Escolar Documentos
Profissional Documentos
Cultura Documentos
INTEGRATED CIRCUITS
Untold Important Facts
This page intentionally left blank
ASSET
International Series on Advances in Solid State Electronics and Technology
Founding Editor: Chih-Tang Sah
INVENTION OF
INTEGRATED CIRCUITS
Untold Important Facts
Arjun N Saxena
Emeritus Professor & Patroon
Rensselaer, USA
World Scientific
NEW JERSEY • LONDON • SINGAPORE • BEIJING • SHANGHAI • HONG KONG • TA I P E I • CHENNAI
Published by
World Scientific Publishing Co. Pte. Ltd.
5 Toh Tuck Link, Singapore 596224
USA office: 27 Warren Street, Suite 401-402, Hackensack, NJ 07601
UK office: 57 Shelton Street, Covent Garden, London WC2H 9HE
For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance
Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA. In this case permission to photocopy
is not required from the publisher.
ISBN-13 978-981-281-445-6
ISBN-10 981-281-445-0
Disclaimer: This book was prepared by the authors. Neither the Publisher nor its Series Editor thereof,
nor any of their employees, assumes any legal liability or responsibility for the accuracy, completeness,
or usefulness of any information. The contents, views, and opinions of the authors expressed herein do
not necessarily state or reflect those of the Publisher, its Series Editor, and their employees.
Dedication
v
February 20, 2009 9:39 B732 9in x 6in b732-fm FA
This book is the first and only in the literature to give a technically authoritative and
historically in-depth, correct account of the invention of the silicon integrated circuit,
because the invention has been attributed to two inventors, each from a combating
manufacturer in rivalry, eventually with compromised stipulations ruled by the judge in a
settlement reached in court, compounded subsequently by popularity recognition of the
U.S.Presidential National Technology Award in engineering and also the World's Nobel
Prize in Applied Physics, all of which further confused the facts underlying the invention
of the integrated circuits. It is hoped that this monograph puts an end to the stipulations
and the reader-attracting tales about the IC invention history and its two rivalry and
colorful personal-life inventors. The invention of IC is quantitative engineering technology,
thus, amenable to precise solution rather than uncertain extrapolations by those without
the hands-on information from personal involvement in the events of fifty years ago.
This book presents the history as a detective story, by the person who was involved and
was on the scene, and who has continually kept track. But at the end, also discovered
discrepancies and ambiguities in the incomplete historical records, even appeared
meddled intentionally, which have not been known for fifty years, and which would take
further efforts using modern techniques from using the integrated circuits themselves to
figure out, if at all possible, showing the superior of human intelligence in disguise, not
surmountable by human-developed diagnostics techniques using the very integrated
circuits themselves. But nevertheless, after reading this book, we will all have learned
the documented historical truths rather than the grapevine propagated tales originated
from those who had not even the vigorous semiconductor and transistor physics
classroom background necessary to judge and delineate.
vii
February 20, 2009 9:39 B732 9in x 6in b732-fm FA
Preface
The main reason I am writing this book is to get history right and set
the records straight on the invention of the integrated circuits (ICs). Getting
history right can be a monumental, daunting and an unrewarding task. This
is especially true if the subject matter is of fundamental importance and
wide popularity having direct impact on the bottom line of a huge market.
The task is compounded further if its prevailing view has been misleading,
and has lasted for a long time, e.g., for more than four decades in this
case of the invention of ICs. Another dimension is added to its monumental
aspect when world caliber and colorful personalities in highly respected
institutions with impeccable records are involved in this whole saga of the
mystery of the IC invention. Almost everybody in the microelectronics field
involving physics, chemistry, engineering etc in the entire world appear
to have accepted the erroneous information of the IC invention for more
than four decades because they have done nothing so far to correct it.
Nevertheless despite the daunting aspect of the task, I feel that it needs to
be done because it is important to get history right. The previous accounts
of the invention of ICs given by several science history writers have been
erroneous. I shall give all the relevant and well documented facts to set
the records straight, and get history right for the invention of ICs in this
book. These facts can be verified by anybody, as it is the accepted hallmark
procedure in good scientific research: any result claimed must be verifiable
by independent workers. However, this book has not been written like an
esoteric thesis. It reads like a mystery novel, but the intrigues and their
solutions are given in an easy to read style based on well established facts.
The readers will find it amazing that the real truth of the IC invention
has been swept under the rug earlier by the powers to be in a manner
unparalleled in the history of mankind. A capsule of the details given in
various chapters is presented in this Preface. I hope that the brief discussion
given below will make a compelling case for reading and enjoying this book,
ix
February 20, 2009 9:39 B732 9in x 6in b732-fm FA
and to learn about the facts of this historic invention which has changed
mankind forever.
For the sake of convenience of the readers, I shall repeat the discussions
of some of the important facts and published documents in appropriate
places to avoid going back and forth in this book. Since the misleading
information on the invention of ICs has lasted in the literature for such
a long time, it is particularly important to give credible documentation
and discussions even repeatedly. This is meant to convince the readers
and rectify the misinformation that has been prevailing all these 40+
years all over the world. Also those who may not be familiar with the
various scientific abbreviations, acronyms and symbols, I shall repeat them
a few times so that the readers will become conversant with their usage in
this book. However, all the abbreviations, acronyms and symbols are also
defined and listed in Appendix 6.
Preface xi
Section 3. What is an IC?
Earlier, the industrial revolution, the printing press, and the agricultural
revolution also revolutionized the human life. However, even these earlier
outstanding innovations have needed the enhancements enabled by the
present ICs used in their control systems, without which their efficiencies
and efficacies cannot be harnessed as well as it has been attained today.
Almost no business can operate, and to make a bold but true statement,
almost no human at least in the developed countries can carry on with their
lives without an IC. Its use is indispensable worldwide in many applications
(partial list given alphabetically): banking, biotechnology, communications,
computers, education, entertainment, government, hospitals, internet,
medicine,nanotechnology, research, travel, and in almost every commercial,
defense and industrial businesses. All the electronic systems in these
applications currently use Ultra Large Scale ICs (ULSICs) which are much
more advanced and complex than the ICs invented originally. However,
the stems of all ULSICs are rooted in the basic invention of the ICs 50
years ago. The ULSIC business has now grown to multi-hundred billion
dollars annually, which is the heart and soul of the entire electronic systems
market place of several trillion dollars per year. The growth of this ULSIC
business and its impact on the society and newer businesses shall forever be
increasing. Therefore it is important to know what were the basic inventions
of the ICs, who invented them, when and how.
3. What is an IC?
The sole credit for the invention of ICs has been given in the literature
to late Jack Kilby and late Bob Noyce after their inventions were disclosed
in their respective patent filings at the US Patent and Trademark Office
(USPTO) about 50 years ago. Unfortunately this recognition accorded to
Kilby and Noyce is only partially correct and justifiable. Most of the other
authors on this subject have failed to appreciate the fallacy of this exclusive
recognition. Indeed Kilby and Noyce did play key roles in the invention of
ICs, but not quite the way they have been portrayed in the literature.
For example, they have been credited in folkloric manner as if they had
invented the IC all by themselves, thereby given an iconic stature by the
hero worshippers. Moreover, some of the important facts have either been
overlooked or not understood at all by the previous authors and those
publicizing such folklore. These facts have been documented in this book.
They tell an accurate but different story about the real contributions of
Kilby and Noyce.
The readers will find here in my book, that Kilby’s descriptions and
documentations underlying the concepts of the ICs were incomplete. He
had missed the key requirements which are: the devices must be connected
also by monolithic interconnects adherent to the surface layers (not by
gold wires bonded to the devices and flopping in the air above the chip
February 20, 2009 9:39 B732 9in x 6in b732-fm FA
Preface xiii
Section 5. Key Points to Get History Right
which Kilby had used), and the planar (not mesa) technology must be used
to fabricate and electrically isolate the devices in a piece of single crystal
semiconductor such as Si.
The key points given briefly in the subsequent sections of this Preface
are very important to get history right about the invention of ICs. They
February 20, 2009 9:39 B732 9in x 6in b732-fm FA
will become clearer after reading their details in the book. Some of them
will startle and surprise many people, including the scholars. But my key
objective is not to sensationalize such an important invention. I shall present
the facts many of which have not been reported earlier in the literature
correctly. These facts have been derived by me from Kilby’s and Noyce’s
published documents such as their US patents and journal articles, and my
communications with other authoritative sources such as the United States
Patent and Trademark Office (USPTO), Nobel Committee members, and
several other original contributors to the invention of ICs. Some of the facts
are also based on my first hand knowledge of participating and contributing
in the historic events from the inception of the invention by Bob Noyce to
the subsequent advancement of the ICs when working in R & D at Fairchild
directed by Gordon Moore, and elsewhere. The technical nature of the
facts given below can be understood better by the scientists and engineers
in the microelectronics field, than by the laypersons. Nevertheless even
the laypersons can appreciate the significance of most of these facts after
reading this book. All the readers will find that the saga of the invention
of ICs is quite different from what has generally been described in the
literature. It is amazing that the entire popular story of the invention of
ICs has the makings of a mystery novel. But this book is based on facts,
not on fiction; hopefully the readers will find it engrossing, educational and
entertaining as well.
In order to set the stage for this book, the key points of Kilby’s and
Noyce’s inventions of the ICs are summarized at the outset. If some of them
come to you (the reader, even the experienced ones in the IC profession)
as a surprise, please do not get alarmed; be inquisitive with an open mind
and learn from the facts given in this book.
6.1. Kilby did not invent the monolithic-ICs made from silicon (Si), the
only kind sold from the very beginning in the microelectronics business in
1960 and onwards. (An exception to this statement is the microwave IC.)
This fact was also confirmed by a Nobel Committee member in his recent
written communications with me regarding Kilby’s invention and the Nobel
Prize in Physics awarded to him in 2000. It will surprise particularly many
February 20, 2009 9:39 B732 9in x 6in b732-fm FA
Preface xv
Section 6. Key Points of Jack S. Kilby’s Invention
of the diehards who are of the firm opinion that the Nobel Committee
decisions to give the awards in Physics are always non-controversial. They
assume that such decisions to select the Nobel Prize awardees must be
based on precise incontrovertible principles of physics.
6.4. Kilby had used gold (Au) wires bonded to the devices made in
germanium (Ge) pieces to interconnect them, flopping over the chip from
one device to another to complete his IC. The monolithic interconnects
to connect electrically the various devices on a chip, which should also be
adherent to the insulator layers, were neither specified correctly nor used
by Kilby. It was adjudged by the Board of Appeals of USPTO to be a key
omission by Kilby. This fact alone, over and beyond a few other serious
drawbacks in his patent(s), had cost Kilby the sole ownership of the IC
invention.
beginning in 1960 and onwards. If anybody who should have been crystal
clear about the monolithic concept, it should have been Kilby, because the
company he had worked for, Texas Instruments (TI) was among the first
to sell the silicon (Si) monolithic-ICs in the entire market at the time and
soon after filing of the Kilby patents.
6.7. Kilby’s patent(s) were awarded a few years after Noyce was awarded
his patent, even though Kilby had filed earlier than Noyce. The filing date
of Kilby’s Original Application (OA) in 1959 appears never to have been
resolved, as evidenced by the recent communications of USPTO with me in
2005 (46 years after the original filing). In addition, it appears that the laws
of the US Patent code 35 USC 112 and associated protocols were possibly
compromised in the procedures used by USPTO to issue the patents to
Kilby, as documented by their filing dates, specifications and claims.
6.8. The quintessence of Nobel’s Will is that the Nobel Prizes shall be
awarded to person(s) irrespective of nationality who, during the preceding
year has (have) conferred the greatest benefit on mankind. For detailed
discussions of the Nobel Prizes in Physics awarded throughout their entire
history, and for the Nobel Prize given to Alferov, Kroemer and Kilby, see
Appendix 7. See chapters 1 and 12 also for the technical issues related to
the IC invention. Nobel Prize is not awarded in the field of Engineering. The
Nobel Prize in physics 7 was awarded in 2000 jointly to Alferov, Kroemer
and Kilby, and their names were listed in this order. Noyce had died in 1990;
the Nobel Prizes are not awarded posthumously. Had Noyce not died, in my
opinion he would have been definitely included in this award. Perhaps then
the entire Nobel Prize award would have been worded and even awarded
differently. The inventions of Kilby and Noyce did not involve any basic
contributions to physics. However, the work of Alferov and Kroemer did
involve basic contributions to physics.
The Nobel Prize money was distributed as 1/4 each to Alferov and
Kroemer, but 1/2 to Kilby. The citation of the Nobel Award to Kilby did
not explain what was his part to invent what kind of IC? Kilby’s citation
in the Prize was incomplete and inconsistent with his contribution to the
purported invention of monolithic-ICs for which he was given the Nobel
award. No explanation has been offered so far by the Nobel Committee
February 20, 2009 9:39 B732 9in x 6in b732-fm FA
Preface xvii
Section 6. Key Points of Jack S. Kilby’s Invention
for the vague and imprecise citation of such an important world renowned
Prize.
Committee. These facts, however, do not clarify the actions of the Nobel
Committee. (See Appendix 7 for details.)
Preface xix
Section 7. Key Points of Robert (Bob) N. Noyce’s Invention
Most of the key points of Noyce’s invention of ICs given below are not
as surprising as they have been given above for Kilby. Nevertheless some of
the details of Noyce’s invention also have not been covered in the literature
properly. Some of the key points of Noyce’s invention are summarized as
follows.
7.1. There is no controversy about the filing and issuance of the key
patent to Bob Noyce for his invention of the IC. Late Bob Noyce, or perhaps
more appropriately his patent attorney, late John Ralls who was the chief
architect to write his patent, says it all regarding the invention in the
very first paragraph of his patent. He had specified all the key processes
and materials required for fabricating the monolithic-ICs, in particular the
planar technology with silicon (Si), and the monolithic interconnects of
aluminum (Al) adherent to the oxide layers. He had described p-n junction
isolation in his patent and it was used in the reduction to practice of his
invention, but he did not claim it. I have discussed the latter point with
Sah 69 and I have given the technical reasons in Chapter 6 for why Noyce
may not have claimed the p-n junction isolation in his patent. Noyce’s
patent describes essentially how the monolithic-ICs are still made with
silicon, although most of the processes and materials used currently are
much different and more advanced than those given originally.
7.2. Noyce was the top boss at Fairchild, which of course meant that
everybody else including Moore, Hoerni, Last and the others were working
for him. Noyce’s invention and its patent were based on the disclosure in
his notebook which was not witnessed. Some of his co-workers had felt that
he had usurped their contributions which had made the ICs a reality, and
that Noyce had written in his notebook in a hurry to claim it all himself.
February 20, 2009 9:39 B732 9in x 6in b732-fm FA
7.3. Noyce had used Hoerni’s invention of the planar technology and
Lehovec’s invention of the p-n junction isolation, but did not acknowledge
them in his patent. The issue of the sole inventorship of the planar
technology by Hoerni has been discussed briefly in section 3 above. To
repeat, Hoerni had combined all the information from the work of the
others elsewhere earlier than him and at Fairchild, several engineers and
technicians other than Hoerni reduced his invention to practice, but Hoerni
filed and received his patents to claim the planar technology invention. This
is another mystery added to the saga of the invention of ICs by Kilby and
Noyce which has also not been discussed before in the literature. However,
its details are discussed in Chapter 6.
The sole credits for their respective inventions, coupled as the only
two co-inventors of ICs, have been bestowed to Kilby and Noyce for all
these years in the technical and popular literature. The above points
regarding their invention clearly warrant clarifications of what actually
their contributions were, and what did they actually invent? Much of
the information perpetuated all this time for more than four decades, has
been misleading and gives a distorted view of who did what? Additional
incongruous events have accentuated the mystery which needs unfolding by
incontrovertible documented and other facts from credible sources.
8. Other Contributors
Efforts other than those of Kilby and Noyce to invent the ICs shall also
be described briefly in this book. There were several other inventors and
contributors who had given their concepts for the invention of ICs earlier
than Kilby and Noyce. It is rather unfortunate that an accurate and a
thorough account of their contributions along with Kilby’s and Noyce’s,
has not been published previously by the other experts and scholars in
the microelectronics industry, academia and patent law. I have published
a few papers recently which give the issues and facts of the inventions of
ICs by Kilby and Noyce. I could describe the roles and contributions of
the others to the invention of ICs only briefly in them, because of the
February 20, 2009 9:39 B732 9in x 6in b732-fm FA
Preface xxi
Section 8. Other Contributors
Preface xxiii
Section 10. Historical Perspective and Accuracy
very purpose of this book is to present the history from the positive point
of view to document the truth of the invention of IC.
For the sake of convenience of all the readers, especially the scholars, I
shall give in this book some of the key original patents and papers which
may not be easily accessible to some of you. I shall also analyze each claim
of a few of these patents which, to the best of my knowledge, have not been
published before. Nevertheless these analyses and interpretations of the
claims are strictly mine, so I regret sincerely if some errors and omissions
may have crept in unwittingly.
Further, I shall not only give all the facts comprehensively in this book
to prove unequivocally within their realm who invented what in ICs, but
I shall also present a few questions at the end on the invention of ICs.
Surprisingly they have remained as yet unanswered even after 48 years
since Bob Noyce’s patent was issued in 1961, and the IC invention had
become public knowledge. I hope that all the readers, experts and laymen
alike will find the pedantic historical accounts given here to be educational
and thought provoking as well. The latter is particularly interesting when
one realizes that many a participant (technical contributors as well as non-
technical contributors), who have played key roles from the inception and
during the progression up to the ULSICs, have remained silent rather than
to publish and set the records straight earlier. The silence on the part of
some may be understandable if they did not have first hand knowledge or
if they were not involved directly in the various key events.
I have given all the facts in this book based on the various documents
I could obtain so far, which is more than ever done before in the literature.
However, I will concede at the outset that some of these facts may still be
incomplete. As an example, despite my best attempts to get the case history
files of Kilby’s and Noyce’s patents from the US Patent and Trademark
Office (USPTO), I have not yet been able to get them so far. (See Chapter
15 on Conclusion.)
Preface xxv
Section 11. My Experience
entire business history of the world. It is natural for the entire society to
expect them to publish and document the facts that they were privy to
during these exciting years. Or they should at least speak up and correct
the wrong information which had been given about the key inventions for
more than four decades. But why haven’t they done so? This fact adds to
the unanswered questions which are discussed in this book.
11. My Experience
knowing them all personally. At Stanford, I had received the single crystals
of Si from Texas Instruments in 1956-57 before Jack Kilby had joined
it in 1958.
Preface xxvii
Section 14. Acknowledgements
Unlike several of the other authors who have written about the
Invention of ICs, I would like to state up front that I have neither solicited
nor received any financial support from various sources to write this book.
All of the facts presented by me about the invention of ICs in this book are
well documented and available in the public domain. I have, however, also
used the inputs on the invention of ICs from a few key individuals which
have not been published before. In this regard, I have used my judgment
and professional discretion not to disclose some of their names. However
in most such cases, I have the inputs shared with me in writing, e. g., via
letters and e-mails. My aim is to put all these facts of the invention of ICs
on historical record, and interpret them objectively and accurately to bring
out the truth as an insider to this whole saga of their birth and progression
to the ULSICs. The fervent desire to get history right for the benefit of
everyone: all professionals, laymen and laywomen, especially the younger
generation, has helped me in overcoming many an impediment, disparaging
moments, even portentous reactions and diatribe.
14. Acknowledgements
and comments. It is almost impossible to list them all. A few key ones
to whom I am grateful are as follows (listed alphabetically): Julius Blank;
Federico Faggin; (late) S. N. Ghoshal; Jay Last; Kurt Lehovec; Toshiaki
Masuhara; Dan Maydan; Gordon Moore; Bob Norman; (late) Bob Noyce;
(late) W. K. H. Panofsky; Sheldon Roberts; Chih-Tang (Tom) Sah; Simon
Sze; Rob Walker. I wish to single out Tom Sah particularly to acknowledge
with thanks his invitation to write this book for his ASSET series of books,
and critical reading of the manuscript to affirm the accuracy of all the facts
documented by me. I appreciate helpful comments on the patent and legal
issues from Roger Borovoy and Gideon Gimlan. I am also grateful to my
daughter-in-law, Mrs. Karen Saxena, for her help with the computer, and
to my wife, Mrs. Veera Saxena, for her forbearance and support.
Arjun N. Saxena
Contents
Foreword vii
Preface ix
Chapter 1: Introduction 1
1. What is an IC? 3
2. Key Requirements for making the IC and did
Kilby and Noyce meet them? 4
xxix
February 20, 2009 9:39 B732 9in x 6in b732-fm FA
1. Discovery 32
2. Invention 32
3. Improvement 36
4. Patents 37
5. Publications 38
6. Trade secrets 39
February 20, 2009 9:39 B732 9in x 6in b732-fm FA
Contents xxxi
1. Summary of IC fabrication 69
2. Summary of the inventions of integrated circuit
by Kilby and Noyce as documented in the
literature 71
3. Sequence of relevant patent filings and issue
dates 78
February 20, 2009 9:39 B732 9in x 6in b732-fm FA
Contents xxxiii
Contents xxxv
1. Kilby 313
2. Noyce 317
Contents xxxvii
1. Conclusions 355
2. Combined summary 358
3. Historical facts 362
4. Unanswered questions 368
USPTO response: File History of Kilby’s patents
is lost 370
5. Epilogue 374
1. Introduction 465
2. Will of Alfred Nobel 468
3. Updated summary of the Nobel Prize award 473
4. The field of Engineering missing from the
Nobel Award list 474
5. Difficult job of Nobel Committees; example of the
Nobel Prize to Sir C. V. Raman 475
6. History of all the Nobel Awards in Physics from
the very beginning in 1901 to 2007 476
7. The entire list of all Nobel Prize in Physics
winners to 3-person awardees from the
very beginning with their citations and
distributions of the award money 480
8. Overall comments on the sequence of listing and
the distribution of the financial amounts
of 3-person awardees throughout the
history of Nobel Awards 486
9. Discussion of the Sequence of Listing and the
Distribution of the Financial amounts
in the Nobel Prize Awarded in Physics
in 2000 to Alferov, Kroemer and Kilby 487
10. Conclusion 492
February 20, 2009 9:39 B732 9in x 6in b732-fm FA
Contents xxxix
References 495
List of Tables 504
List of Figures 506
List of Original Patents, Applications and Papers 508
Index 513
Chapter 1
Introduction
Additional facts are that why not any of those who have achieved high
fame and made huge fortunes unparalleled in the history of business ever
in the whole world, have spoken up, published and corrected the mistakes
in the articles in popular and technical literature and patent claims of the
IC inventions? A few are donating their fortune to give back to the society
1
February 20, 2009 9:39 B732 9in x 6in b732-ch01 FA
graciously and generously, the others are not. The readers are privy to read
about both types of such individuals in the popular literature. Many of
them were intimately involved from the very beginning in various aspects of
the greatest invention of mankind even before the inventions actually took
place. On ethical and moral grounds as well as professionally, if anybody
who had the plenipotentiary strength to uphold the truth without the
fear of jeopardizing their careers and jobs, would have been these select
group of people. Why did they remain silent all along? Too busy amassing
fame and fortune and soaring ever higher even though the fundamental
facts of the IC inventions were on cloudy grounds? Maybe or maybe not;
debatable? Sounds mysterious? Yes, it sure does. Is it all fiction? No, the
mystery of the invention of ICs is based on well documented facts given
in this book which nobody can deny. They are all available as public
records.
