418 F2d 937 Kilby v. Nelson

418 F.2d 937

Jack St. Clair KILBY, Appellant,
Herbert NELSON, Appellee.

Patent Appeal No. 8206.

United States Court of Customs and Patent Appeals.

December 4, 1969.

Ellsworth H. Mosher, Washington, D. C., attorney of record, for appellant;

Samuel M. Mims, Jr., Dallas, Tex., Stevens, Davis, Miller & Mosher, Washington, D. C., of counsel.

A. Sherman, Princeton, N. J., for appellee; A. Russinoff, Princeton, N. J., of counsel.

Before RICH, Acting Chief Judge, MATTHEWS, Judge, sitting by designation, ALMOND, BALDWIN and LANE, Judges.

ALMOND, Judge.

view counter

This appeal is from the decision of the Board of Patent Interferences awarding priority of invention of counts 1-4 of Interference No. 92,840 to Nelson, the senior party. The counts correspond to claims 1, 3, 7 and 8, respectively, of Nelson patent No. 2,972,092, issued February 14, 1961 on an application filed August 11, 1959,1 which claims were copied by Kilby in his application serial No. 169,557, filed January 29, 1962.


No testimony was taken by either party and the senior party Nelson is restricted to his August 11, 1959 filing date for conception and reduction to practice. Before the board, Kilby relied for priority on two prior applications of which his involved application is stated to be a continuation-in-part but, before us, he limits his reliance to just one of them, serial No. 811,486 (the '486 application), filed May 6, 1959.2


Nelson contended before the board that none of the Kilby applications3 support the counts and the board agreed with that contention. In determining whether the board was guilty of reversible error, we conclude that the '486 application does not support the counts and that the board's decision must be affirmed for that reason.


The invention relates to a semiconductor circuit element, such as a diode or transistor, of the mesa type. The record shows that a mesa type element may be formed by diffusing into a semiconductive wafer of one conductivity type, such as P-type, an active impurity which will convert the surface of the wafer to the opposite conductivity type, such as N-type, and provide a rectifying P-N junction at the interface between the surface zone and the wafer interior. Thereafter a portion of the wafer surface is removed down to and including the junction region to leave the remaining portion of the surface zone of N-type conductivity extending like a plateau or mesa from the remaining portion of the wafer of P-type conductivity. The size and the shape of the P-N junction of the device between the mesa and the wafer corresponds to the size and shape of the mesa and may be controlled with precision.

Counts 1 and 4 are representative:


1. A circuit element comprising a semiconductor wafer having two opposed major faces; at least one mesa of semiconductor material on one of said major wafer faces; an insulating coating on said one major wafer face around each said mesa; and a conductive film over the top of, and in contact with, each said mesa and the surrounding coating.


4. A two-terminal device comprising a semiconductor wafer having two opposed major faces; a surface zone of given conductivity type on said wafer including one of said major faces; an opposite conductivity type mesa on said other of said major faces; a rectifying barrier between said mesa and said surface zone; an insulating coating on said other major face around said mesa; a conductive film over the top of, and in contact with, said mesa and the top of the surrounding coating; and electrical leads to said conductive film and to the oppsoite major face.


For an understanding of the invention as disclosed by Nelson, reference is made to a sectional view shown in Fig. 2 of his patent and reproduced in enlarged scale in his brief as follows:

view counter



This figure shows a diode comprising a P-type germanium wafer having an N-type mesa projecting upwardly therefrom with a P-N junction between the mesa and the bulk of the wafer. The element is extremely small as evidenced by the disclosure that the mesa may measure less than 2 mils in diameter. An insulating coating 26 is deposited around the mesa and then a conductive film 27, with a thickness of about 0.1 to 0.5 mil, is evaporated over the top of the mesa and a portion of the insulating coating 26. This film forms a conductive member which has a lateral area considerably greater than the mesa. To complete the diode, a conductive pellet 21, which may also have a greater lateral area than the mesa, is bonded to the conductive film 27, a conductive pellet 22 is bonded to the opposite face of the unit, and lead wires 23 and 29 are attached to electrodes formed by pellets 21 and 22, respectively.


