35 U.S. C.A. Sec. 112, merely requires the patentee "[to] set forth the best mode contemplated . . . of carrying out his invention." And normally the claims of a patent will not be restricted to such a "best mode." Smith v. Snow, 294 U.S. 1, 55 S. Ct. 279, 79 L. Ed. 721 (1935); Edward Valves, Inc. v. Cameron Iron Works, Inc., supra; Reiner v. I. Leon Co., 285 F.2d 501 (2d Cir. 1960); Bryan v. Sid W. Richardson, Inc., supra; B. B. Chemical Co. v. Ellis, 117 F.2d 829 (1st Cir. 1941). As Judge Learned Hand stated in B. G. Corp. v. Walter Kidde & Co., 79 F.2d 20, 22 (2d Cir. 1935):

"It is true that [the inventor of the spark plug] did not foresee the particular adaptability of his plug to the airplane . . . . Nevertheless, he did not shoot in the dark; he laid down with perfect certainty what he wished to accomplish and how . . . [H]e is not charged with a prophetic understanding of the entire field of its usefulness."

This rule of requiring the patentee to disclose only the "best mode" is nonetheless tempered by the countervailing policy of disclosure upon which the patent system is based. The root theory of the patent system is that of a trade: the inventor discloses his invention, in return for which the public grants him limited economic privileges.

"As a reward for inventions and to encourage their disclosure, the United States offers a seventeen-year monopoly to an inventor who refrains from keeping his invention a trade secret. But the quid pro quo is disclosure of a process or device in sufficient detail to enable one skilled in the art to practice the invention once the period of the monopoly has expired; and the same precision of disclosure is likewise essential to warn the industry concerned of the precise scope of the monopoly asserted."Universal Oil Products Co. v. Globe Oil & Refining Co., 322 U.S. 471, 484, 64 S. Ct. 1110, 1116, 88 L. Ed. 1399 (1944). Thus, a well-recognized exception to the general rule of Section 112 has developed. That is, where it appears that the "best mode" is in fact the basic teaching and actual invention or is the invention claimed, then the claims of that patent will be restricted to the "best mode." See Fritz W. Glitsch & Sons v. Wyatt Metal & Boiler Works, 224 F.2d 331 (5th Cir. 1955); Phillips Petroleum Co. v. Sid Richardson Carbon & Gas Co., 293 F. Supp. 555 (N.D.Tex.1968), aff'd, 416 F.2d 10 (5th Cir. 1969). Thus, a patentee will be restricted to the use or uses indicated in the patent if that application or use is indeed the basic teaching and actual invention.

We affirm the District Court's finding that Phillips' Borger polybutadiene operation did not infringe the '332 patent.

Phillips argues that it does not infringe the '332 patent because (1) the catalyst itself is different, (2) the starting material on which the catalyst operates is different, and (3) the end product is different.

1. Phillips asserts that the catalyst itself is different from that of the '332 patent because: it is prepared differently and with different components; it is mixed and reacted in the presence of the monomer, rather than out of the presence of the monomer on which it operates, as provided in the '332 patent; and the addition of iodine to the two key "Ziegler catalyst" components differentiates its catalyst from the '332 claim.

2. Phillips contends that the '332 patent is limited to the polymerization of ethylene, so the polymerization of butadiene falls beyond the scope of the '332 patent.

3. Phillips finally argues that the '332 patent is not infringed by the accused operation because a different end product is produced, not polyethylene but rather high cis polybutadiene, in its Borger operation.

As it construed the '332 patent, the District Court found no infringement because:

(a) Phillips does not polymerize ethylene, nor does it produce polyethylene, as disclosed by the '332 patent. Instead, Phillips polymerizes butadiene, a markedly different monomer, to produce a synthetic rubber substantially different from the plastic polyethylene.

(b) Phillips' accused catalyst is formed from three components added to butadiene. The '332 patent not only is limited to a two-component catalyst, but actually teaches away from the use of iodine as a catalyst component.

(c) Phillips' catalyst is not preformed, since all of the three catalyst components are mixed and reacted together in the presence of butadiene. Unlike the '332 catalyst, a reaction product is never formed by mixing two compounds prior to their addition to the polymerization system.