I shall summarize some of the facts and key information of the invention
of ICs in this Introduction Chapter over and beyond those capsulized
already in the Preface. Their details have been given in the subsequent
chapters. They are also repeated in several portions of the book for the
convenience of the readers. This presentation style has been followed not
only for the sake of continuity of the discussions in those places, but also
to help the readers to avoid going back and forth while reading the book, a
popular style also followed by many biography authors of the titans of the
20th century scientists and engineers.
At the outset, I wish to let all the readers know that I have had the
highest regard for both late Jack Kilby and late Bob Noyce. I had known
Bob Noyce both professionally and personally for about 30 years before he
died from a heart attack in 1990. I had also known Jack Kilby professionally
for several years, although we had met only a few times in person before
he died in 2005. In my opinion, as it is also generally known, both men
did make important contributions to the invention of ICs and to the field
of microelectronics, and they were very caring and decent human beings
too. Therefore any critique that I shall present in this book about their
work, is only to uphold the truth of the science and engineering facts, with
no disrespect whatsoever to them. My main aim and the only purpose in
writing this book is to get the history right based on science and engineering
facts, by putting all the available facts on record. I am doing this for the
February 20, 2009 9:39 B732 9in x 6in b732-ch01 FA
Chapter 1. Introduction 3
Section 1. What is an IC?
1. What is an IC?
The fast evolution of the ICs since their invention in 1959 to the current
Ultra Large Scale ICs (ULSICs) is described in Section 5 of Chapter 3.
The ICs sold for the first time in 1960 had only a few devices per chip
fabricated with minimum geometries measured in mils (1 mil = 1/1000 inch,
or one thousandth of an inch), and they were interconnected by a single
level of metalization. Now a typical ULSIC has a few billion devices per
chip fabricated with minimum geometries measured in nanometers (nm),
and interconnected by 8 or more levels of metalizations. To give you an
February 20, 2009 9:39 B732 9in x 6in b732-ch01 FA
Before getting mired into the details of the inventions of Kilby and
Noyce, it will be helpful for the readers to know at the outset of this
Introduction chapter, that the only kind of ICs sold from the beginning
have been the monolithic-ICs fabricated and manufactured in large volume
on single-crystal silicon (Si), except the hybrid microwave IC’s containing
several Silicon monolithic ICs and Compound semiconductor IC’s which are
excluded in this book to focus on monolithic Si ICs. The precise definition of
the word “monolithic” will be given in Chapter 4. Although it is not easy to
coalesce all the information in one-sentence, the definition of monolithic-IC
can be given as:
Chapter 1. Introduction 5
Section 2. Key Requirements for making the IC
Table 1.1. Key requirements for making the monolithic-IC and whether or not they
were met by Kilby and Noyce in their respective inventions.
The sole credit for the invention of ICs has been given to Kilby1
and Noyce2 in the literature since their inventions were disclosed in
their respective patent filings in the USPTO ∼50 years ago in 1959.
Unfortunately this recognition accorded to Kilby and Noyce is only partly
correct and justified, and most if not all of the authors and hence the
readers on this subject have failed to appreciate this fact. Kilby and Noyce
did play key roles in the invention of ICs, but not quite the way they have
been portrayed in the literature so far. They have been singled out as if
they did it all by themselves in inventing the ICs. They have been given an
iconic stature by the hero worshippers without knowing where the credits
are actually due and for what?
Also, hundreds of million dollars and some in the billions have been
made as profits in their pockets by some of the contributors as well as
non-contributors to the IC technology and related businesses, and by those
business savvy wizards who happened to be at the right place, at the right
time and with the right people. As a professional courtesy and propriety,
February 20, 2009 9:39 B732 9in x 6in b732-ch01 FA
Chapter 1. Introduction 7
Section 3. Sole credit to Kilby and Noyce
I shall not mention the names of several of them whom I knew personally.
This is also to avoid the appearance of a conflict of purpose by implication.
My sole purpose is to help straighten out the true IC invention history.
My presentations in this book are based on my own exhaustive research of
the published literature and my own educational science and engineering
backgrounds and job experiences (see Appendix 1).
Only a very select few genuine contributors numbering less than the
number of fingers on one hand, who have amassed huge fortunes, are giving
back to the industry and the society. They are the very admirable few
February 20, 2009 9:39 B732 9in x 6in b732-ch01 FA
indeed who were lucky, smart and courageous too. Nevertheless, while they
and the others have had dame luck smile on them, many other genuine
contributors have been ignored by the society; not only by the laymen
but also by their technological peers of the latter days including today.
Moreover, some of the fundamental facts of the IC invention have either
been overlooked or not understood at all by the previous authors on the
history of IC. These facts have also been documented in this book. While
not taking away the key contributions of Kilby and Noyce, these facts tell
a different story about their actual contributions. The details can be found
in the subsequent chapters of the book, but a few are summarized below to
point out the mystery of the invention of ICs.
4.1. One fact is that Kilby did make a key contribution to the invention
of ICs by disclosing that all the devices can be fabricated in a single piece of
semiconductor. (See Jack Kilby, Fig. 1 in Wolff3 : “A page from Jack Kilby’s
notebook of July 24, 1958, where he first recorded how resistors, capacitors,
and transistors could be made on a single slice of silicon.” As it is clear from
this Fig. 1, Kilby’s handwritten disclosure was not witnessed, as required by
the Patent Law.) But this is only a small part of the complete information
needed to make the monolithic-IC. Did he borrow this idea from and limited
to the same technical constraints as Geoffrey Dummer4 in England, who
had described the same key yet limited concepts earlier? This remains as yet
an unresolved mystery. However the fact also is that Dummer’s publication
(see Chapter 9) pre-dates Kilby’s disclosure handwritten in his notebook.
It is also well known that Kilby had attended Dummer’s presentation of his
paper in 1952 in Washington, DC. Essentially similar suggestions had also
been made earlier than Kilby by Harwick Johnson5 of RCA and Richard
Stewart6 also of Texas Instruments (as Kilby was also from TI but had
joined TI later than Stewart).
4.2. Another important fact is that Kilby did not invent the silicon
monolithic-ICs, the only kind sold from the beginning to this day (except
the hybrid microwave ICs such as those in the cell phones). This fact has
been confirmed by a member of the Nobel Committee (Dr. MNC-1) in
his written communications with me recently (see Chapter 12). In the
February 20, 2009 9:39 B732 9in x 6in b732-ch01 FA
Chapter 1. Introduction 9
Section 4. A few facts in the mystery of the invention of ICs
specifications of his invention, Kilby did not give the correct procedures
for fabricating the devices and their electrical isolation. Moreover he
missed completely giving the correct procedures to interconnect the devices
(rectifying diodes, amplifying and switching transistors, resistors, and
capacitors) in the chip, without which the IC is not complete and cannot
function to process the electrical signals and information. These facts tell
us that this is not a mystery; these were serious omissions by Kilby and
not envisioned by Kilby at that time. But it is a mystery how all this
important information was ignored and swept aside by the USPTO in
granting the patents to Kilby, and by the key technical and patent personnel
all over the world who simply accepted Kilby as the inventor of the IC.
Moreover the authors of books on the invention of the transistor and of the
IC, and also some who wrote their PhD theses in History on the science
and technology of these inventions also ignored or did not understand this
important information.
4.3 The quintessence of Nobel’s Will is that the Nobel Prizes shall be
awarded to those irrespective of nationality who, during the preceding year
have conferred the greatest benefit on mankind. Nobel Prize is not awarded
in the field of Engineering. The invention of ICs belongs more to the field
of engineering rather than any other basic field of science such as Physics,
Chemistry, etc, in which the Nobel Prizes are awarded. Nevertheless the
impact of the ICs on almost all the other basic sciences has been so huge in
the last few decades that their invention definitely merited a Nobel Award.
The ICs are indispensable to do almost any kind of standard and advanced
work in all the fields included in the original Nobel Award list.
The citation for the above Nobel award to the co-winners Alfred,
Kroemer and Kilby was published as “for basic work on information and
February 20, 2009 9:39 B732 9in x 6in b732-ch01 FA
Chapter 1. Introduction 11
Section 4. A few facts in the mystery of the invention of ICs
Chapter 7). Figure 1.1 shows the photograph of Kilby’s reduction to practice
of his invention. Wire bonded interconnects can be seen clearly going from
one device to another flopping in air above each of the device chips.
Noyce did not acknowledge Hoerni8 and Lehovec9 in his key IC patent.2
Why? The reasons can be debated, nevertheless this is a fact whose mystery
remains unexplained. The facts also tell us that the reduction to practice
of his invention was a team effort who had used Noyce’s step-and-repeat
lithography camera. The contributions were made by several key co-workers
who had worked in the 100-person-strong Fairchild Semiconductor R&D
Laboratory of which he was the Director of Research. Whether or not
their indispensable contributions were ignored by Noyce’s patent lawyer
intentionally, cannot be ascertained, but it was more likely that Noyce
did not know who actually made the contributions and could not single
out one person except those other single-inventor patents also filed during
February 20, 2009 9:39 B732 9in x 6in b732-ch01 FA
Chapter 1. Introduction 13
Section 6. Getting history right
5. My qualifications
are used worldwide in many if not all fields and applications (partial
list is given alphabetically): banking, biotechnology, communications,
computers, education, entertainment, government, hospitals, internet,
medicine, nanotechnology, research, travel, and others, and in every
commercial, defense and industrial businesses. All the electronic systems
in these applications use much more advanced ICs such as Ultra Large
Scale ICs (ULSICs) than the ICs invented originally. However, the stems of
all ULSICs are rooted in the basic invention of the ICs. The ULSIC business
has now grown to multi-hundred billion dollars annually which is also the
heart and soul of all electronic systems market of trillions of dollars per
year. It appears that the future growth of ULSICs and its impact on newer
businesses shall still be forever increasing. Therefore it is important to know
what were these basic inventions of the ICs, who invented them and how.
The case of Einstein will be cited here, not to set history right because
there was nothing wrong with his work, but as an example of defending
under difficult circumstances what was right for physics and humanity.
Max von Laue had defended Einstein’s work which was being criticized
as being Jewish in character by the others during the period Hitler’s power
was ascending in Germany. Even though Laue belonged to the majority
group and Einstein to the minority, it was not easy for Laue to stand up
to the other well known physicists in the majority group who were trying
to discredit and criticize Einstein unfairly. It took a few years for Laue
to set the records straight about Einstein’s discoveries, but he succeeded
eventually. Just imagine the compounding of the problem had the situation
been reversed. Had Laue made the errors in his work and Einstein had
February 20, 2009 9:39 B732 9in x 6in b732-ch01 FA
Chapter 1. Introduction 15
Section 8. Why getting history right is important?
corrected and tried to document them, this would have made Einstein a
pariah being a member of the minority group. Even though Einstein’s main
motive to do this would have been only to set history right, he would not
have been given any kudos for this.
Being cognizant of the yore, while forging ahead with the future by
innovations is a prudent wisdom to follow, so that the past mistakes shall
not be repeated. The adage “Those who do not learn from their past
mistakes, are condemned to relive them” has been proven to be right over
and again throughout the entire history of mankind including scientific
research. The inspiration for the younger generation to innovate for the
future is provided by the leaders of the past. If the past accomplishments
are misrepresented by the spin meisters of the present, it negates this key
process of progeny of innovation for the future. Therefore getting history
right is very important.
The invention of ICs has been one of the most important inventions
of the 20th century which has revolutionized mankind forever. In some
respects, the impact of this invention on mankind far exceeds that of
Einstein’s discoveries made earlier. Nevertheless, the fundamental aspects
of IC are rooted in Einstein’s contributions among other pioneering
discoveries. To re-emphasize, ICs are used worldwide in many applications
in various fields, and almost nothing is possible today without using the
ICs and their associated technologies. As best as one can extrapolate into
February 20, 2009 9:39 B732 9in x 6in b732-ch01 FA
9. Documented facts
The documented facts are the only way to get history right especially
in science and engineering. I concede, however, that even the documented
facts in these fields of exact sciences and engineering may have resulted
from intrigues and usurping other people’s rights and contributions. Such
discussions are also debatable and beyond the scope of this book, although
brief comments will be made only on a few relevant issues directly
concerning the invention of ICs. Therefore the main focus of this book
is to give and discuss the documented facts of IC invention. Even the
documentations have left several key questions still unanswered after so
many years regarding this important invention. They will be discussed
briefly at the end in Chapter 15. Comments shall also be made on the
support technologies key to the invention and progression of ICs, which
have not been given their due recognition in the literature so far.
Chapter 1. Introduction 17
Section 10. What are documented facts?
latter has limits, the former is limitless!” Even though these sentiments
have been expressed by many over millennia, Einstein is credited with them
as he has been better known in the modern era. A key implication derived
from these comments of Einstein is to get history right. So this statement
has been made here to reinforce the importance of getting history right,
especially regarding the IC inventions by objective and thorough analyses
of the documented facts.
Documented facts which are used in the scientific and technical world
to assign an invention are the patents issued by a bona fide legal agency
of a government such as the United States Patent and Trademark Office
(USPTO) in the USA or similar agencies in other countries of the world, and
the research publications, pre-publication and now also post-publication or
post-public-disclosure, in peer reviewed professional journals. The latter
have less clout than the former in legal contests. However for professional
recognition as well as awards of prizes like Nobel Prize, research publications
wield an equivalent clout if not higher than the issued patents. While this
practice holds true most of the time, however, it was not the case in the
recognition given to Kilby “for his part in the invention of the integrated
circuit” cited by the Nobel award committee7 in 2000 (for details, see
Chapter 12).
invention to practice. All the claims of the patent must be supported clearly
by its specifications.
11. All ICs sold from day one have been Si monolithic-ICs
All the ICs manufactured and sold from the very beginning when they
had only a few transistors on a silicon chip to those currently having
several billion transistors per chip, have been the monolithic-ICs using
silicon (Si). (As stated earlier, the bybrid microwave ICs, widely sold, such
as in cell phone, are excluded from these deliberations.) As recently as
in 2006, several research historians16,41 have erroneously heralded Kilby’s
invention in 1958 as the advent of the monolithic era. Noyce’s invention
as documented in his patent was simply to prescribe how a monolithic-IC
was to be made using Si, but he did not reduce his concepts to practice.
It was done by the others who were members of Noyce’s Research and
Development Laboratory of which Noyce was the director at that time.
Gordon Moore became its director later. The planar technology mandatory
for fabrication and the electrical isolation of devices necessary in the ICs
were respectively invented by Hoerni,8 a member of Noyce’s Laboratory,
and Lehovec9 who was with Sprague Electric Company. However, even
the invention of the planar technology ascribed to Hoerni can also be
questioned. It was derived by Hoerni from a combination of associated
technologies on which the key technical work was done earlier by several
others (see Chapter 6 for details). What is the monolithic concept and
how do the monolithic-ICs differ from the hybrid-ICs, has also not been
understood or delineated properly in the literature so far. This will be
explained in Chapter 4. Details of the facts of Kilby’s and Noyce’s inventions
shall be given with proper documentations in Chapters 7 and 8 respectively.
Chapter 1. Introduction 19
Section 13. Victory/success and defeat/failure
12.1 Why do we need to know the important facts and to get the
clarifications now after the invention of ICs about 50 years ago (or 55 years
depending on when you start counting)?
12.2 Why nobody else has clarified some of the key technical facts of
their inventions by Kilby and Noyce so far? Did Kilby and Noyce borrow
their ideas from the others?
12.3 What are the questions on the invention of ICs which are still
unresolved, and why have they not been addressed so far?
The answers to the above three questions are given in Section 18 below
after discussing the relevant information in the following sections.
Despite the above well known adage regarding victory and defeat, it
is interesting to note that not many “fathers” have come forward to claim
the parentage of the invention of the ICs during the past 50 years. The
two “fathers”, viz., Jack Kilby1 and Bob Noyce,2 have been regarded as
the undisputed inventors of the ICs, except that a few key facts even about
their inventions have not been scrutinized carefully so far in the literature.
Several others also did make key contributions to make the monolithic-ICs
a reality, which are the true ICs, not the hybrid-ICs, being manufactured all
along in the past 50 years. These monolithic-ICs are the heart and soul of
the multi-hundred-billion dollar per year IC industry, which is also the core
of the multi-trillion dollar per year electronic systems industry. However,
while a few have been recognized (cf. Rostky15 ) for their contributions to
make monolithic-ICs a reality, others have been literally “walked over”.
Was it due to the hero worship of Kilby and Noyce, and/or due to the
lightening speed at which the technology and business developments with
huge profits that took place, and that the timing was propitious for them
to happen? These are debatable issues. The purpose of this book is not
to address this complex subject which is hard to quantify, but to present
several technical facts primarily about Kilby’s and Noyce’s inventions of
ICs and key contributions of a few others which have not been clarified and
documented before. Therefore, it is important to know the truth about who
invented what and how, even after 50 years since the patenting process of
these inventions was begun.
A few authors such as Kilby1,19 (who is widely regarded as, and was,
the co-inventor with Noyce2 of the ICs), Wolff,3 Rostky,15 Riordan &
Hoddeson,16 Berlin,17 Reid,39 Lee,40 Brock,41 and Lojek42 have tried to
February 20, 2009 9:39 B732 9in x 6in b732-ch01 FA
Chapter 1. Introduction 21
Section 13. Victory/success and defeat/failure
tell the story of the invention of ICs in their respective ways. While Kilby21
himself has given a historical account of the invention of the ICs in 1976,
however he addressed and discussed the technical aspects of his invention
and the patent22 only recently19 in 1998, and made some comments also
on Noyce’s invention and his basic IC patent.2 Noyce43 described the IC as
conceived at Fairchild, and referred to the work of Kilby (ibid), Hoerni,8
Lehovec9 and others, but did not describe the technical details of their
patents in his paper. The authors in references 15–17, 39, 41, 42 do not
address the technical issues of Kilby’s and Noyce’s IC inventions and their
patents, and they have ascribed Kilby’s invention incorrectly to be that of
a monolithic-IC. Perhaps this may be due to their efforts more as science
history writers, rather than as scientists and engineers who are educated
and trained to have the technical precision and knowledge and contributed
first hand to solid state devices and IC technologies. Even Kilby’s later
comments19,44 are incomplete engineering physics at best.
Chapter 1. Introduction 23
Section 14. Disclosure of the basic concept for ICs
writing this book is to clarify and present the facts. Like celebrities, only
those inventions are newsworthy and deserving of critical and thorough
investigation, which make a major impact on the business and/or human
life. “The Invention of ICs” is such an invention (see Saxena20 ). Therefore,
its story merits a more thorough technical scrutiny than it has been done
so far in the literature.
For the sake of whetting the appetite of the reader at this point, I shall
quote the key ideas of the IC invention described in their respective patents
and paper by Kilby, Dummer, Johnson and Stewart briefly as follows:
Kilby did not state the above concept in any of his other patents.
“At this stage, I would like to take a peep into the future. With the
advent of the transistor and the work in semiconductors generally,
it seems now possible to envisage electronic equipment in a solid
block with no connecting wires. The block may consist of layers
of insulating conducting, rectifying and amplifying materials, the
electrical functions being connected directly by cutting out areas of
the various layers.”
Chapter 1. Introduction 25
Section 15. Brief comments on the basic concepts of IC invention
“The invention improves over the prior art circuits in that the
necessary circuit elements such as the load resistor may
be made an integral part of the semiconductor element.
One such semiconductor element according to the present invention
replaces several transistors required in the circuits of the prior
art performing the same function. T he interconnecting circuit
wiring is thereby eliminated or reduced. Also the gates of
the “nor” circuit of the present invention provide amplification in
addition to performing a logical function and the gates are well
matched since they are basically one transistor.
The fact that the entire circuit is embodied in a single
transistor element allows miniaturization of the circuit heretofore
not realizable.”
For the sake of continuity and the readers to get a sense of intrigue in
the invention of ICs, brief comments on their respective versions of concepts
quoted above, are given as follows. Detailed comments on the basic concepts
of the IC invention as documented by Kilby, Dummer, Johnson and Stewart
above in Section 14 are given in Chapter 9.
15.1 Kilby had filed his patent(s) in 1959; some were issued in 1964, and
others later. However, Kilby’s key patent no. 3,138,744, the only patent in
which he had stated his basic concept of IC invention, was filed on May 6,
1959, and was issued on Jun. 23, 1964*. Kilby had claimed priority of earlier
filing on Feb. 6, 1959, of his original application (OA) but this was denied
by the USPTO. Its filing date was recorded as May 6, 1959, or that no filing
date was recorded at all. Nevertheless based on OA, patent no. 3,138,743
was also issued on the same date as 3,138,744, viz., on Jun. 23, 1964*.
Strangely enough, Kilby kept on referring to his patent no. 3,138,743 in his
later patents and papers, although he had not stated his basic IC concept
in this patent. Both of these patents were assigned to Texas Instruments
(TI). The asterisk on the issue dates for these two patents of
Kilby (3,138,743* and 3,138,744*), as well as Stewart’s patent no.
February 20, 2009 9:39 B732 9in x 6in b732-ch01 FA
15.2 Dummer did not file for a patent. But he had first published his
concept on May 6, 1952, almost as an afterthought for an IC, without
defining what kind of IC could be fabricated: hybrid or monolithic?
Kilby’s description in his patent no. 3,138,744 filed on May 6, 1959 was
similar and limited to the same criteria as Dummer’s in 1952. Kilby had
attended Dummer’s presentation in 1952, so it is likely that Kilby may
have derived his ideas from Dummer. He seems to allude to it in his Nobel
Lecture in 2000. However, this cannot be proven beyond any reasonable
doubt. Nevertheless, despite their incompleteness and similarities, clearly
Dummer’s documentation of the basic concept pre-dates Kilby’s.