The structure disclosed in the Kilby '486 application is illustrated in plan and cross-sectional views in Figs. 1 and 3, respectively, thereof:




The figures show, in greatly enlarged scale, a miniaturized electronic module made up of interconnected elements, including a transistor designated generally as 14. The module includes a block of semiconductor material 19 which may be of N-type. The transistor which is formed within and upon the block comprises a portion of the block of N-type conductivity together with a mesa portion made up of layers 21 and 22 of semiconductor material which are of conductivity types respectively opposite to and similar to the type of the main body of the block. The layer 21 constitutes the base region of the transistor and the upper layer 22 the emitter region. A portion of the block 19 itself forms the collector. After the block with the layers 21 and 22 is formed, a coating 20 of insulating material is provided over the entire block. Thereafter, small apertures are etched in the coating at locations for the emitter, base and collector contacts or electrodes shown at 12, 13 and 11 as by using a mask in conjunction with a photo-resist process. It is disclosed that the electrodes may be formed by evaporating or otherwise depositing "suitable ohmic-contact-making material" ("antimony-doped gold or other suitable material" for the emitter and collector and a "material such as aluminum" for the base) in the apertures and the entire assembly heated to a predetermined temperature "at which the deposited material alloys with the base, emitter, and collector to form severally distinct ohmic contacts therewith." Next, a film of highly conductive material, "such as copper or gold," may be applied by vacuum deposition techniques to those areas shown in Fig. 1 with a cross hatch line extending upwardly to the right, including areas 15, 17 and 23 connected to collector, base and emitter electrodes 11, 13 and 12, respectively.


In reaching the conclusion that the Kilby applications do not support the counts, the board held unsupported the term in count 1 reading:


* * * and a conductive film over the top of, and in contact with, each said mesa and the surrounding coating.

The board's reasoning was as follows:


In our opinion the term "in contact with" is normally used to denote touching physically. In the electrical arts a "contact" is used to provide an electrical connection with a cooperating part or "contact." However, the electrical connection connotation urged by Kilby is not a normal one. It is true that an electrical connection results from physical contact or touching of a conductive body with another such body, but this does not mean that every electrical connection between two components of an electrical circuit can properly be the basis for stating that the two parts are in contact with each other. The two parts may be spaced a considerable distance from each other and connected by a wire or bus.


Viewing count 1 as a whole it is considered to be a simple and straightforward description of the structural embodiment disclosed in the Nelson patent. It describes the mesa as being of semiconductive material and it states that a conductive film is "over the top of, and in contact with," the mesa and surrounding coating. We think that this clearly indicates an intention that the conductive film and the mesa top be physically touching. Clearly, Kilby's conductive film is in contact with an intermediate metal button or contact plate, not with the mesa, which is of semiconductive material. The natural way to broadly claim merely an electrical connection between the film and the top of the mesa would be by requiring the conductive film to be over the top of and electrically connected to the mesa.


In addition to arguing in support of the board's position, Nelson urges that the Kilby applications fail to support the immediately preceding recitation in count 1 describing the insulating coating as "around each said mesa."


Kilby disagrees with the board's interpretation of the counts, urging that the limitation cited by the board demands "nothing more than that there shall be an electrical contact between the conductive film and the mesa and the surrounding coating." He states that the counts are not ambiguous and that the issue here revolves around the "holding" of the board "that the claims copied from the Nelson patent require the insertion of the limitation physical into that part of the counts which reads `and a conductive film over the top of, and in contact with, each said mesa and the surrounding coating.'"


Thus, the basic issue here involves the meaning of "contact" as used in the counts. While Webster's New International Dictionary of the English Language, Unabridged, 2nd Ed., G. & C. Merriam Co., 1949, sets forth ten definitions of that word, all that need be considered here are those designated 1 and 6 and reading as follows:


1. A union or junction of bodies; a touching or meeting; as, a disease communicated by contact; contact with the air.


* * * * * *


6. Elec. The junction or touching surface of two conductors through which a current passes; also, a special part, as a platinum stud, made for such a junction for temporary or momentary connection. Cf. ARCING CONTACT.