(d) Without the use of iodine or diiodobutene as a third catalytic component, Phillips' high cis 1,4- polybutadiene rubber cannot be satisfactorily obtained.

(e) The conduct of the '332 patentees confirms that the only invention disclosed, and intended to be disclosed, was a catalyst for the polymerization of ethylene consisting of the reaction product of only two components, separately prepared prior to any contact with the ethylene to be polymerized.

(f) The Phillips catalyst does not perform the same function in substantially the same way to achieve substantially the same result as the catalyst claimed in the '332 patent.

(g) The Phillips catalyst using molecular iodine or diiodobutene was not a known equivalent of the '332 patent at the time of its invention.

Ziegler contests this finding of noninfringement, arguing that the Court construed the '332 patent too narrowly on each point.

Ziegler's most important contention is that the District Court erred when it found that the '332 patent is limited to a catalyst for the polymerization of ethylene. Ziegler proposes an "all uses" theory for chemical catalysts. He contends that a patent on a chemical catalyst is a "composition of matter" patent and is infringed under 35 U.S.C.A. Sec. 271 whenever anyone, absent the patentee's permission, "makes, uses or sells" the claimed composition. This argument is too facile.

First, such an "all uses" theory presupposes that the '332 patent claims a composition of matter that is reproduced in all material respects in the accused Phillips operation. Yet there was no convincing proof that the chemical components disclosed in the '332 patent ever combined together in the Phillips process to produce that composition of matter. To the contrary, the expert testimony indicated that the presence of iodine necessitated a complicated series of reactions leading to the polymerization of butadiene, and we cannot conclude, on the record evidence, that the catalyst claimed in the '332 patent ever existed during the accused Phillips operation.

Second, we hesitate to apply so automatically an "all uses" theory to a chemical catalyst. Although the District Court here found that the '332 patent claimed a composition of matter, we are not convinced that such a limited view of a chemical catalyst is warranted. As Ziegler quite correctly points out, a chemical catalyst is a special kind of composition. In the chemical art, a catalyst is defined as a substance that affects the rate or course of a given chemical reaction or reactions without becoming a significant part of the reaction product. As Ziegler further points out a catalyst should be analogized to a "tool." Without it, nothing (or very little) happens when particular chemicals are reacted together. But different reactions require different catalysts. Hence, a catalyst must be defined in terms of the reactions that it catalyzes. In other words, the use of a composition of matter chemical catalyst is an integral part of the definition of that particular catalyst.

With this analysis, the limited utility of viewing a chemical catalyst as simply a composition of matter becomes apparent. Clearly, a catalyst system such as that described in the '332 patent has both composition of matter and process characteristics, and to pigeonhole the '332 patent's invention as one or the other, exclusively, would be needlessly formalistic and might produce a distorted view of the true patented invention.

Nevertheless, we are not holding as a matter of law that the patentee of a chemical catalyst is protected in his monopoly only so far as he specifically claimed the reactions in which the catalyst finds use. That question is not before us. Rather, our specific task is to construe the '332 patent in a way that gives effect to the real invention, and nothing more or less. We acknowledge that a patentee should not be charged with foreseeing all of its possible uses. See B. G. Corp. v. Walter Kidde Co., supra. But where the patentee has evidently limited his claims to a particular use or has implicitly not claimed a particular use, then a proper construction of the patent must take such a limitation or exclusion into consideration. In such a case, the "best mode" would be not merely an example, it would be the actual teaching of the patent. See, e. g., Fritz W. Glitsch & Sons v. Wyatt Metal & Boiler Works, supra.

The proper construction of a patent claiming a chemical catalyst system, therefore, looks not only at the components of the system but also at the reaction or reactions catalyzed and the reaction products. The basic question is whether the accused Phillips operation employs a catalyst system that is substantially equivalent to that disclosed in the '332 patent. Hence, we must consider, first, the components of each system, second, the particular chemical reactions catalyzed (that is, the particular monomer polymerized), and third, the reaction product (the particular polymer produced). But no one similarity or difference will be determinative.