15.4 Richard Stewart’s patent no. 3,138,747 was filed on Feb. 12, 1959,
and it was issued on June 23, 1964*. The asterisk on the issue date of
this patent and the two patents of Kilby (3,138,743* and 3,138,744*) given
February 20, 2009 9:39 B732 9in x 6in b732-ch01 FA
Chapter 1. Introduction 27
Section 17. Inputs for Noyce’s patent and investigation of Kilby’s patents
above, is to draw the reader’s attention that they were all issued by the
USPTO on the same date, and assigned to Texas Instruments (TI). Stewart
had been working at TI before Kilby had joined it in 1958. Why was Stewart
ignored when he had given essentially similar concepts to Kilby’s, and all
the kudos were being given to Kilby, is a mystery. Stewart’s disclosure of
the basic concept in his patent filed ahead of Kilby’s patent no. 3,138,744
in which Kilby’s concept was stated, was quite similar. Stewart wrote,
“. . . the necessary circuit elements such as the load resistor may be made
an integral part of the semiconductor element. One such semiconductor
element according to the present invention replaces several transistors
required in the circuits of the prior art performing the same function. The
interconnecting circuit wiring is thereby eliminated or reduced.” Filing of
Stewart’s patent no. 3,138,747 predates the filing of Kilby’s patent no.
3,138,744. For detailed discussion, see Chapter 9.
facts of Kilby’s and Noyce’s inventions which had not been published in the
past.
The second question, “Why nobody else has clarified some of the key
technical facts of the inventions of ICs by Kilby and Noyce so far?” is not
easy to answer. Nevertheless I have given in Chapters 7 and 8, these key
technical facts and why they are important. Hundreds and thousands of
qualified personnel (scientists, engineers and patent attorneys) all over the
world must have noticed these technical facts. Why they did not address
and document them so far is surprising indeed.
The answers to the last question, “What are the unresolved questions
still on the invention of ICs, and why have they not been addressed so
far?” are also somewhat difficult. However, these questions and comments
are given in Chapter 15.
February 20, 2009 9:39 B732 9in x 6in b732-ch01 FA
Chapter 1. Introduction 29
Section 20. For whom is the book primarily intended and at what level?
Many of the engineers and scientists, even those who are currently in the
fields of microelectronics, computer and information and communication
technologies (ICTs), have taken Ultra Large Scale Integrated Circuits
(ULSICs) for granted. While some may know how these ULSICs came to be,
and how they emerged from the original ICs to the super chips of today,
many are probably too busily immersed in the narrow topic and task of
developing their respective advanced technologies to meet the target dates
demanded by their jobs. As it is well known, these billion-device super
chips are used in many applications in various fields of education, research,
medicine, government, and in the entire commercial, industrial and defense
industries. Therefore, even though the number of engineers, scientists and
other workers in these fields may be huge, those who really understand
the overall facts regarding the invention of the original ICs may be rather
small. In order to appreciate the mind boggling progress of the entire IC
industry, it will be helpful to understand how did it all begin, and to know
the facts about how the original ICs were invented and by whom? Knowing
the true history of the invention of ICs and their development can only help
in having a better insight into the future technologies needed for the even
more advanced chips.
This book may also be used as the key text book to establish
graduate level courses on “The Invention of ICs which Revolutionized
February 20, 2009 9:39 B732 9in x 6in b732-ch01 FA
Chapter 2
Books have been, and many more can be, written for the answers to the
above questions. This is because they involve the inventions and inventors
in many disciplines, scientists, policy makers in different governments, and
the spin-meisters in the legal/patent fields all over the world. The answers
arrived at to define an invention, for example, can also be debated forever
in some cases. However, for the purpose of this book, we shall focus only
on the invention of the ICs. All the ICs manufactured and sold from the
very beginning when they had only a few transistors per chip to those
currently having over a billion transistors per chip have been monolithic-
ICs made with Si. But what is the monolithic concept which is mandatory
for fabricating such ICs, and how do these ICs differ from the hybrid-ICs?
This is explained in Chapter 4.
31
January 20, 2009 9:57 B732 9in x 6in b732-ch02 FA
1. Discovery
2. Invention
Other serious matters which also should be noted are the confused
record keeping by the USPTO regarding the filing and issuing of Kilby’s
patents, and possible compromise of the laws of US Patent code 35 USC
112 and associated protocols. For details, see Reference 12 and Chapter 7.
2.3. We can go to the moon and return by using rockets and space
vehicles. [The exact proven methods and designs are not given.]
2.4. ICs can also be made from semiconductors other than Si by using
appropriate surface passivation, planar, and interconnect technologies. [The
exact proven details are not given.]
2.6. The energy crisis can be solved by using alternate energy sources
instead of gasoline. [The exact proven details of the choice of alternate
energy source(s) and their methods and techniques are not given.]
January 20, 2009 9:57 B732 9in x 6in b732-ch02 FA
Noyce’s invention and his patent2 could also be questioned about their
uniqueness deserving the sole credit. As discussed above, Noyce’s invention
was based only on the disclosure in his notebook which was not even
witnessed. It had used Hoerni’s8 and Lehovec’s9 inventions, without which
Noyce’s invention would not have been feasible. Its reduction to practice
was not done by Noyce; it was done by others several of whom claim that
had it not been for their key contributions, Noyce’s invention would not
have worked.
The above conclusions are not meant to tarnish the iconic images of
Kilby and Noyce, but the documented facts have been given here to get
the history right. I have had, and still do have, the highest respect for
both Kilby and Noyce. The leadership qualities of Noyce are well known.
Noyce went on after co-founding Fairchild to co-found Intel Corporation
with Gordon Moore which is the top most IC manufacturer in the world
today. At Intel, the most advanced IC technologies are being developed
and used in large volume manufacturing. Kilby also made other important
contributions to TI and the microelectronics industry. But the documented
facts about their respective inventions of ICs tell a different story than it
has been generally accepted so far for the past five decades regarding the
credits accorded to them.
3. Improvement
4. Patents
— making,
— using, or
— selling the invention.
To get a patent, the inventor must provide written description, and the
method of fabricating and using the invention in
— full,
— clear,
— concise &
— exact terms.
After getting the patent, the inventor can license it to the others to use
the invention for fees which are acceptable mutually. (Such fees can amount
to millions of dollars.)
in his invention several key technologies correctly which are mandatory for
the IC to be completed and function properly: monolithic interconnects
adherent to the surface of the insulating film to connect all the devices;
planar technology for fabricating the devices; monolithic isolation between
the devices.
During the time of Kilby’s invention, each solid state device had very
poor yield and it was definitely not obvious to one in the state of the art
that the overall yield for multiple devices on a single chip could be brought
up to an acceptable level. Jack Morton, Head of Bell Labs at that time
didn’t put any of his team’s effort into ICs because of his confidence that
the overall yield of an IC would be disastrous. No one would have expected
that the inclusion of multiple resistors or capacitors could have had a poor
yield, so the additional inclusion of a multiplicity of resistors and capacitors
may not have been an issue for the Patent Department judges at the time.
5. Publications
of ICs are the patents and the papers by Jack Kilby (see Chapter 7) which
got issued and published despite their questionable and/or limited merit.
6. Trade secrets
Still another way to protect one’s proprietary idea is to use the “Trade
Secret” route. A famous example for this is the Coca Cola formula, which
has never been patented. However in short, this route is not easy to
protect one’s key technologies for the ICs and ULSICs, although several
corporations have invoked “Proprietary Information” as the reason for not
divulging them in conferences and publications.
January 20, 2009 9:57 B732 9in x 6in b732-ch02 FA
Chapter 3
41
January 20, 2009 9:57 B732 9in x 6in b732-ch03 FA
Fig. 3.1. A typical cross section of twin-tub CMOS structure (not drawn to scale).
January 20, 2009 9:57 B732 9in x 6in b732-ch03 FA
When solid state active devices such as transistors and diodes replaced
the vacuum tubes, and were used in conjunction with the passive
components such as resistors and capacitors in a chassis or on a PCB, the
resulting electronic circuit was a sort of integrated circuit. As explained in
the next Chapter 4, such circuits are called hybrid-ICs. However, to further
clarify their evolution and differentiate between those hybrid-ICs fabricated
initially with “Packaged Chips” and later with “Unpackaged Chips”, I shall
introduce a new terminology as follows.
None of the IC products initially sold had all the devices monolithic
within one Si chip. They were individual devices which were wire-bonded
between the various active and passive devices to connect them to form the
desired circuit. Such products were not monolithic-ICs; they were hybrid-
ICs. In hybrid-ICs, the entire structure of the circuit on the chip is not
monolithic, because the bonded wires go from one device to another, up
above the Si chip, and/or to the pre-metallized interconnects on the ceramic
or other substrates.
Many of the older readers of this book will remember the days
of assembling radios, amplifiers and other circuits with transistors,
diodes, etc., using “Heathkits” and other electronic kits sold by several
manufacturers in the 1950s and 1960s. Already packaged solid state devices
with PCBs or other substrates having some of the interconnections were
supplied with the kits, along with the circuit diagrams and instructions
on how to assemble them. Such kits represented circuits which were
miniature hybrid-PC-ICs. So we shall not use this example for the onset of
miniaturized integrated circuits, otherwise we will have an onslaught from
many a “father” claiming to have invented the ICs after having bought the
Heathkits for themselves and/or their kids.
The solar cells also are not characterized as miniature ICs, because
their p–n junctions and interconnections are huge in size as compared to
those fabricated in transistors or even hybrid-ICs. However, since the early
workers in this field did use Si for the solar cells inter-connected with Al
adherent to the oxide films, even though their technology relatively speaking
was crude, some of them may feel entitled to be credited with the invention
of the integrated circuit (e.g., see Queisser24 ). Therefore, any person who
has constructed such solar cells earlier could also claim recognition as an
inventor of the IC. But this has not been practiced in the literature so far,
or given recognition by the Nobel Committee either, so we shall not deviate
from this practice as well.
Not only the state of the art ULSIC chips, but from the very beginning
of manufacturing and selling the ICs about 48 years ago, all the IC products
sold have used monolithic chips of single crystal Si, although the monolithic
Si and compound-semiconductor chips may be connected by wires to form
a larger hybrid IC product. Such monolithic Si chips are fabricated in
single crystal Si substrates, and the planar technology is mandatory for
the fabrication of the devices in them. Each chip is one solid block of Si
single crystal substrate in which the various devices are fabricated and
isolated monolithically, and they are connected with single and multilevel
interconnects adherent to SiO2 films and contiguous to the Si substrate.
Within a package containing a die-attached (or chip-attached) monolithic-
IC chip, a few wires which are not monolithic are only those bonded from
the pads on the chip to the external leads of the package, to communicate
with the outside circuits external to the chip. These wires do not dangle
loose and flop around above and in between the devices of the chip in the
package, but the two respective ends of each wire are securely bonded.
Thus all the IC products manufactured and sold from day one of the IC
industry contained Si chips which were entirely monolithic. The practice of
referring to these products has been to call them simply in short the ICs,
and not to elaborate them as the monolithic-ICs.
Figure 3.2 shows the roadmap of the evolution of ICs.20,27 Figure 3.3
is the same with additional comments on the role of scaling (viz. shrinking
January 20, 2009 9:57 B732 9in x 6in b732-ch03 FA
Fig. 3.3. Roadmap of ICs.20,27 Same as Fig. 3.2, but it also gives additional comments
on the role of scaling, multilevel interconnections, and the device integration regimes.
The largest segment of the present ULSIC business of multi-hundred dollar volume is
Si CMOS.
Faggin’s classic editorial comments61 and his paper co-authored with several
colleagues as follows:
Since the total world semiconductor market in 2007 was about $250
billion, the market directly related to the microprocessor was $175 billion
approximately. The total systems market using them was over $1 trillion
annually. (See Chapters 13 and 15.)
A very important fact the reader must also realize is that all these
microprocessors and other advanced chips are all 2D-Si-ULSICs. To explain
it briefly, such 2D (2-Dimensional) chips have all the devices laid out
in the 2-dimensional silicon surface layer of the Si wafer, and the 3rd-
dimension is used primarily for the multilevel interconnects above the
2-D silicon surface layer. Some thin-film devices, such as resistors and
capacitors, may be fabricated in multiple levels in the 3rd dimension above
the main level, the 2-D silicon surface layer level, of the single crystal Si
wafer. But these thin-film devices are not fabricated from single crystal
Si films. However, almost all the ULSIC materials and device technologies
are reaching their engineering if not also fundamental limits that prevent
their further performance improvements as well as higher functional packing
density for further microminiaturization. To understand these limitations,
one has to dig deep into various aspects of reaching these limits. Only a brief
list of the rather broad topics is given below without giving their details.
5.1. Materials.
5.2. Equipment.
5.3. Processing.
5.4. Diagnostics/Monitoring.
January 20, 2009 9:57 B732 9in x 6in b732-ch03 FA
5.5. Test.
5.6. Reliability.
5.7. Design.
5.8. Applications/Marketing.
The debate has gone on regarding how small dimensions of the devices
and interconnects can be achieved and manufactured economically and
profitably in large volume. There are various aspects of this debate which
are also beyond the scope of this book. The current status of the practical
limit of the small dimensions is expected to reach at about 0.018 µm (or
18 nm). This regime is referred to as the nano-scale CMOS technology,
because the dimensions have been shrunk from one-tenth of micron (0.1 µm
or 100 nanometer) to the tens of nanometer (nm) range. The latter range
is also referred to as the nano-technology encompassing a wide variety of
disciplines including nano-tubes made of molecular-size carbon fibers.
January 20, 2009 9:57 B732 9in x 6in b732-ch03 FA
Since CMOS has such a strong history behind its development and
in large volume manufacturing, the natural question arises what can be
done to milk its attributes further before its limit is reached. Many
excellent papers have been written to address this. The challenges and
issues of the nanoscale Si CMOS have been reviewed and analyzed. Various
technologies and device structures discussed include different types of Si-
on-insulator (SOI), double-gate-field-effect-transistor (DG-FET), stacked
FET, strained-Si FET, SiGe, high-k gate-dielectrics, low-k inter-layer-
dielectrics, metal gate electrode, and carbon nanotube FET (CNFET).
Also new device geometries at even smaller dimensions to provide higher
function density such as the recent discovery (March 2007) by Sah and
Jie63 on the simultaneous presence of electron and hole surface channels
and currents in the experimental data of the nanometer double-gate thin-
base MOSFETs, would immediately double the function density of the
CMOS ICs without change in the dimensions used in the current nanometer
technology. Already, SiGe, high-k gate-dielectrics and metal gates are
being implemented in production. Although the proprietary details are not
known, a high-k gate-dielectric material of choice is based on Hafnium Oxide
(HfO2 ). Its dielectric constant has been reported recently as about 20. It
is interesting to note that it is quite close to the value reported by me and
Mittal57 about 33 years ago. We had calculated it from the accurate film
thickness measurements by ellipsometry. Earlier, the dielectric constant of
HfO2 was claimed to be about 3 times larger by a few other investigators,
who had made errors in the film thickness measurements of HfO2 .
The bottom line is that economics will dictate the continued use of
conventional Si devices and related materials, until they become obsolete
to deliver cost effective performance. However, paradigm shifts in the
technologies are needed to extend the validity of Moore’s Law and introduce
new products.48
All the processes developed for the above 3D-Si-ICs and UPICs should
be applicable for large volume manufacturing. See recent patents.37,38
2. With 3D-Si-ICs, the design rules can be relaxed. This means that the
minimum geometries and the alignment tolerances used in the conventional
<65 nm 2D-Si-IC technology, can be relaxed to wider CDs such as >65 nm
when the new 3D-Si-IC technology is used. Relatively speaking, the former
requires more stringent manufacturing technologies than the latter. It is
important to note that the criterion of the number of devices per chip, or the
device density, is not compromised by going to the more relaxed design rules
with 3D-Si-ICs. The device density achieved with a conventional <65 nm
2D-Si-IC technology, can also be obtained if not increased with the new
>65 nm 3D-Si-IC technology. Moreover, the chip size with the latter having
wider CDs can be smaller than with the former having smaller CDs. This
is due to the layout advantages of the new >65 nm 3D-Si-IC technology.
Such layout advantages are even more striking when an apple-to-apple
January 20, 2009 9:57 B732 9in x 6in b732-ch03 FA
comparison is made for the wider CD (e.g., 65–90 nm) technologies. The
chips with the 3D-Si-IC technology can be about 50% smaller, and have less
metal and improved performance than those with the conventional 2D-Si-IC
technology. The reliability of the 3D-Si-ICs having wider CDs will also be
better than that of conventional 2D-Si-ICs with smaller CDs. Thus, several
aspects of manufacturing technologies, such as photolithography, reactive
ion etching (RIE) and multilevel metallizations should be easier for the
3D-Si-ICs than those needed for the conventional 2D-Si-IC technologies for
chips having the same device densities.
Figure 3.4 shows the cross section of an UPIC chip using both Si and
compound semiconductors.38 It shows an example of four layers of devices
fabricated on single crystal films of Si and compound semiconductors,
and connected to each other with multilevel interconnects. Only the
ohmic contact technologies in UPICs will require different metals for Si
and compound semiconductors. However, the multilevel interconnects are
essentially the same as in the 3D-Si-ICs. The thermal budget requirements
in both UPICs and 3D-Si-ICs have to be considered carefully because of
the growth of the single crystal Si and compound semiconductor films, and
metallizations thereafter.
Fig. 3.4. Cross section of an UPIC chip38 using both Si and compound semiconductors.
It shows four layers of devices fabricated on single crystal films of Si and compound
semiconductors, and connected to each other with multilevel interconnects.
3. Existing 6 (150 mm) and 8 (200 mm) fabs can be used for
manufacturing the 3D-Si-ICs and UPICs, with essentially the same
equipment, except for a few changes in the epitaxial growth, ion
implantation and metallization equipment. The process integration will
January 20, 2009 9:57 B732 9in x 6in b732-ch03 FA
need to be adjusted for the new process flow. Other key processes such as
photolithography, RIE, multilevel metallizations and planarizations shall
remain essentially the same.
4. The present trend to increase the wafer size from 8 (200 mm) to
12 (300 mm) and beyond to 16 (450 mm), thereby escalating the wafer
fab cost from about 5 billion dollars to colossal larger amounts, and to
require equipment migration to ever larger sizes can be stemmed by the
3D-Si-IC and UPIC technologies. In fact these new technologies offer a
significant advantage to extend the life of the older 6 (150 mm), current
8 (200 mm) and 12 (300 mm) fabs. The main reason for this advantage
is that they allow unique custom chips with high profit margins, which
cannot be produced with the current 2D-Si-IC manufacturing technologies
and fabs.
6. The new 3D-Si-IC and UPIC technologies will also have a significant
impact on the large area flat panel displays, TVs and other unique
applications. The former has been already in the large volume production
which has driven the price of the LCD TV down by about factor of two
annually. However this technology does not employ the single crystal Si
films, which are described in the recent patents.37,38
9. Future
This implies large volume manufacturing, meeting and exceeding the needs
and lowering the cost of existing applications, creating new applications,
acceptance by the political and social changes, marketing and distribution.
Despite some promising new technologies, there is plenty of room still for
advancing the current well established CMOS technologies. Nevertheless,
paradigm shifts in technologies need to be explored in the future. As an
example, a technology will be very welcome to have a silicon light emitter
with sufficient optical efficiency at low power dissipation, compatibility with
existing processes, and economic manufacturing. A few advanced research
labs have claimed that they are close to achieving this, but it has not yet
been proven in manufacturing. Not only existing product designs can be
implemented with these new technologies to produce improved and more
profitable chips, but newer chips with unique capabilities heretofore not
possible to serve the markets can be manufactured. Thus, the future of the
microelectronics industry is quite promising indeed.
January 20, 2009 9:57 B732 9in x 6in b732-ch04 FA
Chapter 4
Riordan and Hoddeson16 gave historical account of the era from the
birth of the transistor to the beginning of integrated circuits. Their last
chapter is on “The Monolithic Idea”, as they gave their concluding remarks
in their book on the advent of integrated circuits. However, they ascribed
erroneously the accomplishment of Kilby’s reduction to practice as “The
monolithic idea was finally a reality.” It was not a reality in totality, but only
a partial reality, and that too in only a small way. That is because Kilby’s
reduction to practice was a hybrid-IC, not a monolithic-IC (see Chapter 7
for details). Briefly, Kilby had used mesa devices fabricated monolithically
in Ge, but they were interconnected by wire-bonding; none of these are
used to fabricate monolithic-ICs.
59
January 20, 2009 9:57 B732 9in x 6in b732-ch04 FA
Neither of the above two types of hybrid-ICs necessarily has all the
active and passive devices fabricated within or on the same piece of the
semiconductor. Moreover, these devices and the wire-bonded interconnects
in such hybrid-ICs are not contiguous to a single surface, be it may of the
substrate or the semiconductor.
between the devices and the semiconductor pieces were not monolithic to
the entire circuit. Thus such a circuit was not one whole block of a solid unit.
Therefore, it was a hybrid-IC, not a monolithic-IC. Precisely speaking, this
is exactly how Kilby had reduced his invention to practice using Ge mesa
devices and wire bonded gold (Au) interconnects to connect the devices,
viz., Kilby’s reduction to practice was a hybrid-UC-IC.
Kilby’s invention of the ICs, for which he was awarded the Nobel Prize
in Physics in 2000, was a hybrid-IC, perhaps more hybrid than all the
subsequent hybrids. (See above Section 1, and Chapters 7 and 12.) The
reduction to practice, i.e., demonstration of his invention, was not one
whole block of solid unit, so it was not monolithic-IC (see above Section 1).
To give some details, Kilby’s first demonstration device was a phase-shift
oscillator which used only one mesa transistor fabricated in a piece of Ge.
Kilby’s second demonstration device was a multivibrator which used two
mesa transistors fabricated in a piece of Ge. The resistor(s) and capacitor(s)
were fabricated in another piece of Ge. Each respective Ge piece was glued
to a substrate such as glass slide. The transistors in one piece of Ge were
wire-bonded to the resistors and capacitors on another piece of Ge, which
was also glued to the same substrate glass slide. Thus, Kilby’s invention
more precisely was for a hybrid-UC-IC, not a monolithic-IC. The IBM
silicon Solid-Module for the IBM360 mainframe was perhaps the closest
to Kilby’s invention of hybrid-UC-IC. Instead of using dangling gold wires
a la Kilby, IBM replaced them by evaporated aluminum films, patterned
by etching into narrow aluminum metal lines, and interconnected several
single-device chips of bipolar silicon transistors, silicon p/n junction diodes,
and resistors.