Despite Kilby's arguments, we are in full accord with the analysis of the board quoted hereinabove. In our opinion, the statement that "the conductive film is in contact with" both the "mesa" and "the surrounding [insulating] coating" plainly dictates that the word "contact" has the same meaning as applied to both the mesa and the coating. As applied to the coating, "in contact with" necessarily means "a touching or meeting" in the physical sense since the coating is of insulating material and cannot be part of a junction or touching surface of two conductors through which a current passes. Thus, the normal, and at the same time the broadest reasonable, understanding of the language is that no less than physical touching is required in both cases. Such touching additionally provides a connecting surface through which current passes in the case of the contact with the mesa of semiconducting material.4


While we think our reasoning is warranted even if the counts are deemed unambiguous and are interpreted without reference to the Nelson disclosure, the fact that Kilby urges that different interpretations of "contact" are applicable may properly be treated as justifying analysis from the viewpoint that ambiguity is involved. See Smith v. Wehn, 318 F.2d 325, 50 CCPA 1544 (1963). Therefore, interpretation of the counts in the light of the Nelson patent in which they originated is appropriate. Such consideration reinforces the conclusion that a physical contact or touching is required between the conductive film and the top of the mesa. Thus, Nelson attains a construction where connection to the very small mesa area, without short-circuiting the P-N junction, is facilitated by having the mesa surrounded by an insulating coating which is as thick as the mesa height, but does not cover the top of the mesa, at the time the conductive film is applied. That construction results in the film being in physical contact with the entire top of the mesa and the surrounding coating. It avoids any need for the delicate task of etching an extremely small aperture accurately through an insulating coating on top of the mesa, which itself is of very small lateral area, and depositing a contact-forming metal therein prior to depositing the conductive film.


Kilby also argues that the button-like electrodes of his structure, as 12 and 13, should be considered as part of the mesa or, alternatively, as part of the conductive film. With respect to considering the electrodes 12 and 13 as part of the mesa, he refers to the law of fixtures as it pertains to personalty such as machinery firmly attached to the earth or a building being legally annexed to real property. That point is entirely too remote to even suggest consideration of the semiconductive mesa and the conductive metallic electrodes as one and the same.


As to the alternative argument, Kilby refers to the possibility of "the use of a common material for producing a rectifying junction, such as aluminum, and, then, because aluminum is an excellent electrical conductor, using aluminum as the conductive film." He states that the first deposited portion of aluminum would be the "button" and the second deposited portion the conductive film, and asks "who could say which was the `button' and which was the conductive film * * *?" Kilby further argues that the conductive film and the electrode between it and the semiconductor, when made of dissimilar materials, would be "mechanically, electrically and chemically" connected together.


The first of those contentions involves a hypothetical construction not disclosed in Kilby '486, where the electrodes or "buttons" are described only as making ohmic-contact in contrast to a rectifying junction. Moreover, there is no justification in any event for regarding the electrodes 12 and 13 of Kilby as being but a part of the conductive film. The fact is that these electrodes are shown as separate from the film and are described as separate. Thus the electrodes are described as "distinct ohmic contacts" and the connections to them are described as "highly conductive films."


Finally, Kilby contends that the board has ignored the "`gist of the invention' doctrine applicable in interferences involving a patent and an application based on claims copied from the patent." It is true that consideration of the "gist of the invention" has been found useful in resolving the question of support for counts copied from a patent in certain circumstances, notably in Hall v. Taylor, 332 F.2d 844, 51 CCPA 1420, involving limitations in the counts to certain range limits. However, we think the basic rule that an applicant who copies a claim from a patent must show that he is entitled to make the claim and that all limitations in the copied claim will be considered material in determining applicant's right to make the claim, Smith v. Wehn, supra, and Segall v. Sims, 276 F.2d 661, 47 CCPA 886 (1960), compels the conclusion that the Kilby '486 application does not support the present counts. Moreover, it seems to us that consideration of what the gist of the claimed invention here is can only lead to the determination that it constitutes the improved structure which offers the previously set forth advantages in making the connection to the mesa of minute area and is defined by the counts in a manner not supported by the Kilby '486 construction.


Since we are convinced that the only application antedating Nelson's record date which Kilby relies on does not provide support for the counts, the matter of whether the directly involved Kilby application provides such support is moot.


The decision is affirmed.





Serial No. 833,031 for "Semiconductor Devices."


Patent No. 3,138,744 issued on that application on June 23, 1964


Kilby's right to make in his directly involved application was raised initially through a motion to dissolve by Nelson, which motion was denied by the examiner


An interesting analogy exists between this case and Rajchman v. Lockhart, 339 F.2d 233, 52 CCPA 853 (1964), wherein the court agreed with the board's refusal to assign a purely magnetic meaning, as contrasted to a meaning requiring physical linking or embracing, to the word "linking" used in conjunction with windings on a magnetic core