At this point in our analysis, we feel constrained to point out that no magic formula can be discerned that will lead us inexorably to a finding of infringement or noninfringement. Many possible permutations of the above factors are possible. For example, one might employ the same components to polymerize a different monomer. Or one might employ slightly different components to polymerize the same monomer. Or one might employ slightly different components to polymerize a different monomer to produce a different polymer. Or one might employ slightly different components to polymerize the same monomer to produce a different polymer. In each case, the extent of identity or similarity between the catalyst system disclosed by the '332 patent and that of the accused catalyst system will guide our determination of infringement or noninfringement.

With this approach, we examine the District Court's conclusion that the '332 patent's catalyst system is not infringed by the accused Phillips operation. After a thorough reading of the briefs, the trial testimony, and the patent itself, we agree with the District Court. First, although the components of the '332 patent's catalyst system are employed in the accused operation, this similarity alone is not determinative of the question of infringement. Second, a proper construction of the '332 patent indicates that it claims only a catalyst system for the polymerization of ethylene. Third, the catalyst system claimed in the '332 patent was not intended by the patentees to polymerize butadiene into high cis 1,4-polybutadiene. Taking these three conclusions together, we hold that the basic teaching and actual invention of the '332 patent discloses exclusively a catalyst system for the polymerization of ethylene into polyethylene.

Ziegler argues, first, that the '332 patent specifically claims an application to polymerization reactions other than the polymerization of ethylene. Second, he argues that, since it is a "pioneer patent" and is therefore entitled to a very broad construction under the doctrine of equivalents, Westinghouse v. Boyden Power Brake Co., supra, the '332 patent should be construed to claim a catalyst for the polymerization of butadiene.

As to the literal infringement argument, the claims of the '332 patent nowhere mention the polymerization of butadiene. Ziegler fails completely to make out a claim of literal infringement.

As to his second contention, we conclude that, even given a broad construction, the '332 patent is limited to a catalyst for the polymerization of ethylene. Not only the patent itself, but the background of the patent, as exposed by record testimony and exhibits, makes it clear that Ziegler had no intention of claiming a catalyst for butadiene to produce the kind of polybutadiene made by Phillips in its accused Borger, Texas, operation.

Although the District Court made no explicit finding that the '332 patent was a "pioneer patent," it found that the claimed composition of matter catalyst was "new and useful" and it construed the patent broadly. We follow that approach, giving the '332 patent the broadest construction possible. In other words, we treat the '332 patent as a "pioneer" or "generic" patent. Nonetheless, we find untenably broad any construction of the patent that would encompass a catalyst for the polymerization of butadiene. No intent to claim a catalyst for the polymerization of monomers other than ethylene may be gleaned from the patent as a whole. The '332 patent is entitled "Polymerization of Ethylene;" its preamble states emphatically and succinctly that "[t]his invention relates to new and useful improvements in the polymerization of ethylene for the production of high molecular polyethylenes;" the preamble lists five "Objects of Invention," all five of which deal only with "ethylene" or "polyethylene;" the description preceding the claims focuses solely upon the polymerization of ethylene; and the Examples without exception are limited to the production of "high molecular polyethylenes" from ethylene.

In addition to the wording of the patent itself, the evidence presented at trial indicates that the '332 patent was not intended to claim a catalyst that polymerized butadiene into high cis butadiene.

First, the evidence indicated that the '332 catalyst performed poorly in polymerizing butadiene. For example, in 1954 one of Ziegler's co-patentees, Dr. Martin, attempted to use the '332 catalyst to polymerize butadiene. The experiment was unsuccessful, and Martin noted in his lab journal that "butadiene was poorly polymerized."

Second, it is apparent that butadiene reacts differently than ethylene. Phillips' expert, Dr. Reynolds, testified that, unlike ethylene, butadiene is a "conjugated diolefin" that is "very different" in its "chemical reactivity." Dr. Reynolds further testified that butadiene has "very special properties" and that the "conjugated diolefin concept" plays an important role in the polymerization of butadiene. Even Ziegler's expert, Dr. Mark, testified that the polmerization of butadiene is complicated and that, depending upon the catalyst and other conditions, butadiene can polymerize to either a resinous substance or a synthetic rubber. Moreover, Ziegler himself pointed out, in a pending patent application relating to the polymerization of propylene, that just because a catalyst polymerizes ethylene does not mean that it will necessarily operate on a higher monomer such as butadiene.