2.1. The active devices are fabricated (e.g., using mesa technology
used by Kilby, but not restricted to it) singly or collectively in a piece
of semiconductor (e.g., Ge used by Kilby, but not restricted to it).
2.2. The passive devices are fabricated (e.g., using mesa technology
used by Kilby, but not restricted to it) singly or collectively in the same or
another piece of semiconductor (e.g., Ge used by Kilby, but not restricted
to it).
2.5. The entire circuit constructed as above (by Kilby) was a hybrid-
UC-IC, not a monolithic-IC (explained in detail in sections 1 and 2 above,
and below in Section 3).
In addition to the above facts, the readers should keep in mind that
huge numbers of IBM360 mainframes were produced in which the Solid
Modules were used. It is well known that IBM360s dominated all the
computer applications at that time. The concept and the use of Solid
January 20, 2009 9:57 B732 9in x 6in b732-ch04 FA
Modules was just one step beyond Kilby’s hybrid-UC-IC “obvious to those
familiar with the art”.
In monolithic-ICs, all the active and passive devices are formed and
fabricated in and on the surface of a single piece (chip) of a single
crystal semiconductor, e.g., Si, wafer (substrate). But correct technologies
for forming and fabricating the devices must be used. Mesa technology
specified and used by Kilby for his invention is wrong. It will not work for
monolithic-ICs; the planar technology must be used as was done in Noyce’s
invention. Also, fabrication alone of the active and passive devices in the
same chip in one block (monolith idea) is not enough, because they must
be interconnected contiguous and adherent to the insulating layer over the
same body of the semiconductor to produce a monolithic-IC. If the devices
are fabricated within the same body of the semiconductor, but they are
interconnected by bonding wires dangling over the chip, such an IC is not
a monolithic-IC anymore; it is then a hybrid-UC-IC. This is explained in
detail in the previous Section 1. To re-emphasize, only after all the steps
of fabrication and isolation of devices within the same semiconductor body
(chip) with the correct technology, and their interconnections contiguous
and adherent to the insulating layer over the semiconductor surface are
completed, only then a monolithic-IC is produced.
In practice, each wafer has a large number of chips laid out in arrays.
In monolithic-ICs, each fabrication step is done on the wafer as a whole,
i.e., simultaneously on every chip on the wafer. The only exception to
this is in photolithography with the advent of steppers a few decades ago,
when each die is aligned and exposed individually on the wafer. However,
the associated processes of applying and developing the photoresist are
done simultaneously at each level over the entire wafer. Each respective
process step of depositions and/or growth of the various films/layers are
done contiguous to the entire surface of the wafers, and the respective
January 20, 2009 9:57 B732 9in x 6in b732-ch04 FA
Noyce did not reduce to practice his invention, which was written
but un-witnessed in his lab notebook. However, he had specified Si planar
technology, Al interconnects adherent to and going over SiO2 layers without
shorting p–n junctions, photolithography and etching techniques which are
all used in manufacturing the ICs. Turning Noyce’s invention into reality
was done by several other people.
Even though Noyce received a few patents after his basic IC patent,
and had published a few papers, he made no direct contributions to the
advanced or the next generation ICs himself. However, Noyce made major
contributions to the entire field when he co-founded Intel Corporation with
Moore where many of the most advanced USICs have been developed, and
is now the leading IC manufacturer in the world.
Chapter 5
1. Summary of IC fabrication
How do the monolithic-ICs differ from the other types of ICs such
as hybrid-IC, has been explained in Section 3 of Chapter 4. The key
steps for its fabrication and whether or not they were met by Kilby and
Noyce in their respective inventions have also been given in Table 1.1
in Chapter 1. But I shall re-state them with a few additional comments
69
February 18, 2009 14:32 B732 9in x 6in b732-ch05 FA
as follows:
1.2. Planar technology must be used to fabricate the above devices. This
technology protects the edges of the junctions from being shorted by the
contact and interconnect metallizations. (Mesa technology used by Kilby
exposes the junction edges to metallizations causing shorting and leakages.)
Hoerni8 has been credited in the literature with the invention of planar
technology. However, as it has been documented in the next Chapter 6,
Hoerni did not invent the planar technology all by himself; it was built
upon a combination of important discoveries of a few other people earlier,
the crucial work done by Sah,10 and key inputs from the test engineer(s)
doing the electrical characterizations of the transistors.
1.3. All the devices must be isolated from one another by an appropriate
planar technology (e.g., p–n junction, LOCOS, trench). The most important
technique to isolate the devices initially was invented by Lehovec9 by
using the p–n junctions. It was soon replaced almost completely by the
LOCOS technique invented by Kooi25 and the trench isolation invented by
several technologists later. (See next Chapter 6). Nevertheless, it must be
remembered that Lehovec’s invention was crucial for the invention of ICs.
Therefore keeping the above four steps in mind as the readers proceed in
this chapter and subsequently, it will be easier to understand the important
facts of the IC inventions of Kilby and Noyce. They are given in Table 5.1
below. Since the IC inventions of Kilby and Noyce are the main focus of this
book, the reader will benefit by knowing these important facts also ahead of
time before reading the detailed analyses of Kilby’s and Noyce’s patents in
Chapters 7 and 8 respectively. It is easy to get bogged down in the details
as evidenced by the lack of actions on the part of most if not all the senior
February 18, 2009 14:32 B732 9in x 6in b732-ch05 FA
Next in this chapter, I shall give the updated Table 5.2 to summarize
the relevant patent filings and their issue dates, and earlier unpatented
but documented work, of Kilby, Noyce and several other key contributors.
All the references of the documents have been given, so that any reader(s)
can follow up for details of the information given by me. For the sake
of comparison with a similar table published by me in the earlier NSTI
paper,12 I shall also list them in Table 5.3. The NSTI paper was limited
in scope due to the length limitation of the conference proceedings. The
listings are chronological according to patent filing and publication dates.
(Continued )
February 18, 2009 14:32 B732 9in x 6in b732-ch05 FA
1. Figure 6.1 shows Figs. 1–4 copied from Kilby’s1 patent no.
3,138,744. Note the mesa structures in Kilby’s Figs. 3 & 4 (instead
of the planar structures which are necessary in monolithic-ICs).
Mesa technology was invented at Bell Telephone Laboratories
and first used in mass-produced silicon transistors by Fairchild
Semiconductor which actually used also Hoerni’s planar process
to define the emitter of the first commercial silicon n/p/n and
p/n/p double diffused silicon transistors.
2. Figure 6.2 shows Figs. 1-5 copied from Kilby’s23 patent
application no. 03/791,602. Note the mesa structures in Kilby’s
Figs. 4 & 5 (instead of the planar structures which are necessary
in monolithic-ICs).
3. Figure 6.3 shows Figs. 6 & 8 copied from Kilby’s23 patent
application no. 03/791,602. Note the wire-bonding in Kilby’s
Figs. 6 & 8 (instead of the monolithic interconnects which are
necessary in monolithic-ICs).
STRONG: 1. Noyce’s concept for ICs was for monolithic-ICs, which are
the only type manufactured and sold in the IC market from the
February 18, 2009 14:32 B732 9in x 6in b732-ch05 FA
1. Figure 6.4 shows Figs. 3, 4 & 5 copied from Noyce’s2 patent no.
2,981,877. Note the planar structures and monolithic interconnects
in Noyce’s Figs. 3 and 4 (instead of the mesa structures and
wire-bonding). Figure 5 just shows the circuit diagram of Noyce’s
invention.
February 18, 2009 14:32 B732 9in x 6in b732-ch05 FA
Table 5.2. Filing and issue dates of patents and publications relevant to IC inventions
(Updated version for this book).
Filing date/
Patent no. Publication date Issue date
(Continued )
February 18, 2009 14:32 B732 9in x 6in b732-ch05 FA
Filing date/
Patent no. Publication date Issue date
Filing date/
Patent no. Publication date Issue date
Note: The asterisk * in Table 5.2 is to draw attention to the same date on which
Kilby’s patents 3,138,743 and 3,138,744, and Stewart’s patent 3,138,747 were awarded,
all assigned to Texas Instruments. Note also that Kilby’s patent numbers differ by
only 1, and Stewart’s patent differs from Kilby’s by 4 and 3 respectively.
in this table for the sake of continuity and sequencing, although their
details and other documents relevant to the IC invention are given in
Appendix 2. The patents and earlier unpatented but documented work are
listed chronologically with patent filing dates and public disclosures from
publications. To emphasize, listing patent filing dates and public disclosures
chronologically is important because it documents the origin and sequence
of conception of an invention, which is not reflected by the issue dates of the
patents. The process in between the filing and the issue dates of the patent,
as well as what each inventor did beyond his respective invention to advance
its technology to what it is today, are also important to acknowledge and
when critiquing each contribution. For Kilby and Noyce, they are discussed
in detail in Chapters 7 and 8 respectively, whereas for the others, they are
given in Chapter 9.
Table 5.3. Filing and issue dates of patents and publications relevant to IC inventions
(Copied from the earlier NSTI paper12 for the sake of comparison with the above
Table 5.2. The references in the following table correspond to those in the NSTI
paper.12 ).
The key invention of this patent was to form the well-defined patterns
of interconnects and contacts selectively without doing any etching of the
metal films. The main purpose of listing this patent here is to give example
February 18, 2009 14:32 B732 9in x 6in b732-ch05 FA
of the continuity of my work on interconnects and ICs from the early years
to the present, a credible feature for valid ownership of patents.
Please note that this patent of mine was filed on March 10, 1971, which
was after Kooi’s25 patents on Local Oxidation of Silicon (LOCOS) which
were filed on Oct. 3, 1966, and Jun. 4, 1970. But it was granted on Aug.
29, 1972, which was before both patents of Kooi were granted (July 20,
1976; August 14, 1973, respectively). Prior to Kooi’s patents25 on LOCOS
process, which is used for the isolation of devices in ICs, Lehovec9 had
been awarded the patent for the p–n junction isolation of devices in ICs.
Lehovec’s patent was important and crucial to isolate the devices in the
invention of monolithic-IC by Noyce.2 Kooi’s patent, while not crucial for
the invention of ICs, was important for IC manufacturing for many years
until it was also replaced by trench isolations in advanced ULSICs.
the application #03/791,602 does not have an official filing date.” This was
sent by USPTO to Saxena on November 02, 2005. (See Fig. 5.6)
It is also important to note (see Table 5.2) that according to the public
records, no further action was taken either by Kilby or by the USPTO
on Kilby’s Application No. 03/791,602, and no patent was ever issued
for this application. Even though Kilby’s patent no. 3,138,743 was based
on his Original Application No. 03/791,602, which was never patented,
their claims were quite different. No patent per se was issued directly on
03/791,602, as confirmed by USPTO during communications over telephone
(with “Chris” of Electronic Business Support, USPTO; April 11, 2007).
awarded to Kilby and cited “for his part in the invention of the integrated
circuit” in physics in 2000 is rife with controversy regarding what was his
part in inventing which kind of IC. Kilby was also awarded double the
financial amount than to his co-winners Allferov and Kroemer. My direct
written communications with two members of the Nobel committee (Dr.
MNC-1 and Dr. MNC-2) did not result in clarifying either the official
citation of the award to Kilby or the unequal financial awards to him and
the other co-recipients. However, after I did my own research into the matter
using the Nobel website for the documented facts, I can explain the unequal
financial awards but not the citation of the award. This unusual fact remains
a mystery. For details, see Section 6.8 of Preface, and Chapter 12.
The following figures are given from Kilby’s and Noyce’s patents, and
correspondence from USPTO, with brief comments on them.
1. Figure 5.1: Figs. 1–4 copied from Kilby’s1 patent no. 3,138,744. Note the
mesa structures in Kilby’s Figs. 3 & 4 (instead of the planar structures
which are necessary in monolithic-ICs).
2. Figure 5.2 shows Figs. 1–5 copied from Kilby’s23 patent application no.
03/791,602. Note the mesa structures in Kilby’s Figs. 4 & 5 (instead of
the planar structures which are necessary in monolithic-ICs).
3. Figure 5.3: Figs. 6 & 8 copied from Kilby’s23 patent application no.
03/791,602. Note the wire-bonding in Kilby’s Figs. 6 & 8 (instead of
the monolithic interconnects which are necessary in monolithic-ICs).
4. Figure 5.4: Figs. 3, 4 & 5 copied from Noyce’s2 patent no. 2,981,877.
Note the planar structures and monolithic interconnects in Noyce’s
Figs. 3 and 4 (instead of the mesa structures and wire-bonding). Fig. 5
just shows the circuit diagram of Noyce’s invention.
5. Figure 5.5 Response from E. Bornett,29 Certifying Officer, USPTO,
sent to Saxena on Kilby’s Application No. 03/791,602 on September
26, 2005, regarding its filing date.
6. Figure 5.6 Response from Customer Service Department,30 USPTO,
sent to Saxena on Kilby’s Application No. 03/791,602 on November 02,
2005, regarding its filing date.
February 18, 2009 14:32 B732 9in x 6in b732-ch05 FA
Fig. 5.1. Figures 1–4 copied from Kilby’s1 patent no. 3,138,744. Note the mesa
structures in Kilby’s Figs. 3 & 4 (instead of the planar structures which are necessary in
monolithic-ICs).
February 18, 2009 14:32 B732 9in x 6in b732-ch05 FA
Fig. 5.2. Figures 1–5 copied from Kilby’s23 patent application no. 03/791,602. Note
the mesa structures in Kilby’s Figs. 4 & 5 (instead of the planar structures which are
necessary in monolithic-ICs).
February 18, 2009 14:32 B732 9in x 6in b732-ch05 FA
Fig. 5.3. Figures 6 & 8 copied from Kilby’s23 patent application no. 03/791,602. Note
the wire-bonding in Kilby’s Figs. 6 & 8 (instead of the monolithic interconnects which
are necessary in monolithic-ICs).
February 18, 2009 14:32 B732 9in x 6in b732-ch05 FA
Fig. 5.4. Figures 3, 4 & 5 copied from Noyce’s2 patent no. 2,981,877. Note the planar
structures and monolithic interconnects in Noyce’s Figs. 3 and 4 (instead of the mesa
structures and wire-bonding). Fig. 5 shows the circuit diagram of Noyce’s invention.
February 18, 2009 14:32 B732 9in x 6in b732-ch05 FA
Fig. 5.5. Response from E. Bornett,29 Certifying Officer, USPTO, sent to Saxena on
Kilby’s Application No. 03/791,602 on September 26, 2005, regarding its filing date to
be May 6, 1959 (NOT Feb. 06, 1959) cliamed by Kilby.1,19,23
February 18, 2009 14:32 B732 9in x 6in b732-ch05 FA
Fig. 5.6. Response from Customer Service Department,30 USPTO, sent to Saxena on
Kilby’s Application No. 03/791,602 on November 02, 2005, stating that it “does not have
an official filing date.”
February 18, 2009 14:32 B732 9in x 6in b732-ch05 FA
A. Quotations from the original patents (text and claims) and papers are
given in ITALICIZED FONT.
B. Important points in the above quotations are highlighted in BOLD
ITALICIZED FONT.
C. My comments on any of the above are given in NORMAL FONT.
D. Important points in the above comments given by me are highlighted
in BOLD NORMAL FONT.
The above methodology makes it easier for the reader to recognize when
certain material is being quoted from the original document, and what
portion of the quote is actually important. The same comment applies to
my comments on the original documents.
9.2. Numbering
The numbering of the tables and figures in each chapter follows the
convention defined below.
For example, Table 1.1 is the 1st table in Chapter 1. Table 6.2 is the 2nd
table in Chapter 6.
Similarly, Fig. 1.2 is the 2nd figure in Chapter 1. Fig. –6.3 is the 3rd figure
in Chapter 6.
February 18, 2009 14:32 B732 9in x 6in b732-ch05 FA
Chapter 6
95
January 20, 2009 9:58 B732 9in x 6in b732-ch06 FA
on the Si surface. The inventions of both Hoerni and Lehovec were crucial
to make Noyce’s invention of the monolithic-IC to function properly.
Many scientists all over the USA and the rest of the world were working
with Germanium (Ge), Silicon (Si) and other semiconductors to make
solid state devices. The original credit for choosing Si over Ge for making
the devices belongs to William Shockley.55 The reasons for this choice
were primarily technical, not technological. The insight and the genius of
Shockley as well as some of his preferences and personality traits have
been recognized already in the literature. They are not the subject of this
book. Nevertheless, without shortchanging the former qualities, it is fair to
say that Shockley’s acumen did not follow through a rigorous technology
development of the potential of high quality silicon-dioxide (SiO2 ) films that
could be grown on Si at high temperatures. These temperatures were in the
1000 degree centigrade (C◦ ) range, such as those used for the diffusion of
n- and p-type dopants in Si. The SiO2 films protect and passivate the p/n
junctions (needed to fabricate diodes and transistors) electrically and from
the surrounding ambient on the Si surface. This important quality of the
SiO2 films to preserve the electrical characteristics of the p–n junctions, and
prevent these electrical characteristics from deteriorating by the gaseous
ambient environment was discovered by Atalla at Bell Laboratories.64,65
As it will become clear from the following discussions, Atalla’s important
finding would have led to the invention of the planar technology by Hoerni
while working at the Shockley Transistor Corporation instead of at a new
company in 1959, the Fairchild Semiconductor Corporation. As it is well
known, Hoerni and his seven colleagues (including Noyce and the others —
see Preface, Section 10) had worked together at Shockley, left and co-
founded Fairchild.
The choice of silicon (Si) over the other semiconductors was fortuitous
and timely because the oxide film (SiO2 ) that could be grown on it was
an excellent insulator film. Appreciation of this important property47,55
was due to a combination of knowledge, serendipity, luck, timing, ingenious
contributions of various process and test engineers, and quick turn around
January 20, 2009 9:58 B732 9in x 6in b732-ch06 FA
Atalla and his group at Bell Laboratories were the first to show in
1959 that high quality insulator films of SiO2 could be grown thermally
on the Si surface to protect the underlying Si p/n junction diodes and
transistors.56,64,65 Frosh and Derrick11 were the first to show in 1958 that
the thermally grown SiO2 films on Si can mask against the diffusion of
the vapor of the p-type acceptor impurities (Boron, Gallium, Indium)
and n-type donor impurities (Phosphorus, Arsenic, Antimony) into Si.
These results were extended quantitatively for practical engineering use in
processing by Sah10 in 1958 first at Shockley Transistor Corporation. Later
they were further enhanced by Sah and other colleagues at Fairchild.56 They
modeled the processes mathematically, and designed the experiments and
fabrication processes necessary for the selective diffusions for p/n junction
diodes and transistors passivated by the SiO2 films on the surface of Si. In
1959, Jean Hoerni8 combined these discoveries of the unique properties of
the thermally grown Si-dioxide (SiO2 ) film on Si. They had been reported
and published earlier by Atalla,64 Frosch and Derrick,11 engineering design
data of Sah,10 and Sah with his co-workers.56 Hoerni also seized on the key
inputs from the electrical test engineering group at Fairchild (which have
never been described before in the literature; see discussions given below).
With the knowledge of all this combined important information, Hoerni8
filed for patents on what is now known as the famous planar technology
(see also Table 5.2 in Chapter 5).
for the brain-wave in Hoerni’s mind regarding the invention of the planar
technology. I have used some inputs related to the invention of ICs from
a few reliable sources elsewhere in this book, for which I cannot provide
published documents. But I have quoted and referred to them as personal
communications. I have also stated up front that I have used my professional
discretion and courtesy to refrain from divulging the names in a few cases.
I was privy to and participated in several events that I have discussed in
this book. In the same spirit, I shall refer to the following inputs66 on
the additional sequence of events which were important in the evolution
of planar technology at Fairchild. Therefore I feel compelled to describe
them below in this book. They have never been reported before in the
literature.
3.1. First was the 1954 work of Fuller and Ditzenberger68 on the
diffusion of boron and phosphorus impurities into silicon.
3.2. Second was the 1957 work of Frosh and Derrick11 on the surface
protection and selective masking during diffusion of dopants into silicon.
January 20, 2009 9:58 B732 9in x 6in b732-ch06 FA
3.3. Third was the 1958 work of Sah10 at the Shockley Semiconductor
Laboratory and Shockley Transistor Corporation, (in the six months during
summer 1958 to spring 1959) with the help of Doug Tremere in the second
half of the nearly 100 experiments for diffusion masking by the oxide. Sah
had developed severe allergy to the P2 O5 vapors, and reported it in the full
report of the Shockley Semiconductor Laboratory Technical Memorandum
Series, No. 61, dated 2 March 1959. All the results were presented by Sah
to the Fairchild engineers and Hoerni during Sah’s job interview arranged
by Noyce at Fairchild during the following week in March 1959. This was
also the exact manuscript submitted on June 17, 1959 and published10 in
September 1959 after Sah had left Shockley and joined Noyce at Fairchild.
The manuscript and the final journal article (since there were no revisions
and changes) were given to Hoerni during Sah’s job interview at Fairchild.
This paper10 provided the experiment-based theoretical design equations
and the complete family of engineering design curves for the necessary oxide
film thicknesses needed to mask against phosphorus impurity diffusion into
the Si underneath which oxide masking was not possible as stated in the
original paper by Frosh and Derrick.11
the planar n+/p/n double diffused and oxide protected silicon bipolar
transistors. The diligence of Sah’s work on engineering based technology
of oxide masking against dopant diffusion was recorded by a series of
experiments published56 in the Journal of the Applied Physics in 1964
and 1965 by Sah and his co-workers at Fairchild Semiconductor Research
and Development Laboratory. These experiments were designed to give the
oxide film thickness curves against diffusion of the other donor and acceptor
impurities also into Si.
3.4. Fourth was the 1959 work of Atalla, Tanenbaum and Scheibner64
on the stabilization of silicon surface by thermally grown oxide.
3.5. Finally, the combination of the above four items reported in the
1959–1960 work of Hoerni8,67 was first disclosed to the public in 1960
at the December 1960 IEDM.67 In this presentation, Hoerni coined the
combination (or borrowed from Noyce’s expression) as the planar process
pursuant to benefiting from the work of ALL the prior scientists and
engineers.
3.7. The p–n junctions and their peripheries were sealed at the edges at
the surface by the SiO2 films, thus they were protected from the variations
in the ambient above the films. This enabled the p–n junctions to be very
stable with low leakage currents.
3.8. Because of the point No. 3.7 above, the metal contacts to the p–n
junctions made in their centers did not short the junction edges.
Hoerni got the patents8 for the planar technology essentially covering
the key points 3.6, 3.7 and 3.8 given above. Hoerni’s patents were single-
inventor patents commanded by the Fairchild Patent Attorney’s concern
on the validity and/or the strength of multi-inventors in a patent. It is
unlikely that the criterion of single-inventor was enforced by the senior
attorney, John Ralls (who had died after filing for the IC patent of Bob
Noyce2 ). However it may have been insisted upon later, as an example, for
the patent on the CMOS inverter circuit by Wanlass. Point no. 3.9 was the
necessity that was built-in in Hoerni’s innovative combination of 3.1, 3.2,
3.3 and 3.4. As discussed above, they were obvious to those familiar with
all the work done previously. However Hoerni was at the right place and
time to combine all this prior knowledge and invent the planar technology.
It worked successfully and produced the original Si planar transistor, and
provided Noyce to take the next obvious step, i.e., to put two and two
together and invent the monolithic-IC. The continued use of the planar
technology from the inception to the ULSICs of today attests to the
fact that it is indispensable to the entire microelectronics field. However,
despite all the technical and technological factors, it must also be recognized
that the Nature’s gift to mankind of high quality SiO2 films that can be
grown on Si with a reliable and stable interface reproducibly, was and is
absolutely crucial.
The credit due to both Hoerni and Lehovec for their fundamental
contributions to the invention of the IC had been essentially ignored, or
certainly not recognized in the entire literature so far, until my paper12
was published. Nevertheless, their innovations did not fall like manna from
the sky, but they were based on the four indispensible prior arts and the
knowledge gained by the tedious and long experiments listed in 3.1, 3.2,
January 20, 2009 9:58 B732 9in x 6in b732-ch06 FA
3.3 and 3.4. The historical fact cannot be denied that Sah was the first
to design and perform the crucial experiments which enabled Hoerni to
accomplish “fait accompli”, and they were described by Sah in 1988.10
The credit at least due Hoerni was acknowledged recently by Riordan51
after my paper12 had stated it explicitly for the contributions from both
Hoerni and Lehovec. But nobody has recognized the importance of the
engineering-designed prior arts described in 3.1 to 3.4, and the key input
from an electrical test engineer, that were the seeds sowed in Hoerni’s
mind to combine them to blossom into the ubiquitous planar process of
the IC industry. As the readers can now appreciate that the Si planar
process was built upon the key contributions of many, but it is referred
to as Hoerni’s planar process. This may also be due to the fact that the
basic patents for this process were filed and awarded to Hoerni8 (two years
ahead of Kilby’s patent, and one year after Noyce’s patent; see Table 5.2,
Chapter 5).
(b) opening a hole through the coating, thereby exposing a limited surface
area of the semiconductor;
trenches (grooves) in the bulk Si wafer is covered and filled with deposited
oxide. Thus any junctions if delineated in the trenches due to etching
are covered by the oxide layers, thus they are protected from ambient
effects also.
Lehovec specifies the use of p–n junction isolation of devices in ICs. This
is Lehovec’s key invention. Nevertheless, this specification is too general
and not valid under all conditions. It only works to electrically isolate if
the p/n junction is not strongly forward biased, i.e., VPN is less than about
+300 mV, above which it will not isolate effectively. Therefore, in such cases
Lehovec’s patent is not valid, and his claims cannot be allowed. Lehovec
wrote a rejoinder to Kilby’s paper21 of 1976, because according to Lehovec,
Kilby had misquoted his p–n junction claims. Hence Lehovec corrected it in
the letter to the editor published in IEEE Trans. ED26 given in Section 10
at the end of this chapter.
It should also be evident to anyone familiar with the state of the art
why Bob Noyce may have chosen not to specify p/n junction for isolating
the transistors on a piece of silicon. Noyce probably knew the trouble due
to the p/n/p/n switch back, which was not obvious to anyone else at that
time, that for certain conventional circuit designs and operations Lehovec’s
simple p/n junction isolation scheme will not work. This limitation can
be obviated substantially when additional circuit elements are added to
the original circuit. This has been done in the scaling of CMOS ICs,
but it adds complications to the design. Therefore, Lehovec’s invention of
p–n junction isolation is at a disadvantage for widespread use in ICs, in
particular CMOS ICs. Consequently it has been replaced by Kooi’s25 and
trench isolation inventions. It is another example of why just a theoretical
idea of an invention without meaningful reduction to practice and detailed
scrutiny is usually inadequate.
The technical reasons have been given above why Noyce may have
chosen not to file claims for p–n junction isolation, and also not to
acknowledge Lehovec in his patent. Please note the word “may” that I
have used in the previous sentence, which I have done to give the benefit of
January 20, 2009 9:58 B732 9in x 6in b732-ch06 FA
I did not have a chance to review personally the above with Noyce
before he died in 1990. Nevertheless based on my technical reviews with
Sah,69 I have given the most likely technical reasons in the previous
paragraphs above why Noyce may have chosen not to claim the p–n junction
isolation in his IC patent2 because of the circuit design issues and the
pnpn switch back which Noyce was very familiar with while working at
Shockley Laboratory on the pnpn diode switch for telephone switching
office.
Lehovec sent me a personal copy of the first page of the final hearing
on March 16, 1966, of Kilby vs. Lehovec. It is given in Section 11 at the
end of this chapter. See also Reference 8 of Lehovec’s paper.26 It refers to
“Decision of the Board of Interferences in the patent interference Kilby vs.
Lehovec,” No. 93,612, April 5, 1966. Quote from last but one paragraph of
Lehovec’s paper26 is as follows:
“To quote from the decision by the Board of Interferences [8]: ‘We have
carefully examined Patent 3,138,743’ [7] ‘but nowhere can we find support
for the subject matter of the counts . . .. Since Kilby has no reduction to
practice prior to the filing date of Lehovec, . . . priority of counts 1–5 is
awarded to Kurt Lehovec, the senior party.’ The counts of interference are
the claims of the Lehovec patent [2].”
12. Copy of the first page of the final hearing on March 16,
1966, of Kilby vs. Lehovec. See reference 8 of Lehovec’s paper,26
given as “Decision of the Board of Interferences in the Patent
Interference Kilby vs. Lehovec,” no. 93612, April 5, 1966. Quote
from last but one paragraph of Lehovec’s paper is as follows:
To quote from the decision by the Board of Interferences [8]: “We have
carefully examined Patent 3,138,743” [7] “but nowhere can we find support
for the subject matter of the counts . . .. Since Kilby has no reduction to
practice prior to the filing date of Lehovec, . . . priority of counts 1–5 is
January 20, 2009 9:58 B732 9in x 6in b732-ch06 FA
awarded to Kurt Lehovec, the senior party.” The counts of interference are
the claims of the Lehovec patent [2].
Chapter 7
Kilby’s Original Application (OA) and the three patents for his IC
invention that are discussed in this chapter are as follows.
127
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 129
Section 1. Introductory comments on Kilby’s invention
have been lost! I am still pursuing this matter with higher authorities
at USPTO, but I have had no satisfactory response from them so far.
(See Section 1.5).
1.5. Why the facts of the filing date of Kilby’s OA (which was so
important to claim the earliest filing date of the IC invention by Kilby) were
not ascertained correctly earlier? It appears still to be in a state of confusion
at USPTO as evidenced by their written communications29,30 with me
in 2005. As of writing this book, my attempts to obtain the “Certified
Copy each of the File History of patent numbers 3,138,743 and 3,138,744,
both being based on the original application number 03,791,603” from the
USPTO have not been successful. For detailed discussions, see Section 4.2
in Chapter 15. The response from the USPTO is given in Figure 15.1 at
the end of Chapter 15. It states, “We are unable to fill your order because
the files for the above patent numbers are unavailable due to being in the
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
lost category. . . .” I have initiated further inquiry to find out why such
important files have been “lost” by the USPTO. So far I have not received
any reply from the USPTO.
If the files of “Certified Copy each of the File History of patent numbers
3,138,743 and 3,138,744, both being based on the original application
number 03,791,603” ordered and paid for by me were not “lost” by the
USPTO, they would have helped to clarify at least two mysteries:
Given and discussed below are a few important facts and comments to
help the readers in appreciating the details of Kilby’s invention. Some of
these facts and comments may be repeated throughout the book for the
sake of continuity and emphasizing their relevance to the discussion at a
given point.
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 131
Section 2. A few important facts and comments on the invention of ICs
2.3. Kilby wrote his original patent application (OA) and claimed to
have filed it with USPTO on Feb. 6, 1959. However, the USPTO recorded
it as filed on May 6, 1959, as documented in recent communications with
me,29 or that it “does not have an official filing date.”30 These conflicting
responses from the USPTO cannot be explained. A staff member “Chris”
of USPTO told me over telephone that, “If an application does not have
an official filing date, then no action was taken by USPTO on it, and no
patent was awarded for it.” Kilby wrote repeatedly in the text of his issued
patents and in his subsequent papers that his patent was filed on Feb. 6,
1959. Kilby’s most recent paper19 was published in 1998, which was only
two years before he was awarded the Nobel Prize in 2000.
2.5. Kilby received his IC patents1,22 in 1964 after Noyce’s patent2 was
issued in 1961, even though he had filed earlier than Noyce.
2.6. Kilby lost his patent interferences against Noyce (for monolithic
interconnects) and against Lehovec (for p–n junction isolation in ICs).
These rulings by USPTO clearly suggest that Kilby did not qualify for
the Nobel Prize, especially as it was cited by the Nobel Committee for his
award. They also provide evidence that the Nobel committee had not done
a thorough job of evaluating the contributions of Kilby for the Nobel award
(see Chapter 12 and Appendix 7).
was well known that the planar technology and other advanced technologies
were essential to fabricate them.
2.9. Kilby was awarded the Nobel Prize7 in 2000 “for his part in the
invention of the ICs”. But what was his part in the invention of what
kind of ICs was not specified by the Nobel Committee. His invention was
only for hybrid-ICs with Ge mesa transistors and wire-bonded interconnects
which are not manufactured and sold because they are not monolithic-ICs.
Also it is inexplicable why Kilby was given twice the financial award than
to each of the other two co-recipients of the Nobel Prize (for details, see
Chapter 12).
2.10. Kilby’s patent no. 3,138,743, serial no. 791,602 was issued on
June 23, 1964, the same date on which his patent no. 3,138,744 (note the
difference in the patent nos. sequentially by only 1), serial no. 811,486
was issued. Also 3,138,743 was issued by USPTO, and wrote on this
patent as filed on Feb. 6, 1959. This filing date is same as that of original
application (OA) no. 791,602 claimed by Kilby, although this filing date was
not confirmed by USPTO.29,30 Another patent 3,261,081 was also filed on
Feb. 6, 1959 (see Section 1.4). The texts and figures of OA, 3,138,743 and
3,261,081 were exactly the same in all three, but their claims were entirely
different. Huge differences in their claims can be read in the discussions
in Sections 15.2, 16 and 20 in this chapter. The original copies of OA,
3,138,743, 3,138,744 and 3,261,081 are given at the end of this chapter.
They all contain the same sets of figures and texts. The claims of 3,138,743
and 3,261,081 revised from the OA did not support the text and the figures,
so this violates the US patent code of Section 35 USC112. But still the
USPTO allowed these patents. All of these facts are quite unusual and
highly circumspect legal procedures of USPTO. Those readers who are
interested in the details should read and compare the original documents
given at the end of this chapter.
2.11. To inform the reader briefly, Section 35 USC 112 requires that:
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 133
Section 2. A few important facts and comments on the invention of ICs
2.11.3. The written description must prove that the inventor (e.g., Jack
Kilby) was in appreciative possession of the later claimed subject matter
as of his original filing date.
Under Section 35 USC 112, one is allowed to re-write the claims entirely
while keeping the text the same, as long as the specification (text and
drawings) provides full support for the new set of claims and one is not
double patenting (claiming the same thing twice).
3. Original patents
5. Key figures
In patent no. 3,138,744, Figs. 1 and 2 show the top view and the circuit
diagram of the test device, and Figs. 3 and 4 show the mesa structures.
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 135
Section 7. Reduction to practice
6. Key claims
7. Reduction to practice
The last two sentences from this quote do not quite convey the message
accurately. This sentiment is also echoed by Berlin17 (see below), when
she writes about Kilby’s circuit as, “. . . his was undoubtedly an integrated
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Berlin17 also writes: “In the fall of 1958, a young Texas Instruments
researcher named Jack Kilby set out to build an integrated circuit. By early
1959, he had built a complete circuit on a single germanium substrate.
Kilby’s circuit was meticulously hand assembled with a network of gold
wires connecting the components to each other. The wires precluded the
device from being manufacturable in any quantity, a fact of which Kilby
was well aware, but his was undoubtedly an integrated circuit of sorts.”
This also confirms the wire bonding used by Kilby.
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 137
Section 8. Choice of semiconductor
also confirms the wire bonding used by Kilby. Moss had tested the “solid
circuit” flip-flop (7 resistors, 2 capacitors, 2 transistors, “interconnected
with very fine wires”) given to him in 1960 by Kilby. For details, see
Chapter 11.
In 1998, Kilby19 refers only to his patent no. 3,138,743 in his discussions
on the invention of ICs, not to any other patent of his. Kilby writes,19
“To demonstrate the feasibility of the idea, germanium wafers which had
previously been diffused for transistors were selected. These were cut and
masked with wax to form the necessary circuit elements for a simple phase
shift oscillator. It was operated successfully on September 12, 1958. Within
a few weeks a similar process was used to produce a flip-flop circuit.
A patent application covering both germanium and silicon was prepared
and filed in February, 1959.”
8. Choice of semiconductor
all the claims, Kilby writes only “semiconductor material” and does not
specify the semiconductor to be either Ge or Si.
Kilby wrote in his patent1 : Column 3; Lines 23–25 & 28–34: “Initially,
a block of semiconductor material is procured and doped either in its
entirety or over an area in which an active element is to be formed. . . . Thus
impurities may be diffused in successive layers into the surface of the
semiconductor block to form the emitter, base and collector regions. (Para)
After doping has been accomplished, upper layer areas other than those
required for the active circuit elements may be eliminated by etching,
thereby leaving only those desired . . . .”
Kilby wrote in his patent1 : Column 2; Lines 70, 71; Column 3; Lines 1–
4: “Immediately over film 8 is located a dielectric film 7 which may be of any
suitable material, such as silicon monoxide; and immediately above film 7 is
positioned relatively low resistance film 6, which as heretofore mentioned,
comprises the upper conducting film of the capacitor C.” (Underlining of
silicon monoxide done by me here to draw attention to it, and emphasize
that it was not silicon dioxide.)
Kilby did not write how silicon monoxide was deposited or grown. It
cannot be grown easily, because the grown film is normally silicon dioxide.
So, it is implied that it was a deposited film. Vacuum evaporation was one
of the commonly used processes to deposit silicon monoxide films during
those years, and it is used even currently in a few non-IC processes. Also,
silicon monoxide was used by Kilby in his patent1 as the dielectric for the
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 139
Section 12. Impact of 35 USC 112
11. Interconnects
The impact of US Patent Code 35 USC 112 has already been discussed
above in Sections 2.10 and 2.11.
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 141
Section 14. Kilby’s OA No. 791,602
Table 7.1. Important items of Kilby’s Original Application No. 791,602, analyses and
comments on related patents based on it.
4. Another patent issued anyway Patent no. 3,261,081 with “Original Field
with OA 791,602, but with Feb. 6, 1959”; with OA “791,602”, but
“Divided and this application with “Divided and this application Mar.
Mar. 16, 1964, Ser. No. 16, 1964, Ser. No. 352,380.”
352,380.”
5. Texts in OA & the patents. Same in all three: 791,602; 3,138,743;
3,261,081. None of these three state the
monolithic concept even partly as was done
by Dummer in 1952 (although Kilby gives
it only partly in his patent no. 3,138,744).
A few key points described in the texts of
these three patents are: 1. Shaping for
isolation between components; 2. Shaped
“mesas” for transistors, diodes, and
capacitors; 3. Evaporated Si-oxide for
dielectric layer for capacitors; 4. Au-plated
Kovar leads “attached by alloying”; 5. Au
evaporated through a mask to make ohmic
contact with n-regions; 6. Al evaporated
through mask to provide emitter areas;
7. Au wires thermally bonded; 8. Au “laid
down on the insulating material” for
interconnects. None of these are ever used
in Si monolithic-ICs. Kilby also states on
p. 7 of this application, “. . . examples of
suitable semiconductor materials Ge, Si,
intermetallic alloys such as GaAs, AlSb,
InSb, as well as others.” Technologies need
for Ge, GaAs, AlSb, InSb, etc are not the
same as those used in Si monolithic-ICs.
6. All Figs. 1–8. Exactly same in all three: 791,602; 3,138,743;
3,261,081. A few key points from the
figures are: 1. Figures 1–5 show mesa
structures; 2. Figures 6 and 8 show
wire-bonded interconnects. None of these
are ever used in monolithic ICs.
7. Claims. Entirely different in all three: 791,602;
3,138,743; 3,261,081. But they are all for
the same invention. Moreover they are not
supported by the texts and figures. For
possible violation of 35 USC 112, see next
item below.
(Continued)
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 143
Section 14. Kilby’s OA No. 791,602
8. Possible violation and irregular 1. Patent no. 3,138,743: Under 35 USC 112,
use of the patent law? re-writing the entire set of claims in
3,138,743 which are completely different
from 791,602 was not allowable, because
they do not support the specifications of
the text and the figures. But this was
allowed by the USPTO. Why?
2. Patent no. 3,261,081: Even though all the
texts and Figs. In 3,261,081, OA 791,602
and 3,138,743 are the same, and their
claims are entirely different, if this patent
3,261,081 was filed as Ser. No. 352,380 on
Mar. 16, 1964, then the patent law 35 USC
112 would not have been violated in issuing
this patent. However, the revised filing date
Mar. 16, 1964 was not written on the front
pages of 3,261,081; instead “Original filed
Feb. 6, 1959” was written. This appears
rather strange action by
USPTO.
All of the above actions were very
unusual, and difficult to understand now as
we know the laws under 35 USC 111 and 35
USC 112 to be. During 1959–1964, the field
of planar transistors, devices and monolithic
ICs were in its infancy as compared to what
we know now in 2008. Therefore lack of
appropriate expertise in USPTO examiners
may have contributed to the unusual
decisions made by them when confronted by
teams of aggressive patent attorneys trying
to secure the maximum for their clients’
patents.
9. Significance of the same issue 3,138,743 and 3,138,744 (note that they differ
date of 3,138,743 and 3,138,744. in number by only 1, i.e., they were
successive patents) were issued on the same
date June 23, 1964. Note the following
points:
1. USPTO took over 5 years to review both of
these patents since their filing in 1959 and
finally issued them on the same date June
23, 1964.
(Continued)
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 145
Section 14. Kilby’s OA No. 791,602
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 147
Section 14. Kilby’s OA No. 791,602
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 149
Section 14. Kilby’s OA No. 791,602
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 151
Section 14. Kilby’s OA No. 791,602
Claim 22. “Each and every novel aspect of the invention disclosed
hereby.”
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 153
Section 15. Kilby’s Patent No. 3,138,743
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 155
Section 15. Kilby’s Patent No. 3,138,743
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 157
Section 15. Kilby’s Patent No. 3,138,743
being the base and the emitter regions of said junction transistor,the
emitter region being substantially smaller than the base region
on said one major face, a base contact being positioned on said
base region spaced from the emitter region.”
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 159
Section 16. Kilby’s Patent No. 3,138,744
semiconductor material with said first and second conductive means being
separately connected to opposite ends of such elongated region and with said
means for applying operating bias being connected to a centrally located
portion of such elongated region.”
Comments on Claim 23: What are the 1st and 2nd conductive
means which connect the bases of the two transistors, and how
are they fabricated? NOT supported by the figures or the text.
Kilby’s patent no. 3,138,744 is the only patent in which he had stated
the monolithic concept, although it was only partly correct. As discussed
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 161
Section 16. Kilby’s Patent No. 3,138,744
Thus Kilby’s statement alone of his concepts, and that too partially,
of monolithic-ICs is not enough to credit him with their invention. Most
of the materials and technologies written by him in his patent #3,138,744
and original application #03/791,602, are not suitable for the fabrication
of monolithic-ICs. Planar technology and interconnects going over and
adherent to the insulating films are essential to fabricate monolithic-ICs.
Key figures from Kilby’s patent no. 3,138,744 are Figs. 3 & 4. They
show clearly mesa structures of the transistor.
Comments on the claims of the patent which are important for Kilby’s
invention are as follows.
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 163
Section 16. Kilby’s Patent No. 3,138,744
This has already been discussed above in section 7. Further quotes and
comments are:
This has been already discussed above in section 8. Further quotes and
comments are:
Column 2; Lines 70, 71; Column 3; Lines 1–4: “Immediately over film 8
is located a dielectric film 7 which may be of any suitable material, such as
silicon monoxide; and immediately above film 7 is positioned relatively
low resistance film 6, which as heretofore mentioned, comprises the upper
conducting film of the capacitor C.”
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 165
Section 16. Kilby’s Patent No. 3,138,744
16.9. Interconnects
Table 7.2. Summary of comparisons between Kilby’s two IC patent nos. 3,138,743 and
3,138,744, and a few additional comments.
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 167
Section 17. Summary of Comparisons
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 169
Section 18. Detailed analyses of all the Claims of 3,138,744
All the claims of Jack Kilby’s patent no. 3,138,744 (filed May 6, 1959;
issued June 23, 1964) and their detailed analyses are given below.
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 171
Section 18. Detailed analyses of all the Claims of 3,138,744
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 173
Section 18. Detailed analyses of all the Claims of 3,138,744
Comments on Claim 13: What are the conductive means, how are
they deposited and etched to define conductive patterns?
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 175
Section 18. Detailed analyses of all the Claims of 3,138,744
Comments on Claim 14: What are the conductive means, how are
they deposited and etched to define conductive patterns? Re-hash of Claim
13; minor addition of “applying a bias voltage”.
thereto, one end of said elongated layer being electrically connected to one of
said conductive means by a conductive strip overlying said insulating layer
and being contiguous thereto, one end of said conductive strip contacting
said one of said conductive means, the conductive strip extending over said
portion and one of said regions.”