Third, Dr. Reynolds testified that polybutadiene comes in two basic forms, cis and trans. The chemical distinction between the two forms lies in the positioning of the hydrogen atoms along the chain of carbon atoms. More importantly, each form has very distinct physical characteristics. Trans polybutadiene is a hard, inelastic synthetic rubber, while cis polybutadiene is a soft, highly elastic synthetic rubber. Dr. Reynolds testified that polybutadiene containing less than eighty-five percent cis-type is essentially similar in physical properties to trans-type, but that, once the eighty-five percent level is achieved, the physical properties associated with cis-type polybutadiene increase dramatically. Thus, even if the '332 catalyst had been able to polymerize butadiene to some extent, there is no convincing evidence that it would have been able to produce a polybutadiene containing at least eighty-five percent cis-type. In fact, in a 1961 article for a technical journal, one of Ziegler's expert witnesses wrote that "[n]o really high . . . cis polybutadiene has as yet been made with the aid of Zieglertype catalysts."

Thus, on the basis of the patent itself, as construed in the light of its "pioneer" character, and on the basis of its background and development, we affirm the District Court's determination that the '332 patent is limited to the polymerization of ethylene and was not intended to encompass the formation of high cis polybutadiene. We affirm the finding of noninfringement as not "clearly erroneous" under Rule 52(a), F.R.Civ.P.

We reverse the District Court's finding that the '115 patent was not infringed by Phillips' Adams Terminal and Monument processes. Although Phillips' arguments are similar to those made against the claimed infringement of the '332 patent, we conclude that a correct construction of the patent and a proper application of the doctrine of equivalents requires a finding that the '115 patent is valid and infringed.

As with the '332 catalyst, Phillips argues that it does not infringe the '115 patent because (1) its catalyst is different, (2) the starting material on which its catalyst operates is different, and (3) its end product is different. Phillips contends that

1. the accused catalyst itself is different from that of the '115 patent because it is prepared differently and with different components: it is mixed and reacted in the presence of the monomer rather than out of the presence of the monomer on which it operates, and the addition of aluminum trichloride to the two key "Ziegler catalyst" components differentiates its catalyst from that of the '115 patent;

2. the '115 patent is limited to the polymerization of ethylene or ethylene-dominated monomers, so it does not extend to the polymerization of propylene by Phillips; and

3. the '115 patent is not infringed by the accused Phillips operations because they produce a different end product, not polyethylene or an ethylenedominated copolymer as claimed in the patent, but rather polypropylene.

Construing the '115 patent, the District Court found no infringement because:

(a) Phillips does not polymerize ethylene nor does it produce polyethylene or an ethylene-dominated copolymer. Instead, Phillips polymerizes propylene to obtain polypropylene.

(b) The accused catalyst does not "consist essentially" of the two components required by the '115 patent claims. Rather, the Phillips catalyst is formed from diethylaluminum chloride, and a complex of titanium chloride and aluminum trichloride.

(c) Phillips catalyst is not preformed, as required by the '115 patent. Instead, its three components are mixed and reacted together in the presence of propylene.

(d) Phillips' propylene polymerization catalyst does not perform the same function in substantially the same way to achieve the same result as the '115 catalyst. On the contrary, Phillips' three component catalyst achieves a result materially different from the result obtained with the '115 patent's catalyst.

(e) Phillips' titanium trichloride-aluminum trichloride complex is not the equivalent of titanium tetrachloride or even titanium trichloride alone.

(f) The use of a complex of titanium trichloride and aluminum trichloride was not a known equivalent of anything disclosed in the '115 patent at the time of its invention of the '115 patent.

We have already rejected Ziegler's "all uses" theory in our analysis of the '332 patent, and that same theory is equally inapplicable to the '115 patent. Thus, for Ziegler to establish infringement he must show that the accused Phillips operations either literally infringe the '115 patent or infringe a proper construction of the patent.

As to the starting monomer, we think that a proper construction of this "pioneer" patent encompasses the polymerization of propylene.

First, unlike the '332 patent, the title of the '115 patent ("Polymerization Catalyst") does not limit the claimed catalyst to the polymerization of any particular monomer.

Second, the "Objects of Invention" listed in the specification indicate that monomers other than ethylene were within the contemplated scope of the '115 patent. Particularly, the specification states that an "object of the invention is a polymerization catalyst for lower olefins up to about C5 and particularly ethylene." This language includes the accused Phillips polypropylene operations because propylene is a "lower olefin" within the class of olefins specified by the quoted language. Propylene is an olefin having the chemical formula CH2 = CH-C3 . Having three carbon atoms, it is referred to as a C3 olefin. Thus, since the "object of invention" language discloses that the '115 patent encompasses lower olefins up to and including C5 , propylene, being a C3 , falls squarely within the range of olefins that the '115 patent's catalyst was intended to polymerize.

Third, the language of Example 15 expressly encompasses propylene, speaking of a "dried, oxygen free gaseous mixture of 12 vol. percent propylene and 88 vol. percent ethylene . . . ."

Fourth, unlike the '332 patent with respect to butadiene, the historical background of the '115 patent indicates that the polymerization of propylene with the "Ziegler catalyst" components was well established. For example, at the time the '115 application was filed, Ziegler already had an application pending (No. 514,068 filed June, 1955) for a patent on a process using the "Ziegler catalyst" components of the '332 patent to polymerize propylene. Thus, unlike the '332 patent, extrinsic circumstances do not indicate an intent to limit the '115 patent to the polymerization of ethylene.

Fifth, the '115 patent has been understood and accepted by those skilled in this particular area of organic chemistry as covering catalysts polymerizing lower olefins, specifically including propylene. Major United States corporations have paid substantial royalties to Ziegler under the '115 patent for licenses for the use of these catalysts in processes for polymerizing propylene, processes substantially identical to that of Phillips.

Phillips contends that the predominance of ethylene polymerizations in the '115 patent's Examples indicates that the contemplated scope of the patent was limited to a catalyst for ethylene polymerization. We disagree. The Examples are merely illustrative and should not be construed as limitations on the scope of the claims. See Sterner Lighting, Inc. v. Allied Electrical Supply, Inc., supra; Edward Valves, Inc. v. Cameron Iron Works, Inc., supra. Moreover, nearly all of the references to ethylene in the '115 patent are preceded by the illustrative language "such as."

Thus, we conclude that the District Court erred when it construed the '115 patent as being limited to a catalyst for the polymerization of ethylene or an ethylene-dominated mixture. And since the patent covers a catalyst for the polymerization of propylene, it therefore encompasses an end product of polypropylene.

The final factor to be considered is that of the catalysts and their components. Phillips seeks to differentiate its catalyst on four points, none of which is persuasive.

First, Phillips argues that its catalyst has three components, while the '115 patent is limited to a two component catalyst. The thrust of Phillips' argument is that the "essentially consisting of" language constitutes a file wrapper estoppel limiting Ziegler to a catalyst of two components.

The basic theory underlying the doctrine of file wrapper estoppel is that a patentee cannot recapture through the doctrine of equivalents those claims that he has surrendered, amended, narrowed, or otherwise limited in response to the objections of the Patent Office. Graham v. John Deere Co. of Kansas City, supra; Williams Bit and Tool Co. v. Christensen Diamond Products Co., supra; Dry Hand Mop Co. v. Squeez-Ezy Mop Co., supra. Here, the critical language "essentially consisting of" was inserted in the '115 patent's claims to overcome the Examiner's objection that the original claims were "too broad and based on insufficient disclosure." But Phillips failed to show anything in the file wrapper that indicated that third components were being excluded by this language. Nowhere was it necessary to amend or to restrict the claims to exclude a third component, nor did the Examiner (or Phillips) cite any prior art to show a third component that had to be avoided by excluding it from the claim.