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 177
Section 18. Detailed analyses of all the Claims of 3,138,744
adjacent said one surface, third conductive means contacting said third
region adjacent said one surface, each of said conductive means being
positioned in areas free of said insulating layer, an elongated layer of
resistive material overlaying said insulating layer and being contiguous
thereto, one end of said elongated layer being electrically connected to said
first conductive means, means for applying a bias voltage between said third
conductive means and the other end of said elongated layer, means for
applying an electrical potential to said second conductive means, and output
means connected to said first conductive means and comprising a thin
conductive strip adherent to said insulating layer .”
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 179
Section 19. Kilby’s Patent No. 3,261,081
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 181
Section 19. Kilby’s Patent No. 3,261,081
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 183
Section 19. Kilby’s Patent No. 3,261,081
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 185
Section 19. Kilby’s Patent No. 3,261,081
said method including the step of treating the wafer toprovide electrical
isolation between said active and passive elements.”
said method including the step of processing the wafer to obtain electrical
separation through the wafer between wafer portions which in the
completed device form said active and passive circuit elements.”
(b) diffusing into one area of said wafer an opposite type conductivity-
producing ingredient to form a transistor region selected from the class of
emitter, base and collector,
said method including the step of treating the wafer to provide electrical
isolation between active and passive circuit elements.”
Claim 15. “The method of fabricating a solid state circuit device having
at least one active element and one passive element comprising the steps of:
said method including the step of treating the wafer to provide electrical
separation between the active and passive elements.”
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 187
Section 19. Kilby’s Patent No. 3,261,081
said method including the step of treating the wafer to provide electrical
isolation between the transistor and the resistor and capacitor .”
said method including the step of treating the wafer to obtain electrical
separation between the transistor and the resistor and capacitor .”
said method including the step of treating the wafer to provide electrical
isolation between the diode and the inductor .”
(b) shaping another part of said wafer to form one region of said
inductor,
said method including the step of treating the wafer to provide electrical
isolation between the transistor and the inductor .”
Claim 21. “The method of fabricating a solid state circuit device having
a transistor and a resistor comprising the steps of:
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 189
Section 20. Jack Kilby: Original Application no. 791,602
(c) and applying to the wafer electrically conductive means to the wafer
to connect the transistor and the resistor in a circuit,
said method including the step of processing the wafer to provide electrical
separation through the wafer between said parts of said wafer.”
A copy of the original application filed by Jack Kilby with the USPTO
TO is given below.
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 191
Section 20. Jack Kilby: Original Application no. 791,602
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 193
Section 20. Jack Kilby: Original Application no. 791,602
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 195
Section 20. Jack Kilby: Original Application no. 791,602
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 197
Section 20. Jack Kilby: Original Application no. 791,602
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 199
Section 20. Jack Kilby: Original Application no. 791,602
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 201
Section 20. Jack Kilby: Original Application no. 791,602
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 203
Section 20. Jack Kilby: Original Application no. 791,602
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 205
Section 20. Jack Kilby: Original Application no. 791,602
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 207
Section 20. Jack Kilby: Original Application no. 791,602
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 209
Section 20. Jack Kilby: Original Application no. 791,602
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 211
Section 21. Jack Kilby: US patent no. 3,138,743
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 213
Section 21. Jack Kilby: US patent no. 3,138,743
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 215
Section 21. Jack Kilby: US patent no. 3,138,743
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 217
Section 21. Jack Kilby: US patent no. 3,138,743
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 219
Section 21. Jack Kilby: US patent no. 3,138,743
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 221
Section 22. Jack Kilby: US patent no. 3,138,744
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 223
Section 22. Jack Kilby: US patent no. 3,138,744
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 225
Section 22. Jack Kilby: US patent no. 3,138,744
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 227
Section 23. Jack Kilby: US patent no. 3,261,081
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 229
Section 23. Jack Kilby: US patent no. 3,261,081
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 231
Section 23. Jack Kilby: US patent no. 3,261,081
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 233
Section 23. Jack Kilby: US patent no. 3,261,081
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 7. Kilby’s Invention of IC: Key Patents, Claims and Analyses 235
Section 23. Jack Kilby: US patent no. 3,261,081
January 20, 2009 9:58 B732 9in x 6in b732-ch07 FA
Chapter 8
1. Original patent
237
February 18, 2009 14:19 B732 9in x 6in b732-ch08 FA
state of the art may not need to read the original patent at the outset. The
key points are given succinctly which can be grasped easily. Nevertheless,
the original patent has been given for the sake of record, and for those who
may be interested in the technical and legal details of Noyce’s IC invention.
The original patent also documents how Noyce and his patent attorneys
chose to describe and claim the invention.
Chapter 8. Noyce’s Invention of IC: Key Patent, Claims, and Analyses 239
Section 4. Key claims
which are used in the Ultra Large Scale ICs (ULSICs). The progress of each
has been absolutely mind boggling.
3. Key figures
Comments on the figures of the patent which illustrate the key points
of the invention are given below.
Figure 1 shows the top view, and Fig.2 shows a cross section of a
double diffused transistor. The ohmic contact metalizations to emitter and
base regions going over the oxide layers across the respective junction edges
are also shown. The junctions and their edges are passivated and protected
by the oxide layer. Only the contacts are made in the center away from
the edges of the junctions. Therefore the leakage currents of the junctions
are low and not affected by the ambient. This is the essence of Hoerni’s
invention of planar technology. The junction edges in mesa technology
used by Kilby are exposed to the ambient, therefore the leakage currents
of the junctions are high, unstable, and vary as the ambient conditions
change.
Figures 3, 4 and 5 show the circuit used by Noyce for his invention.
Figure 3 shows the top view, Fig. 4 shows the cross section of the structure,
and Fig. 5 shows the circuit diagram. The key things to note are that the
“dished” planar junctions extend beyond the opening in the oxide whose
edges are all protected by the oxide layer, and the metalizations make
contacts to the junctions away from their edges, and these metalizations
adherent to the oxide layer go over and across the junctions without shorting
them to complete the circuit. The “discoid” shaped metalizations are just
the shapes of the interconnects used at that time by Noyce for his invention.
Figures 6 and 7 show an example in which the emitter and base contacts
are parallel strips.
4. Key claims
described in the patent. However, one or two are usually the key claims
of the invention, and the rest of the claims are written to cover the various
ramifications and subsets of the invention.
Chapter 8. Noyce’s Invention of IC: Key Patent, Claims, and Analyses 241
Section 6. Choice of semiconductor
5. Reduction to practice
The reduction to practice was led by Noyce; it included not only Noyce’s
step-and-repeat lithography camera but also the contributions made by
the others. Several of them claim that had it not been for their key
contributions, Noyce’s invention would not have worked. This may have
been true, but Noyce’s own lithography contributions were crucial too.
Keeping in mind that the various technologies used in 1959 to reduce the
invention to practice were rather primitive by today’s standards, however,
they are summarized as follows:
The oxide layers were grown by thermal oxidation; masking was initially
done by contact printing; buffered HF was used to etch patterns in oxides;
patterns were defined by photolithography; thermal diffusions of boron
and phosphorus through patterns etched in the oxides were accomplished
to dope selectively the desired areas of silicon to give the p- and n-type
junctions respectively, and the rest of the regions were masked from
diffusions by the oxide layers of appropriate thicknesses; metallizations were
done by evaporating aluminum which was adherent to the oxide layers;
metal patterns were delineated in aluminum by contact masking and wet
chemical etching.
6. Choice of semiconductor
Many scientists all over the USA and the rest of the world were working
on Germanium, Silicon and compound semiconductors. However, Noyce
used Si. The choice of Silicon over others was fortuitous because the oxide
February 18, 2009 14:19 B732 9in x 6in b732-ch08 FA
film that could be grown on it was an excellent insulator film. For details
of the various factors resulting in this choice, see Chapter 6.
8. Insulating layers
Chapter 8. Noyce’s Invention of IC: Key Patent, Claims, and Analyses 243
Section 9. Isolation of devices
The above lines clearly state that the insulating layer used in
Noyce’s invention is SiO2 film which is thermally grown on silicon
substrate. For pedantic reasons during 1959 when Noyce’s application was
filed, “disproportionated silicon suboxide” was also included in Noyce’s
description to cover all possible aspects of thermally grown oxide.
Also, deposited silicon oxide films were not mentioned because the
“semiconductor device” being claimed then was a single-level metallization
device, not a multi-level metallization device as produced thereafter.
As it is well known now, after the 1st-level insulator film of SiO2 is
thermally grown, insulator films of doped and undoped Siv Ox Py Bz
films are deposited above thermal oxide and at higher-levels by various
processes such as low pressure chemical vapor deposition (LPCVD),
plasma enhanced chemical vapor deposition (PECVD), sputtering, etc,
to be used as inter-layer dielectrics (ILDs). Planarization of these films
has also been used in most of the advanced technologies to reduce the
interconnect delays and the steps of thinner interconnect metal giving
higher resistance and current density leading to higher failure rate due to
electromigration.56
9. Isolation of devices
10. Interconnects
Chapter 8. Noyce’s Invention of IC: Key Patent, Claims, and Analyses 245
Section 12. Overall comments
A few important facts other than the technical details and comments
on the invention of ICs by Noyce are as follows.
12.1. Noyce became aware of Hoerni’s planar process in 1958, wrote his
concepts for monolithic-ICs in his notebook at Fairchild in 1959 which was
not even witnessed by anyone. However, he must have disclosed his concepts
at least to some of his seven co-founders at Fairchild Semiconductor and
his key managers during that time. Proof of such disclosure by Noyce
has not been available, although several of the seven co-founders are still
professionally active at this moment. Thus, existence of such disclosures by
Noyce during 1958–1959 is debatable at best, until one or some of the key
contributors would step forward.
12.2. Noyce filed for his IC patent on July 30, 1959; was awarded his
patent on April 25, 1961, which was ahead of Kilby’s.
12.4. Noyce did not receive the Nobel Prize in 2000 because he had
died in 1990. The Nobel Prizes are not awarded posthumously. (For details
of the Nobel Prize, see Chapter 12 and Appendix 7.)
Chapter 8. Noyce’s Invention of IC: Key Patent, Claims, and Analyses 247
Section 13. Detailed analyses of all the claims
Bob Noyce says it all in the very first paragraph of this patent.
This is essentially how the present monolithic ICs are made.
P- and N-type regions “extending to said surface” and “an insulating
layer consisting essentially of oxide of said semiconductor on
and adherent to said surface, said layer extending across a
different portion of said junction” are essentially derived from Hoerni’s
invention of planar technology. Noyce’s key invention lies in “comprising a
conductor adherent to said layer, said conductor extending from
said one contact over said layer across said different portion
of the junction”, in particular the adherence of the conductor to the
insulator films. The key issue of monolithic interconnects adherent to oxide
was settled by the decision of the CCPA, the highest appeals court to
hear the case, that Kilby did not disclose the adherent interconnects.19
(To clarify the legal aspects of patent interference proceedings, it is an
adjudicative hearing within the auspices of an administrative agency, the
US Patent Office. It is run like a mini-trial, but technically speaking it is
not occurring in a court of law. It is occurring before the Board of Appeals
of the USPTO.26
Chapter 8. Noyce’s Invention of IC: Key Patent, Claims, and Analyses 249
Section 13. Detailed analyses of all the claims
discoid contact but spaced and insulated from the ends of said C-shaped
contact.”
said layer being congenitally united with said body and extending
across said junction, and an electrical connection to said contact
comprising a metal strip adherent to said layer . . . ” Why was P-N
junction isolation, which was awarded to Lehovec (who had filed earlier but
awarded his patent later than Noyce — see Chapter 6, Table 6.2) was not
claimed per se by Noyce? See my comments in Claim 1 above.
Chapter 8. Noyce’s Invention of IC: Key Patent, Claims, and Analyses 251
Section 13. Detailed analyses of all the claims
Chapter 8. Noyce’s Invention of IC: Key Patent, Claims, and Analyses 253
Section 14. Copy of the original Patent No. 2,981,877 of Bob Noyce
February 18, 2009 14:19 B732 9in x 6in b732-ch08 FA
Chapter 8. Noyce’s Invention of IC: Key Patent, Claims, and Analyses 255
Section 14. Copy of the original Patent No. 2,981,877 of Bob Noyce
February 18, 2009 14:19 B732 9in x 6in b732-ch08 FA
Chapter 8. Noyce’s Invention of IC: Key Patent, Claims, and Analyses 257
Section 14. Copy of the original Patent No. 2,981,877 of Bob Noyce
February 18, 2009 14:19 B732 9in x 6in b732-ch08 FA
Chapter 8. Noyce’s Invention of IC: Key Patent, Claims, and Analyses 259
Section 14. Copy of the original Patent No. 2,981,877 of Bob Noyce
February 18, 2009 14:19 B732 9in x 6in b732-ch08 FA
Chapter 9
Besides Kilby and Noyce being recognized in the entire literature as the
sole inventors of the ICs, the only other name occasionally associated with
the invention has been that of Dummer. But this is restricted primarily
only to the technical literature, because Kilby had referred to and quoted
Dummer in his papers and in his Nobel speech. Kilby did not make any
reference to Dummer in any of his patents but instead, Kilby gave his own
version of the IC concepts in only one of his patents. These concepts given
by Kilby were strikingly similar and limited to the same constraints as
those given earlier by Dummer. Noyce did not refer to nor quote Dummer
at all in his papers and patent, perhaps because Noyce did not regard
Dummer’s concepts to be worth quoting or Noyce may not have been
aware of Dummer’s article. Regardless, it is proven by me in the preceding
descriptions that Dummer’s descriptions were irrelevant to the concept of
monolithic-ICs given by Noyce. There were others too who had given the
concepts for the ICs earlier than Kilby and Noyce and contributed to the
IC inventions. But why were they ignored? The answers to this question
are still elusive.
261
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
RRE had pursued grown junction technology for the transistors, which
was extremely difficult to control and make contacts to the necessary
layers of devices. At Bell Labs, the focus was on finding clever ways to
eliminate as many components and interconnections as possible, and to
fabricate functional blocks using mesa technologies for the transistors and
diodes. Mesa technology had been known by that time before the planar
technology was invented by Hoerni. So mesa technology was chosen by
Bell Labs. The results at both RRE and Bell Labs were dismal failures
to invent the ICs. While their efforts can be termed as gallant, they
“were barking up the wrong tree.”16 Even to this day, the technologies
initially pursued to invent the ICs, both at RRE and Bell Labs, have
made no contributions to the invention, or to the advancement of ICs,
and they are not used to manufacture them. However, Bell Labs did make
outstanding contributions to several technologies beyond the IC invention,
which are well known and documented in the literature. Similarly, key
contributions to several technologies for and beyond the IC invention were
made in the universities and research labs of various corporations not only
within the US, but also in other countries and regions of the world (a
partial alphabetical list includes Belgium, Canada, England, France, West
Germany, India, Israel, Italy, South Korea, Japan, Singapore, Taiwan, The
Netherlands).
1. Geoffrey Dummer
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 263
Section 1. Geoffrey Dummer
have been defined. These criteria used for such comparison, and whether or
not Dummer and Kilby met them are summarized in Table 9.1.
Table 9.1. Similarity between and limitations in both Dummer’s and Kilby’s concepts.
“At this stage, I would like to take a peep into the future. With the
advent of the transistor and the work in semiconductors generally, it seems
now possible to envisage electronic equipment in a solid block with no
connecting wires. The block may consist of layers of insulating conducting,
rectifying and amplifying materials, the electrical functions being connected
directly by cutting out areas of the various layers.”
Dummer4 did not file for a patent. But he had first published his
monolithic concepts in 1952, although they were incomplete to enable the
fabrication of the monolithic-IC. He had written that he was taking “a peep
into the future” at the very end of his paper. This was akin to wishful
thinking or the increasingly popular science and technology fictions such as
those presented in the TV programs and popular science fiction articles and
books, about the new things that could be achieved because the transistor
had become a reality. Dummer had only “envisaged” but did not give any
details of the materials and technologies needed in reality to fabricate the
“electronic equipment in a solid block with no connecting wires.” The three
sentences of Dummer quoted above were like dreaming or thinking out aloud
of the new possibilities. But they had many shortcomings and limitations,
and the details were missing to specify clearly how a monolithic-IC could
be fabricated following his concepts. Kilby’s description in his patent1 filed
in 1959 was similar and limited to the same shortcomings as Dummer’s in
1952. Kilby had attended Dummer’s presentation in 1952, so it is possible
to assume that Kilby may have derived his ideas from Dummer. He seems
to allude to that in his Nobel Lecture in 2000. However, this cannot
be proven beyond any reasonable doubt. (See comments in Chapter 7,
Section 17.1.)
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 265
Section 1. Geoffrey Dummer
Dummer had also quoted his concepts4.1 in a later paper that he had
published in 1964.4.2 A complete reprint of this paper is also given in
Section 7. However for readers’ convenience, portions relevant to Dummer’s
concepts and the work on integrated electronics in England are copied below
from this paper.
Towards the end of 1956 the use of silicon as a resistor was investigated
and resistors of approximately 1000 ohms were made from sliced crystals,
about 2 cm long × 1/2 mm square. A contract was placed with the
Plessey Company Research Labs. In April, 1957, for the development of
semiconductor integrated circuits on the lines indicated by the author.
A model demonstrating the technique of shaping the silicon crystals
to control semiconductor properties was shown at the International
Components Symposium at the RRE in September, 1957, as an illustration
of the possibilities of semiconductor integrated circuit techniques.
Early in 1958, work was begun in the RRE labs. on the development of
complex thin-film circuits using resistive, capacitive, and conductive films.
It was envisaged that the transistors would be inserted separately. The
aspect of reliability was considered to be important, and therefore inorganic
materials were considered essential for the required high standard.
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 267
Section 2. Harwick Johnson
material was chosen initially because of the ease with which it could be
evaporated. However, by the time of the circuit failure, silicon monoxide
was being satisfactorily evaporated and this was used in future work.
2. Harwick Johnson
Johnson5 had filed his patent in 1953, and it was issued in 1957. A
complete copy of this patent is given in Section 8 at the end of this chapter.
Kilby1,23 had filed his patent(s) in 1959, and they were issued in 1964 and
later. So clearly, Johnson patent predates Kilby’s patents.
1. Filed May 21, 1953; Serial no. 356,407; issued Dec. 10, 1957; assigned to
RCA.
2. Column-1; lines 15–17: “This invention pertains to semiconductor
devices and particularly to semiconductor phase-shift oscillators and
devices.”
3. Column-1; lines 18–22: “One basic form of semiconductor device
is known as a P–N junction transistor and comprises a body of
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 269
Section 4. Richard Stewart
3. Arjun Saxena
4. Richard Stewart
They shall not be elaborated here. Such intricate details are deemed
unnecessary for these patents regarding the invention of ICs.
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 271
Section 4. Richard Stewart
Why was this patent of Stewart not referred to in any of Kilby’s papers
and patents? It claims an integrated circuit with a plurality of transistors,
resistors, isolation by P–N junction, and interconnections by conductive
means.
Analyses of all the relevant claims to ICs in the patent are given; others
are coalesced.
(f ) first contact means engaging said small portions of the wafer on said
one major face, said first contact means being connected to said third crystal
means and effective in operation to provide substantially ohmic connections
to the emitters of the plurality of transistors,
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 273
Section 4. Richard Stewart
(g) a plurality of second contacts to the wafer on said one major face
ohmically engaging the second crystal means at said small portions so
that the second contacts will provide separate inputs to the bases of the
transistors,
(h) conductive means on said one major face ohmically engaging said
first crystal means at positions closely adjacent said small portions of the
wafer to provide collector contacts for the plurality of transistors,
(j) third contact means engaging the wafer on a major face thereof
ohmically contacting the other end of the resistor region, the third contact
means being spaced from the parts of the first crystal means which function
as the collectors of the transistors by distances much greater than the
spacing between such parts and the positions where the conductive means
engage the first crystal means,
(k) and means for applying operating bias voltage to the first and third
contact means.”
(h) means for applying operating bias potential across said second and
third conductive means to reverse bias the collector-base junction of each of
the transistors, the resistor region thereby providing a collector load resistor;
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 275
Section 4. Richard Stewart
said one face making low resistance ohmic contact to one end of said
resistor portion and to the collectors of the plurality of transistors
so that the collectors are effectively connected together ; separate
electrical connections including contacts on said one face to the bases of the
transistors; conductive means engaging said one face contacting the emitters
of the transistors; and an electrical connection to the other end of
said resistor portion.”
Stewart’s drawings also do not show the correct planar technology for
devices and interconnects, which has the same limitations as Dummer and
Kilby. However, Stewart’s claims have the additional concept of electrical
connections by conductive means, and lack only capacitor in the integrated
circuit.
5. Jay Last
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 277
Section 5. Jay Last
4. Key summary: Silicon is etched in between the devices all the way
through, and the space opened up is filled with a bonded insulating material
such as epoxy resin to isolate the devices and maintain the integrity of the
wafer.
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 279
Section 6. Reprint of the paper presented by Geoffrey Dummer
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 281
Section 6. Reprint of the paper presented by Geoffrey Dummer
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 283
Section 6. Reprint of the paper presented by Geoffrey Dummer
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 285
Section 7. Reprint of the paper by G. W. A. Dummer
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 287
Section 7. Reprint of the paper by G. W. A. Dummer
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 289
Section 7. Reprint of the paper by G. W. A. Dummer
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 291
Section 7. Reprint of the paper by G. W. A. Dummer
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 293
Section 7. Reprint of the paper by G. W. A. Dummer
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 295
Section 7. Reprint of the paper by G. W. A. Dummer
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 297
Section 7. Reprint of the paper by G. W. A. Dummer
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 299
Section 8. Harwick Johnson
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 301
Section 9. Richard Stewart
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 303
Section 9. Richard Stewart
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 305
Section 10. J. T. Last
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 307
Section 10. J. T. Last
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 309
Section 10. J. T. Last
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 9. Other Efforts to Invent and/or Contribute to the Invention of ICs 311
Section 10. J. T. Last
January 20, 2009 9:58 B732 9in x 6in b732-ch09 FA
Chapter 10
1. Kilby
After Kilby1,23 was awarded the IC patent nos. 3,138,743 and 3,138,744,
a few other patents were also awarded to him (see Chapter 7 and Table 10.1
below). However none of these, and no additional patents were awarded
to him which had described the Si planar technology and interconnects
that were adherent to SiO2 . It had been well established that they were
mandatory for fabricating the monolithic ICs. Kilby did not make any
contributions later to the planar and other technologies which were, and
are, essential to manufacture the conventional and advanced monolithic-
ICs. The subsequent patents that he did obtain were on miniature electronic
calculators and in other fields which were important contributions in their
own right, but not to the invention and further development of ICs.