Ziegler argues that the "essentially consisting of" language was inserted to define more accurately and precisely the claimed catalyst as one in which the recited components, the diethyl aluminum chloride and the titanium salt, were essential elements. This revision, according to Ziegler, was necessary to avoid claiming language that would be broad enough to encompass a catalyst system in which these components might incidentally be present and yet not be essential constituents of a polymerization catalyst system.

Significantly, in its Findings of Fact, the District Court simply repeated the wording of the '115 claims and rejected Phillips' proposed language that the "essentially consisting of" phrase constituted "a compelled limitation on the claims, the effect of which was to exclude components not recited therein and which make a material difference in the catalyst formed."

We agree with Ziegler that Phillips' contention stretches both the doctrine of file wrapper estoppel and the quoted language beyond reasonableness. In the absence of any evidence that a third component was being excluded by the "essentially consisting of" language, we cannot read those words as meaning "consisting solely of" or "consisting exclusively of." Rather, the most logical and persuasive construction of the phrase would be to read it as limiting the '115 catalyst to one in which both elements are necessarily present.

Second, Phillips contends that its catalyst is different because it has different components. This argument has two prongs. Phillips' first point is that its use of titanium trichloride takes it out of the wording of the '115 claims, both because the '115 claims do not encompass titanium trichloride and because the '115 claims are limited to titanium tetrachloride. Phillips' second point is that the addition of aluminum trichloride to its catalyst system so changes the catalyst that it is no longer the equivalent of the '115 catalyst. This argument must fall on both points.

As to the first point, the '115 claims unquestionably include titanium trichloride. The precise wording of claims 1 and 2 of the '115 patent requires a "heavy metal compound from the group consisting of salts . . . of metals of groups IV-B, V-B, and VI-B of the periodic system . . . ." Titanium is a group IV-B heavy metal, and both titanium tetrachloride and titanium trichloride are titanium salts. The evidence showed conclusively that this particular form of the titanium salt, the trichloride, has found favor as the heavy metal salt component in Ziegler catalysts and is widely marketed for this purpose. A brochure of the Stauffer Chemical Company states:

"Titanium Trichloride is an important co-catalyst used with Aluminum Alkyls, in the polymerization of olefins. Titanium Trichloride and an Alumimum Alkyl is the Ziegler-Natta catalyst system. The Specialty Chemical Division, Stauffer Chemical Company, produces catalytic grade Titanium Trichloride and other catalysts used in the polymerization of olefins."

It is apparent that titanium trichloride is a titanium salt within the meaning of that term as used in the '115 patent.

Additionally, a reading of the patent does not indicate that titanium tetrachloride was the sole titanium salt intended to be used in the '115 patent's catalyst. Although titanium tetrachloride is the only titanium salt specifically mentioned, the general references in claims 1 and 2 to "salts" indicate that the '115 patent is not to be so limited. Titanium tetrachloride is mentioned in only three of the eighteen claims, and in only eight of the Examples. Other '115 Examples employ such varied salts as zirconium tetrachloride, anhydrous ferric chloride, anhydrous ferrous chloride, anhydrous nicklous chloride, anhydrous manganous chloride, and anhydrous aluminum chloride. Moreover, the specification states that "the term 'salt' or 'salts' designating a compound having a heavy metal of the IV-B, V-B and VI-B groups of the periodic system . . . is employed in its broadest sense, i. e., to connote the reaction product between a base and an acid . . . ." Hence, Phillips' citation of Application of Ziegler, 390 F.2d 762 (3 Cir. 1968), for the proposition that titanium trichloride is not necessarily in all respects the chemical equivalent of titanium tetrachloride is not on point. Because a proper construction of the '115 patent indicates that the patent embraces titanium trichloride, we need not reach this question of equivalency.

As to the second point the effect of adding the third component aluminum trichloride, we do not view this variation as a change of such major import as to avoid the doctrine of equivalents. Rather, the "essence" of the '115 patent's catalyst pervades the Phillips operation, and the function of the doctrine of equivalents is to protect the patentee from those infringers who, without literally infringing, nonetheless appropriate the substance of the patentee's invention. Graver, supra; Chemical Cleaning, Inc. v. Dow Chemical Co., 379 F.2d 294 (5th Cir. 1967), cert. denied, 389 U.S. 1040, 88 S. Ct. 777, 19 L. Ed. 2d 829 (1968).