Nevertheless, Kilby was given the recognition of being the co-inventor with
Bob Noyce of ICs based solely on perhaps two patents, viz., nos. 3,138,743
and 3,138,744. As mentioned also in Chapter 7, even Kilby19 refers only to
his patent23 no. 3,138,743 in his paper in 1998 on “Origins of the Integrated
Circuit”, and not to any other patent of his, when he was critically reviewing
his and Noyce’s fundamental inventions of the integrated circuit.
313
February 18, 2009 14:22 B732 9in x 6in b732-ch10 FA
Table 10.1. List of patents of Jack Kilby on and beyond the invention of ICs.
Kilby was also awarded patents later in various other fields, e.g.,
packaging, light energy conversion, crystal growth, thermal recording head
February 18, 2009 14:22 B732 9in x 6in b732-ch10 FA
Chapter 10. Contributions of Kilby and Noyce Beyond the Invention 315
Section 1. Kilby
for printers, teaching system, telephone answering system, etc. But they
are not listed above because they are not pertinent to the IC invention.
The obvious question is why Kilby did not refer to any of the other
patents when he reviewed critically the “Origins of the Integrated Circuit”
in his paper19 in 1998 after 34 years of receiving his patent no. 3,138,743?
The commonsense type of obvious answer is that since he did not receive
any other patent relevant to the IC invention, that is why he did not or even
could not refer to any patent other than no. 3,138,743 in his paper19 in 1998.
Kilby19 left the question of monolithic-ICs ambiguous and unanswered
by concluding that “Despite these introductions, the monolithic concept
remained controversial.” The fact that he chose to refer to 3,138,743 and
not to 3,138,744, even though he had at least stated the concept partially
for the monolithic-IC in the latter, was probably due to the earlier filing
date Kilby had claimed for the former. This was rather unfortunate indeed
because his claim for the earlier filing date for the former was not confirmed
by the USPTO.29,30 Another scenario, I suggested in the previous chapter,
could be a typographical error in the patent number during editing the
1998 article.
Chapter 10. Contributions of Kilby and Noyce Beyond the Invention 317
Section 2. Noyce
The above facts seem to negate the assertion that Kilby had
“maintained a significant involvement with the company that continues
to this day”, i.e., until 2000 when the Nobel Prize was awarded to him.
These facts are not crucial for the discussion on the invention of ICs in
this book. However, for the sake of completeness of reviewing critically
perhaps the world’s most important invention ever, the above facts have
to be documented based on the analyses I have given. One may infer from
them that after Kilby took a leave of absence7 from TI in 1970 “to do
some independent work” on photovoltaics, printers etc, he was not involved
actively with the ICs and their associated technologies anymore. In defense
of Kilby and to give credence to his busy schedule with other activities, these
facts may explain why Kilby did not, or could not, make any contributions
to ICs and their technologies after his initial patents were granted in 1964.
2. Noyce
After Noyce2 was awarded the IC patent no. 2,981,877, a few more
patents were awarded to him on other process and design related issues
of ICs. They are listed in Table 10.2 below. But no further patents were
awarded to him that went beyond the present 2D-ICs, only for which
Moore’s Law holds (Moore45,46 ; Saxena12,13,20,48 ). For example, Noyce did
not invent 3D-ICs or UPICs having materials and technologies for devices
in addition to Si, and for interconnects in addition to Al, which are used
today in many of the advanced ULSICs which are all 2D-ICs. Examples
of these are the use of Cu interconnects with appropriate barrier and cap
layers, W (tungsten) for contact and via filling, planarization of dielectric
and metal films, RIE, LPCVD, etc. The limitations of Moore’s Law for the
February 18, 2009 14:22 B732 9in x 6in b732-ch10 FA
Table 10.2. List of patents of Bob Noyce on and beyond the invention of ICs.
•Noyce (ICs) 2,981,877 Jul. 30, 1959 Apr. 25, 1961 (Key patent)
•Noyce 3,108,359 Jun. 30, 1959 Oct., 29, 1963
(Co-inventor with Gordon Moore)
•Noyce 3,150,299 Sep. 11, 1959 Sep. 22, 1964
•Noyce 3,117,260 Sep. 11, 1959 Jan. 7, 1964
•Noyce 3,325,787 Oct. 19, 1964 June 13, 1967 (2 co-inventors)
2D-ICs, however, can be removed by invoking the 3D-ICs and UPICs. This
is discussed in two memos of Saxena48 given to Gordon Moore at Intel. See
also two recent patents of Saxena37,38 and three recent papers.12,13,20 (See
Chapter 13.)
Chapter 10. Contributions of Kilby and Noyce Beyond the Invention 319
Section 2. Noyce
In the case of Noyce, the answers to the above questions are difficult
to arrive at even with the dialectic approach. Perhaps the simplest answer
can be the “trade secret”, a practice which actually Intel was following
at those initial formative times, years, well-known to most and proven by
patent filing history. Without any doubt, Noyce was the key idea generator
(inventor) of the monolithic-ICs, although many others did also make key
contributions to make them a reality as well as to the advancements of
the technologies and design of the ULSICs. Being a charismatic leader as
well as a team player with a hands-off management style, can be viewed as
somewhat contradictory.
If the above observations are correct, then Noyce could have co-
invented, if not claimed to have solely invented, several advanced IC
technologies, since he would have had easy access to many of their key
developments. But he did not do that. If anything, Noyce was too generous
to give his ideas freely to the others. Was he over-reacting to the criticism
that he had usurped the IC invention from the others, and that is why he
was being too generous after that? It is hard to tell. It is well known though
that Noyce was deeply involved in starting, managing and running Fairchild
and Intel. These activities would have forced Noyce not to be able to use his
creativity and contribute first hand to the ULSICs and beyond. However,
he did interact with several other scientists, engineers and entrepreneurs in
various technical fields, and helped them to start their own companies. So
there are no easy answers for Noyce’s lack of further direct contributions
and innovations in the IC field, despite the dialectic reasoning given above.
February 18, 2009 14:22 B732 9in x 6in b732-ch10 FA
Chapter 11
Discussion
This was the only patent awarded to Noyce for the invention of IC. It
states and claims the monolithic concept clearly using planar technology
and monolithic interconnects adherent to the insulator layers, and without
shorting to the regions adjacent to the devices (see Chapter 8 for details).
Even though the ICs have advanced to the current ULSICs in which more
than 2 billion transistors per chip with 45 nm dimensions and more than
8 levels of metallizations are used, Noyce’s invention is still fundamental
to them. Outstanding advancements of processes and materials have been
made to enable such ULSICs in large volume production, but Si as the start
material and Noyce’s invention remain the core.
Important reminders to the reader are that all the ULSICs being
sold in the market are 2D-Si-ULSICs, and only electrical functions can be
performed in them. They have all the devices laid out in the 2-dimensions
of the Si wafer, and the 3rd-dimension is used primarily for the multilevel
321
January 20, 2009 9:58 B732 9in x 6in b732-ch11 FA
reason also why Kilby did not refer to 3,261,081 also in his 1998 paper19
(see Section 3 below).
Kilby19 refers only to 3,138,743 with serial no 791,602 and the filing
date Feb. 6, 1959 in his 1998 paper. Kilby19 does not refer to his second
patent 3,261,081 nor to 3,138,744 (filing date May 6, 1959; serial no.
811,486) in his 1998 paper. Kilby also writes in this paper, “Despite
these introductions, the monolithic concept remained controversial.” In
my opinion, Kilby’s patent no. 3,138,744 was his KEY PATENT because
it was the ONLY patent of Kilby in which he had at least stated the
monolithic concept though partially correct and only partly consistent in
its text.
Why did USPTO still keep the filing date of 3,138,743 to be Feb. 6,
1959 is a mystery? The filing date of the second patent 3,261,081 is written
in its Column 1, lines 6–9 as “Original application Feb 6, 1959, Ser. No.
791,602, now Patent No. 3,138,743, dated June 23, 1964. Divided and this
application Mar. 16, 1964, Ser. No. 352,380; 21 Claims, (Cl. 29–155.5).” The
wording in this patent, “Divided and this application Mar. 16, 1964, Ser.
No. 352,380” seems strange, but it is clear that revised filing date was Mar.
16, 1964, and its serial no. was 352,380. As written in Section 2 above, the
claims of this patent 3,261,081 were also not supported by its specifications
(texts and the figures).
5. Monolithic concept:
Noyce’s patent states and claims the monolithic concept clearly using
planar technology and monolithic interconnects adherent to the insulator
layers, and without shorting to the regions adjacent to the devices
(see Section 6 below). Kilby19 wrote in his 1998 paper, “Despite these
introductions, the monolithic concept remained controversial” implying
that he did not agree with the monolithic concept described by Noyce. Kilby
did not refer in this paper to his patent 3,138,744 which at least stated the
monolithic concept though partially correct and only partly consistent in its
text. Instead, he chose to refer only to 3,138,743 which did not even mention
the monolithic concept. Kilby had filed a patent interference against Noyce
for earlier award, and claiming priority that the monolithic interconnects
January 20, 2009 9:58 B732 9in x 6in b732-ch11 FA
6. Monolithic interconnects:
patent for monolithic ICs had been issued, i.e., its contents were public
knowledge, and Kilby’s patent interferences against Noyce and Lehovec
were ongoing. The re-written claims of 3,138,743 and 3,261,081 which were
quite different from those in OA 791,602 seem to have been influenced
by the then public knowledge of Noyce’s patent 2,981,877 and Lehovec’s
patent 3,029,366. Nevertheless, Kilby’s claims were still inconsistent with
monolithic-ICs.
The fact that the claims of Kilby’s OA 791,602, patent 3,138,743, and
patent 3,261,081 were not entirely supported by their specifications (which
were almost the same in all three), would suggest a possible compromise
of the laws of US Patent code 35 USC 112 and associated protocols. The
proceedings in the award of patents 3,138,743 and 3,261,081 to Kilby, in
particular allowing the filing date to be kept as Feb. 6, 1959 in view of the
lawsuits and their outcomes (see Sections 5 and 7 above), appear rather
unusual. Additional issues that further beg clarifications are that all the
figures, which are exactly the same in the three documents of Kilby (OA
791,602, patent 3,138,743, and patent 3,261,081), show mesa structures
for the devices and wire bonded interconnects which are never used in
monolithic ICs. Also the figures in patent 3,138,744 show mesa structure,
and its text specifies materials and technologies most of which are not used
in monolithic ICs. The “shaping said wafer to obtain isolation between said
components in said wafer” appears to include mesa etching, and selective
formation of p–n junction formation. All the figures show the former but
not the latter at all. See also Section 7 above regarding the decision by the
Board of Patent Interference which ruled against Kilby in favor of Lehovec
in this matter of p–n junction isolation. It is also interesting to note that
the patents 3,138,743 and 3,138,744 differ in number by only 1, and they
were awarded to Kilby on the same date June 23, 1964. Grouping more
than one patent to be issued on the same day may not be unusual. But for
two successive patents with different sets of claims for achieving the same
invention, especially with their strange history, extra long review periods
for over 5 years, patent interferences, and awarded on the same date, all
of them raise some concern regarding the decisions of USPTO to award
Kilby’s patents.
January 20, 2009 9:58 B732 9in x 6in b732-ch11 FA
11. Were any of Kilby’s patents ever used to manufacture ICs? The
clear answer is “No”. Perhaps some aspects of the patents may be argued
to have been usable, but it is extremely debatable. Did they affect the huge
sums of royalties earned by TI from the others, and the patent agreement
between TI and Fairchild? Yes.
12. Table 1.1 describes the key requirements for making the monolithic-
IC and whether or not they were met by Kilby and Noyce in their
respective inventions. Re-stating them here, the four requirements are as
follows:
Any other approach without meeting these four key criteria will not
give monolithic-ICs. For a quantitative characterization of adhesion, which
was a key issue in Kilby vs. Noyce patent interference, see Saxena.28
January 20, 2009 9:58 B732 9in x 6in b732-ch11 FA
13. Noyce’s invention covered all the criteria in 12.1, 12.2, 12.3 and
12.4. So his invention was for monolithic-ICs.
14. Kilby’s invention(s) met only the criterion 12.1 of fabricating the
devices on the same single piece of a single crystal semiconductor. But Kilby
did not specify planar technology in any of his patent(s). Kilby used Ge to
demonstrate his reduction to practice, and
Therefore, Kilby’s invention did not meet all the criteria in 12.2, 12.3
and 12.4. Therefore his invention was not for the monolithic-ICs; instead it
was for hybrid-ICs.
15. Moss (Kilby’s co-op student in 1960 for testing Kilby’s “solid
circuits”):
the above facts as follows. The original copy of Moss’ article is given at the
end of this chapter.
17. The conclusion drawn from all the analyses of the documented facts
is that Noyce invented the monolithic-IC. But its reduction to practice was
done by the others who had worked under him. Kilby’s invention was at best
that of a hybrid-UC-IC, whose basic ideas were similar to those described by
several earlier contributors. The actions and the decisions of Kilby, USPTO,
January 20, 2009 9:58 B732 9in x 6in b732-ch11 FA
and the Nobel Committee are inexplicable. Similarly the inactions of almost
all of the other key contributors, scientists, engineers and patent experts for
about four decades to set the records of who invented the real IC straight,
are also inexplicable indeed.
COMMENTARY
Getting History Right is an Important Matter
Jeffery Marque
Jack Kilby won the Nobel Prize in the year 2000 for his role in
the invention of the integrated circuit, and many people assume that
the ICs now in use are due to Kilby’s invention. At his presentation,
however, Saxena presented extensive historical evidence pointing to
major roles by other inventors. I would say that Saxena’s main point is
that the monolithic IC was not invented by Kilby, but rather by Robert
Noyce. Instead, Kilby invented a hybrid IC, a type that is not used
commercially.Saxena emphasized the distinction between the monolithic
and the hybrid IC, referring in his paper to the former as “. . . the only
kind sold from the inception in the IC industry. . . ”
Quotations from his paper give very clear voice to the points that he
wished to emphasize:
“The issues in the inventions of ICs by Kilby, Noyce and the others
are intricately entwined technically, chronologically, and legally patent
wise . . . the key concepts for the monolithic IC were first documented
by Noyce, even though the reduction to practice of his invention was
done by others, and it depended crucially on Hoerni’s and Lehovec’s
inventions.”
“Kilby missed the key concepts of monolithic interconnects and
planar technology necessary to fabricate monolithic-IC. The reduction
to practice was done by Kilby [himself] using Ge [germanium] mesa
technology and wire bonded interconnects dangling above the chip which
are not used in monolithic-ICs. Kilby was awarded the Nobel Prize
in 2000, and he is generally regarded as the inventor of ICs, implying
monolithic ICs, which is not pedantically accurate.”
One of the more intriguing ideas presented by Saxena concerns
the filing date of one of Kilby’s patents. Kilby claimed a filing date
of February 6, 1959. However, when Saxena dealt with the patent
officeapplication, he received two contradictory responses: One response
indicated a filing date of May 6, 1959, and the other response said,
“The product or service you requested cannot be fulfilled because the
application . . . does not have an official filing date.” Saxena wrote in his
paper, “The above seemingly contradictory responses from the USPTO
cannot be explained . . . one fact is clear from the above responses: the
official filing date of Kilby’s [application] . . . was not February 6, 1959,
as claimed by Kilby . . . Either it was May 06, 1959 . . . or it did not have
an official filing date at all. . . ”
January 20, 2009 9:58 B732 9in x 6in b732-ch11 FA
Chapter 12
An unbiased, knowledgeable and serious reader, who has read this book
up to this point, would have found that I have given facts with proper
documents, not hearsay evidence of the IC inventions. My analyses of
various contributors have been objective and thorough, and I have given
the credits as well as criticism where they are due. In this chapter, the
remarks are based primarily on published facts of both Kilby and the
Nobel Committee. As I have stated in this book and my paper(s) published
recently,12,13 Kilby did make a key contribution to the invention of ICs,
but it was only a small part of the complete invention needed. With due
333
January 20, 2009 9:58 B732 9in x 6in b732-ch12 FA
respects for both Kilby and the Nobel Committee, in my opinion several
of their actions and decisions are rather inexplicable. I will concede though
that a casual reader of this book whose mind is made up and refuses to
accept or understand the truth may still harbor an erroneous impression as
if Kilby and the Nobel Committee are being assailed. But as I have written
before, such is not the case at all in this book.
The following are the highlights of the updated Nobel Awards copied
from the Nobel website. The original list of the five subjects in which the
Nobel Prizes were awarded was “physics, chemistry, physiology or medicine,
literature and peace”. A sixth subject, Economics, was added to this list
in 1968.
Kilby was listed last in this Nobel Prize,7 awarded 1/2 of the prize
money, recognized solely for the invention of the ICs, and was cited as “for
his part in the invention of the integrated circuit.” Kilby’s invention1,22,34
did not involve any contribution which was fundamental to physics. It
had used the technology concepts which had been described earlier by
the others (see Chapters 7 and 9). In the specifications of his invention,
Kilby did not give the correct procedures for fabricating the devices and
January 20, 2009 9:58 B732 9in x 6in b732-ch12 FA
The other two co-recipients of this Nobel Prize, Alferov and Herbert
Kroemer, were listed first in alphabetical order, each were awarded 1/4
of the prize money, and were cited jointly “for developing semiconductor
heterostructures used in high-speed- and opto-electronics.” Unlike Kilby’s
Nobel Award, the contributions of Alferov and Kroemer did involve
contributions to fundamental physics.
Why was Kilby given twice the amount of financial award than to
each of the other two co-recipients, unlike the equal amounts given in the
Nobel Prize in Physics awarded to Shockley, Bardeen and Brattain for their
invention of the transistor earlier in 1956? No explanation has been offered
January 20, 2009 9:58 B732 9in x 6in b732-ch12 FA
so far by the Nobel Committee for the vague and imprecise citation of such
an important world renowned Prize, nor for the unequal and the double
amount awarded to Kilby. It is generally known that the highly revered
group of the Nobel Committee acts as a powerful closed institution which
feels strongly that they do not owe any explanation to anyone regarding
their choices and decisions. Without casting any aspersions on anybody,
respectfully and punctiliously I only wish to state that the manner in which
Kilby’s Nobel citation and award had been handled is quite inexplicable.
Nevertheless one cannot obliterate the facts as published by the Nobel
Committee itself and elsewhere in the literature. They are cast in concrete
and therefore stated as such in this book.
If the Nobel Committee credited Kilby with the Nobel Prize award
for his part in inventing only a hybrid-UC-IC with mesa devices and wire-
bonded interconnects, then an obvious question arises, “Why the other
inventors of hybrid-UC-ICs earlier than Kilby were ignored?”
Noyce had died in 1990; the Nobel Prizes are not awarded
posthumously. So he was not included in the Nobel Prize awarded in 2000,
although he was recognized as the co-inventor of the ICs by Kilby in his
Nobel Lecture.7 That certainly was a gracious gesture by Kilby. Had Noyce
January 20, 2009 9:58 B732 9in x 6in b732-ch12 FA
not died, he would have been definitely included in this award in my opinion.
Perhaps then the entire Nobel Prize award would have been worded and
even awarded differently.
“You are right when you underline that Kilby did not invent the
monolitihic-ICs. But why stressing this point so many times? As you
correctly state, Kilby received the Nobel Prize ‘for his part in the invention
of the integrated circuit’. We considered this phrasing very carefully. The
prize was not given for the invention of monolithic-ICs! I always stress that
in my lectures (see attachment). If Noyce would have been still alive in
2000, maybe the phrasing would have been different. Who knows? As a
member of the Swedish Academy of Sciences, we are not allowed to give
any information on such issues within 50 years.”
It is nice to get agreement from Dr. MNC-1 when he writes “You are
right when you underline that Kilby did not invent the monolitihic-ICs.”
However, he still did not clarify what was Kilby’s part in the invention of
which kind of IC. Since the Nobel Prize was awarded in 2000, and they
cannot give out any information for 50 years, i.e., until 2050, I shall be long
dead when the Nobel Committee may reveal the truth.
So, quid pro quo, and there you are with the attest of one of my key
conclusions on who really invented the monolithic-IC; Bob Noyce did.
Albert Einstein (in 1930), Vladimir Zworykin (in 1935), Enrico Fermi
(in 1940), John Bardeen and Walter Brattain (in 1950), William Shockley
(in 1951), Robert Noyce (in 1961), J. B. Gunn (in 1968), etc. Most of the
inventions documented in this list revolutionized technology developments,
and changed mankind forever. However, the list does not include Jack
Kilby, whose patent was awarded in 1964. Apparently the criteria used
by the Nobel Prize Award Committee and by Professor Blair at the Pierce
Law Center do not jibe with each other. This is public knowledge and an
interesting fact on record. Pierce Law Center is no Nobel Committee or
vice versa depending on your viewpoint and basis; nevertheless it consists
of distinguished and highly qualified patent law professors and experts in
the USA which the Swedish Nobel Committee does not.
January 20, 2009 9:58 B732 9in x 6in b732-ch13 FA
Chapter 13
Moore’s Law
341
January 20, 2009 9:58 B732 9in x 6in b732-ch13 FA
key insight and business acumen was to observe the behavior of these
parameters of the then existing data of manufacturing various chips. Lo
and behold, Moore found that a semi-log plot of these parameters versus
time gave respective straight lines. To quote from Moore’s Chapter 741
that he wrote in the book by Brock,41 “Over time, the term (Moore’s Law)
was used much more broadly, referring to almost any phenomenon related
to the semiconductor industry that when plotted on semilog graph paper
approximates a straight line.” This empirical correlation of the existing
data provided the key guidance for the subsequent development of the
technologies, materials and equipment of IC manufacturing and forecast
their behavior in the future. This sums up essentially the birth, use, and
the key role of Moore’s Law in the entire microelectronics industry. It
sounds quite simple, but it has been a key beacon for almost all aspects of
the current multi-hundred-billion dollar chip manufacturing industry, e.g.,
chips, processing and test equipment. Nevertheless to repeat a key fact
which has been ignored in the entire industry, Moore’s Law applies only to
these present 2D-ULSICs in which only the electrical signals are used and
processed for circuit functions on a chip.20
Fig. 13.1. A semi-log plot of transistors per die (chip) versus time for the early data of
1965 until the recent years and beyond. The linear behavior according to Moore’s Law is
followed in various segments which were influenced by the associated technologies used
for the chips.