Defendant's evidence showed only that the Phillips catalyst was a more active catalyst, producing approximately 50 percent more polymer, and that it produced fewer undesirable by-products. At no time did anyone ever testify that the addition of the aluminum trichloride altered the fundamental polymerization action of the Ziegler components. Rather, the thrust of the testimony was directed to how much better the Phillips catalyst worked. Indeed, the record testimony, uncontradicted by Phillips' expert witnesses, was that no polymerization would occur unless both Ziegler components were present. In view of this evidence and testimony, we think that the District Court mistakenly viewed the Phillips catalyst as not performing the same function in substantially the same way. The testimony is clear that the Phillips catalyst performs the same function (polymerization of propylene) in the same way (with a catalyst relying essentially on the Ziegler combination) but in a better fashion (more polymer and less undesirable by-products). This catalyst, though, does not avoid infringement. Without doubt, Phillips' catalyst is an improvement, but an improver does not escape infringing the dominant patent just by improving it. Temco Electric Motor Co. v. Apco Mfg. Co., supra; Williams Bit & Tool Co. v. Christensen Diamond Products Co., supra. Here, the evidence is clear that Phillips is an infringing improver.

Third, Phillips argues that its catalyst is different because it is mixed and reacted in the presence of the monomer that it polymerizes. Phillips contends that the '115 patent discloses only a catalyst that is mixed and reacted beforehand, out of the presence of the monomer on which it operates. After reviewing carefully the testimony of Phillips' expert witnesses and the claims and specification of the '115 patent, we must disagree. Remembering that the '115 patent is entitled to a broad construction as a "pioneer" patent, we must also remember the rule that a patent is not to be limited to the preferred embodiments shown in the specification. See Continental Paper Bag Co. v. Eastern Paper Bag Co., supra. The claims here contain no specific limitation requiring preforming, and the Examples do not exclude the alternative method of combining the Ziegler catalyst components in the presence of the monomer to be polymerized. Thus, we need not hold that pre-forming and post-forming are equivalent, because we hold that the '115 patent compels neither and permits both.

Fourth, Phillips contends that Ziegler is precluded by the file wrapper estoppel doctrine from asserting claims 2, 15, and 16 against the polypropylene operation. These claims state that the catalyst is formed by mixing the aluminum compound with the heavy metal compound "in an inert organic solvent with at least one of said aluminum compounds and said heavy metal compounds in solution in said solvent." The original claim said merely "solvent," but the Examiner objected to that term as too broad, pointing out that all of the solvents disclosed in the claims were inert solvents. This contention is meritless. The evidence before the District Court indicated that, before being put into the reactor, both the titanium salt and the diethyl aluminum chloride are dissolved in the inert organic solvent pentane.

In summary, although Ziegler is not entitled to "all uses" of the polymerization catalyst claimed in the '115 patent, a proper construction of the patent indicates that it claims a catalyst for the polymerization of propylene, and the catalysts themselves are substantially equivalent in all essential respects. Hence, we hold that the Phillips Adams Terminal and Monument processes infringe the '115 patent. This cause must be remanded for further proceedings consistent with that holding.

We affirm the District Court's finding that both the '332 and '115 patents are valid. Phillips' basic argument as to invalidity is that, given Ziegler's "all uses" construction, both patents would be invalid for lack of disclosure. See Foster Cathead Co. v. Hasha, supra. Because we reject the "all uses" theory as applied to these patents for chemical catalysts, Phillips' argument is inapposite. Moreover, it does not apply to our construction of the '115 patent, because we hold that the catalyst employed in the infringing Phillips polypropylene operations is substantially equivalent to the catalyst disclosed in the '115 patent.

Additionally we conclude that as construed, the '115 patent is not invalid as anticipated by the prior art.

Costs of the appeal are to be taxed equally, each party paying one-half. We affirm the District Court's allocation of costs.

Affirmed in part and reversed and remanded in part.