To expand a little on this eponymous law, Moore was the first to observe
the trend of the increase of number of devices per chip with time, which
was also followed by the increase of die (i.e., chip) area, and the decrease
of average minimum dimension of the device geometries. When these
parameters were plotted by Moore45,58 as semi-log vs. time, they followed
a linear behavior approximately. Since then, some minor corrections have
been made due to the recent data. However, such linear behavior has been
followed by the chip industry, and is known as the Moore’s Law. It has
predicted remarkably well the evolution and the growth of the technologies
and products in the entire multi-hundred-billion dollar monolithic Si IC
industry for the past three decades.
To give the readers a feel for how fast the industry has grown, I shall
give the total dollar volume of all the Si products including the ICs in
the world market (without discussing the details). The data is only for all
January 20, 2009 9:58 B732 9in x 6in b732-ch13 FA
1970 2.6
1990 49.5
2000 200
2007 250
The above data do not include any numbers for the 3D-ICs and UPICs,
because they have not yet been commercialized. I am quite hopeful that
they will be available commercially in the next few years. Not only they
will extend the validity20,48 of Moore’s Law, but the growth of the total
market will be even higher than given above so far. This is because a new
class of IC products will then be available, which will further enhance their
applications in a variety of systems heretofore not possible by the present
2D-Si-ULSICs.
Some observations on the current status and possible factors for the
limitations of Moore’s Law are summarized as follows. A few key items
January 20, 2009 9:58 B732 9in x 6in b732-ch13 FA
which have been ignored in the literature have been repeated below to
emphasize their importance.
1. All the ICs sold in the market are 2D-Si-ULSICs. This means that the
various devices needed for the ICs are laid out in the 2-dimension on a bulk
single crystal Si wafer. The 3rd dimension is used only for interconnecting
these devices by using multilevel metallizations and inter-layer dielectrics
(ILDs). Moore’s Law applies only to these 2D-Si-ULSICs.
4. One of the key features of Moore’s Law is that it predicts the rate
of growth of the total number of devices per chip with time. This requires,
and it also forecasts, the rate of decrease of minimum geometries, or
critical dimensions (CDs), of the devices and interconnects. Consequently,
increasing the number of electronic functions per chip requires device and
interconnect scaling, and increasing the number of multilevel interconnects.
In addition, photolithography, alignment of various masks, etching patterns,
planarization and zero-defect manufacturing are all becoming more and
more critical, and they are absolutely essential. These approaches are
necessary if we continue along the current technology evolution path.
However, they are becoming very expensive, and cause reliability problems,
because of the limitations of the materials and the manufacturing
technologies below certain minimum geometries. Thus, it is becoming
harder to maintain the historical trends of the yields and profitability of
the more advanced ULSICs.
unlikely that they will be suitable in the limit of atomic dimensions. While
no definite limit has been established yet, it appears that the validity of
Moore’s Law will cease at CDs smaller than about 0.018 µm (or 18 nm). This
means that the rate of increase of the number of devices per chip in a 2D-
Si-ULSIC, will not follow Moore’s Law for CDs <0.018 µm. This rate will
slow down initially, and it shall flatten out eventually if we continue down
the evolutionary path of the 2D-Si-ULSICs. Thus we can define the limit
of the validity of Moore’s Law as 0.018 µm. Alternatively we can define the
regime where Moore’s Law is not applicable, to be the regime when the CDs
are <0.018 µm. The fundamental limits of materials and manufacturing
technologies, and the additional problems in device physics and reliability,
do provide a signal that paradigm shifts are needed to extend the validity48
of Moore’s Law. The criterion to be used for this is to continue to increase
the number of devices, consequently the number of functions, per chip in
the future at a rate comparable to the same rate as predicted by Moore’s
Law so far. Technologies that will enable this,48 will allow the continued
growth of the microelectronics industry, without pushing the sub-0.018 µm
2D-Si-ULSIC technologies.
discussed in this book. For the current status, applicability, limitations and
the technologies to extend Moore’s Law, see Saxena.48
In a recent article,59 Moore was asked, “What would you like your
legacy to the world to be?” He replied shaking his head ruefully, “Anything
but Moore’s Law.” I have known Gordon Moore for about 48 years. In
my opinion, there is more to it than his tongue-in-cheek reply. However,
this book is not the place to expound on it. Moore’s Law with all its
ramifications will continue to impact the technologies and the Quality of
Human Life (QHL) in the future, and is here to stay forever.
Chapter 14
Despite the dominance of the chip industry in the world whose business
volume is now multi-hundred billion dollars, which feeds into the systems
market of several trillion dollars annually, a natural question arises, “What
good has the invention of ICs done for mankind to improve the quality
of human life (QHL)?” I was asked this question recently and to give
my assessments.20 It is a very important question, but its answers will
necessitate volumes to be written on the various aspects that affect the
QHL. Obviously that cannot be done in this chapter, or even in one book.
However, I shall adapt from my earlier paper20 to give the list of possible
answers to at least some aspects of the QHL.
349
January 20, 2009 9:58 B732 9in x 6in b732-ch14 FA
The criteria and their needs by various human beings for a desired
QHL, vary from one person to another, within a family, within a community,
within a state, within a country, and not the least, within the world under
zillions of circumstances. So, it is just infinitely impossible to discuss them
on an individual basis for each human being. Nevertheless, there are some
common denominators to the QHL needs of all mankind. Without even
attempting to give an exhaustive account, at least a few salient criteria
which are common to almost every human being in the modern and
ever changing world shall be listed below. Brief comments will be made
on the impact of the new products generated by ULSICs and UPICs,
and their applications to improve the QHL. At the outset, I wish to
request forbearance from all the experts in different fields of this important
subject. Perhaps I may be providing the paradigm shift in thinking about
QHL in some respects, as I have done in the core issues of ULSICs
and UPICs.
The sequence of writing the criteria below does not necessarily imply
prioritizing. Depending on one’s own circumstances at a given time, they
can be assigned different emphases and re-ordered into a different sequence.
Therefore, it is a continuously evolving and variable set of criteria. This of
course illustrates the complexity of discussing QHL, because even for the
same person, the importance of various criteria and their sequence will
change, and evolve with time. Nevertheless, my comments are as follows.
Instead of repeating “ULSICs and UPICs”, the generic term “ULSICs” shall
only be used. ULSICs are available today; the UPICs, which will further
provide the quantum leap for the QHL, have yet to be commercialized.
Chapter 14. Growth of ICs and Impact on the Quality of Human Life 351
2. Food and Water: Everybody needs food and water to survive. ULSICs
are used in agriculture, processing, purification, monitoring, controlling, etc.
QHL is improved when we have adequate supply of healthy nutritious food
and clean water.
Even walkers and runners use pedometers which use ULSICs. Improved
QHL requires improved travel, thus improved ULSICs.
Chapter 14. Growth of ICs and Impact on the Quality of Human Life 353
maintenance use computers and ULSICs. For a good QHL, we must have
safe environment, thus good ULSICs are important.
15. Strong defense & secure country: For any QHL, we need to live in a
secure country which requires strong defense. Computers and ULSICs are
the core of all the systems used for these purposes. Therefore, advancing
the technologies of computers and ULSICs enhances the QHL.
16. Global business: In today’s economy, global business is the key. QHL
depends on the strength of every country’s capability in this area. Without
the computers and ULSICs, participation in global business is impossible.
The discovery of the transistor led to the invention of ICs, which then
progressed to the ULSICs, enabling the computers and a host of other
equipment of today. They have contributed immensely to improve the
January 20, 2009 9:58 B732 9in x 6in b732-ch14 FA
Chapter 15
1. Conclusions
The salient conclusions from the analyses of all the documented facts
on the invention of ICs given in this book are as follows.
5. Several others also gave concepts for the invention of ICs earlier than
Kilby and Noyce. However, they have been ignored in the literature. In a
way this fact is akin to an initial snowflake growing into a snowball and
then generating a giant avalanche, stampeding over anything old or new
that came in its way. However the fundamental initiation of the snowflake
itself akin to the IC invention has also been railroaded by the powers to
be in a manner unparalleled in the history of science and technology in the
world.
been clarified so far? Why was it not ascertained earlier by the patent
attorneys of TI, and also by those at Fairchild even after getting the
Noyce patent? Apparently this issue was also not brought up at the
proceedings of Board of Patent Interference and Court of Customs
and Patent Appeals during the contesting of Kilby’s and Noyce’s
patents.
The conclusions drawn here are not meant at all to be pejorative and
disrespectful to Noyce, Kilby and all the other contributors, or to impugn
their contributions. The same is also true regarding the actions and the
decisions of the USPTO and the Nobel Committee. My only objective is to
put all the available important facts on record. What is credible evidence
depends to a certain degree, in a way similar to beauty, on the eyes of the
beholder. Nevertheless, these are the records in writing, and in technical
English with engineering precision, not just a pleasure to the eyes.
2. Combined summary
concepts. However, I had not only given the concepts (see Appendix 2),
but I had also pursued and contributed subsequently to the development
and advancement of ICs. Even though the details are given in Appendix 2
following the sage advice given to me by Gordon Moore55 (see Section 11
in Preface also), I have chosen to include myself in Table 15.1 of this last
chapter. I have also chosen to include Dummer in this table despite the
limitations and drawbacks of his IC concepts, and that he did not pursue
their development further after giving his concepts. The Table 15.1 gives
the summary of the invention of ICs by Kilby, Noyce, Dummer and me
in a tabular form. It compares my contributions with theirs on a one-to-
one correspondence basis. At one glance, it gives the relevant comparative
information and distinguishes my insight from the others (following Gordon
Moore’s advice; see Section 11, Preface), not coming as a hindsight but as
it occurred originally in time and as it progressed. The readers hopefully
will find it interesting to read the pros and cons of the similarities and
limitations between all of them in Table 15.1, and get the details of my
contributions in Appendix 2.
The STRONG and WEAK points of Kilby and Noyce have already
been given in Chapter 5, Section 5.2. Similar listing for Saxena is given
below. Such discussion is not applicable to Dummer because he had only
enunciated his concepts theoretically; he did not even reduce his concepts
to practice or give any information on how to fabricate the devices and the
IC, and there is no record that he pursued to develop them further.
by Arjun N. Saxena
INVENTION OF INTEGRATED CIRCUITS: UNTOLD IMPORTANT FACTS
Important
facts Kilby Noyce Dummer Saxena
1. What was Hybrid-UC-IC only; Monolithic-IC. Incomplete concepts for Hybrid-UC-IC and
invented? NOT monolithic-IC. hybrid-UC-IC. monolithic-IC.
B732
prior to
disclosing IC
concepts?
3. First public Original application US patent filed on Jul. Paper published in 1952. Described my concepts in
disclosure of filed on May 6, 1959 30, 1959. a lecture on Jan. 14,
invention. (not on Feb. 6, 1959); 1954; original
US patent filed also handwritten notes on
9in x 6in
on May 6, 1959. transistors; paper
published in 1953.
4. Test circuit Yes; in 1958. No (but done by others) No. Yes; given in the
defined. published 1953 paper.
5. Reduction to Yes, but used Ge mesa No (but done by the No; also no information No, but key approaches
practice of transistor, wire others using Si was given how to described in my lecture
original bonded to other planar technology, fabricate the devices in 1954; unpublished.
invention. devices; not used which is used in ICs). and ICs.
b732-ch15
in ICs.
(Continued)
FA
February 19, 2009 8:52
Section 2. Combined summary
Chapter 15. Conclusions, Combined Summary, Historical Facts
Table 15.1. (Continued)
Important
facts Kilby Noyce Dummer Saxena
B732
limitations of these yield monolithic-ICs. other official
patents to yield only documents.
hybrid-UC-ICs.
9in x 6in
technologies
after original
invention?
b732-ch15
3D-ICs and developed and
UPICs? manufactured.
361
FA
February 19, 2009 8:52 B732 9in x 6in b732-ch15 FA
3. Historical facts
not like the documents such as the issued patents or published research
papers cited in the various references given in the earlier chapters. However
it must be emphasized that they also do not belong to the category of
recollections or hearsay evidence. These historical facts are well known
almost to everybody in the IC field, even though they have not been
published in a precise and cohesive manner so far. Some of them have been
given here only for a few cases to lend or deny support to the documented
facts even in a secondary manner. Nevertheless they are important. While
giving the historical well known facts below, my choice is not to mention
any names of the people other than those already referred to in this book.
However, the old and new references have been given.
3.2. Dummer4 described only a part of the concept for solid state
integrated circuits almost as an after thought by writing “At this stage,
I would like to take a peep into the future . . . ” in the very end of his
paper published in 1952. It was like stretching one’s imagination beyond
what had been known about the transistors and diodes in 1952. Until then,
essentially grown junction and mesa technologies were known to fabricate
them. The planar technology with Si had not been discovered until 1959
(see 3.4 below).
3.3. Kilby also described only a part of the concept for integrated
circuits in his notebook in 1958, which was strikingly similar and limited
also only to the part given by Dummer, and it was unwitnessed (Wolff3 ).
But neither Dummer nor Kilby gave the additional parts of the concept,
viz., the technologies needed to fabricate, isolate and interconnect them
February 19, 2009 8:52 B732 9in x 6in b732-ch15 FA
on a chip for making the IC. Further, Kilby’s reduction to practice of his
invention was that of a hybrid-UC-IC which had used Ge mesa technology
to fabricate various devices in Ge chip(s), and they were interconnected by
gold wire-bonds. (See Fig. 1.1).
3.4. Hoerni8 has been credited for the invention of the planar technology
with Si in 1959. As I have documented in this book (see Chapter 6), his
invention was built upon a combination of important work done by several
others earlier. Nevertheless using this technology, highly reliable planar Si
transistors (much more reliable and reproducible than point contact and
mesa transistors) were sold to the DOD for the Apollo, Minuteman Missile,
and other programs at high premium prices.
3.7. Most of the people including the experts understood only bits and
pieces of the various key issues of the invention of ICs, but not their totality.
As explained by me in the earlier chapters and recent papers,12,13 they were
inextricably entwined technically, chronologically and patent wise. This
may explain why nobody documented and published anything authoritative
earlier to characterize correctly what was invented by whom and what
was not. The characterizations by a few science history writers by getting
second or third hand information from interviews, and not having the first
hand involvement and experience, have been incomplete, misleading and
erroneous.
wise. So it has not been easy to unravel the various issues, discern and
prove beyond any reasonable doubt whether the decisions by the various
parties at a given time were justifiable or not. As an example, both Noyce
and Hoerni were colleagues at Fairchild when Hoerni’s planar patents were
filed on May 1, 1959, and Noyce’s patent on July 30, 1959. But Hoerni’s
invention was not cited and acknowledged in Noyce’s patent filing, even
though it was crucial to his invention. The legality of such a decision may
be debatable, but the fact is clear. Also, Noyce and his senior colleagues and
patent attorneys did not state categorically that Kilby’s invention will not
enable monolithic-IC. The last straw that broke the camel’s back was their
focus primarily on the monolithic interconnects adherent to the insulating
layers, and they did not emphasize the Si planar technology as well. Both
of these key factors were winners in my opinion for Bob Noyce’s invention
of the monolithic-ICs. But their clear superiority over Kilby’s invention
was not enforced by Fairchild. A compromise between Fairchild and TI was
accepted in which Kilby’s mesa technologies for device fabrication, which
are never used in monolithic-ICs, and Noyce’s monolithic interconnects
adherent to the insulating layers were essentially given equal importance;
the concept of fabricating all devices within a single piece of Si was
common to both Kilby and Noyce. Moreover, this concept had already been
published earlier than Kilby by several people such as Dummer,4 Johnson,5
and Stewart.6 Why this fact was also not brought to the attention of
the USPTO?
3.9. The key issue of the filing date of Kilby’s original patent application
(OA) was not established correctly earlier, as evidenced by the recent
written communications of the USPTO with me29,30 in 2005. Three patents
were awarded to Kilby based on OA in a procedure which possibly
compromised the US patent procedures and laws. Kilby kept on referring
to the OA in his early patents and recent paper19 (even though the OA was
essentially rejected by the USPTO), and to his first patent22 which was the
weakest of the three. It can easily be inferred that Kilby’s decision to do this
was primarily because of its alleged earlier filing date. The third patent1
at least had stated a part of the monolithic concept partially correctly, but
it was strikingly similar and limited in scope also to that part described
by Dummer4 earlier. Kilby chose to ignore giving credit to Dummer in
his patents, but acknowledged it later in his papers19,21 and the Nobel
speech.7
February 19, 2009 8:52 B732 9in x 6in b732-ch15 FA
3.10. Richard Stewart’s patent 3,138,747 was filed on Feb. 12, 1959,
and issued on June 23, 1964, exactly the same date on which Kilby’s
patents 3,138,743 and 3,138,744 were issued. All of them were assigned
to TI. Stewart had been working at TI before Kilby had joined TI in 1958.
But neither Stewart nor Kilby referred to each other’s patents. Stewart’s
concepts were similar to those described by Kilby for his version of the ICs.
It is strange indeed that throughout his career at TI, and even afterward
until he had received the Nobel Prize, Kilby did not acknowledge Stewart’s
contribution, a fellow worker at TI, while he did acknowledge Dummer’s
contribution made primarily in England. Kilby also did not acknowledge
Harwick Johnson’s contribution made earlier at RCA.
3.12. A few of the key people who are still alive and claim the
importance of their key contributions to make the invention of ICs a reality,
have done little or nothing to publish and document to prove and claim their
assertions.
3.13. The Nobel citation7 in 2000 of Kilby read, “. . . for his part in
the invention of the integrated circuit”. But what exactly was his part for
inventing which kind of IC has never been spelled out and explained so
far by the Nobel Committee. For details of the Nobel Prize in Physics as
they have been awarded throughout their history of existence since 1901,
and specifically to Alferov, Kroemer and Kilby, see Appendix 7. Also as
discussed in Chapter 12, Noyce would certainly have been included in this
Nobel award had he not died earlier in 1990. Then perhaps both Noyce
and Kilby may have been cited jointly as “the inventors of the integrated
circuit”, even though Kilby had invented only a hybrid-IC and Noyce
had invented the real monolithic-IC. This would have been akin to the
Nobel Prize awarded jointly to Shockley, Bardeen and Brattain in 1956,
even though the latter two had invented only an impractical point contact
transistor, and Shockley had invented the practical bipolar junction and
February 19, 2009 8:52 B732 9in x 6in b732-ch15 FA
3.14. Kilby and Noyce have been universally regarded as the two sole
entrenched American icons who invented the ICs. This was despite the
fact that Kilby had invented only the hybrid-IC in which the devices were
connected together only by bonded wires, while Noyce had invented the
monolithic-IC in which the inventions of the others in addition to his team and
his own (lithography camera) were used and were mandatory for it to work.
Lucky are those who have been able to balance between the two
extremes, each finding their respective happy middle ground. Unfortunately,
however, most of the people in the world regard the latter extreme
(Financial rewards) to be a greater achievement of success than the former
(Professional rewards), whether or not the person has made any important
contributions professionally. This has been proven to be true time and again
throughout the history of mankind that money talks. “C’est la vie” or “So
spielt das leben”. It is also well known, however, that in innovations by sole
or multiple inventors resulting in recent technological advances, usually one
person could see the overall view with so many individual components and
non-traditional concepts pertaining to the key invention. This has been the
key ingredient for those who were or became the leader of the team and
recognized as the originators of a successful invention. An example of this is
the bipolar junction and field-effect transistors invented by Shockley alone,
although they were based on the knowledge he had gained from the work
of the others who had surrounded and reported to him at Bell Telephone
Laboratories.
4. Unanswered questions
The above important historical facts though not belonging to the same
class of documents as the published papers and patents, nevertheless well
known for decades if not centuries, lead to several unanswered questions, a
few of which are listed below.
February 19, 2009 8:52 B732 9in x 6in b732-ch15 FA
4.2. Why the facts of the filing date of Kilby’s OA, i.e., original
application (which was so important to claim the earliest filing date of
the IC invention) were not ascertained correctly earlier? It appears still
to be in a state of confusion at USPTO as evidenced by their written
communications with me in 2005. As of writing this book, my attempts
to obtain the “Certified Copy each of the File History of patent numbers
3,138,743 and 3,138,744, both being based on the original application
number 03,791,603” from the USPTO have not been successful. The
response from the USPTO is given in Fig. 15.1 on the next page. It states,
“We are unable to fill your order because the files for the above patent
numbers above are unavailable due to being in the lost category. . . .” I
have initiated further inquiry to find out why such important files have
February 19, 2009 8:52 B732 9in x 6in b732-ch15 FA
Fig. 15.1. Communication from USPTO to Arjun N. Saxena, dated July 03, 2008,
received by Saxena on July 10, 2008, regarding Saxena’s attempts to get the “Certified
Copy each of the File History of patent numbers 3,138,743 and 3,138,744, both being
based on the original application number 03,791,603” from the USPTO. The response
from the USPTO states, “We are unable to fill your order because the files for the above
patent numbers above are unavailable due to being in the lost category. . . .”
February 19, 2009 8:52 B732 9in x 6in b732-ch15 FA
been “lost” by the USPTO. So far I have not received any reply from the
USPTO.
If the files of “Certified Copy each of the File History of patent numbers
3,138,743 and 3,138,744, both being based on the original application (OA)
number 03,791,603” ordered and paid for by me were not “lost” by the
USPTO, they would have helped to clarify at least two mysteries:
4.3. The USPTO and all the patent experts must have been aware of
the possible compromise of the laws of US Patent code 35 USC 112 and
associated protocols when the key patents of Kilby were processed and
issued. Why did the USPTO allow this? Why no expert in the patent field
has published any report since then on it to disclose and clarify the facts
for the past four decades?
4.4. Why nobody has analyzed Kilby’s patents and discovered the errors
and irregularities, and published them earlier than my recent papers?12,13
4.7. Similar to the question no. 4.6 above, the p–n junction isolation
technology invented earlier by Lehovec9 was also mandatory for isolating
devices in the invention of IC. (Since then, other isolation technologies
such as LOCOS and trench isolation have been developed and are being
used.) I have described already in Chapter 6 the technical reasons why
Noyce may have chosen not to file claims for the p–n junction isolation,
and also why he may not have acknowledged Lehovec in his patent. Please
note the word “may” that I have used in the previous sentence, which
I have used to emphasize my doubt in Noyce’s favor. I am making this
qualifying disclaimer because the technical reasons are more apropos for the
CMOS circuit designs, which comprise the largest segment of the entire chip
business today. The concept of CMOS did not even exist in the literature in
1959 when Noyce’s patent was filed, although it was soon (in less than two
years ∼1961) the highest secret project at the Fairchild R&D Lab, known
only to three persons (Grinich, Moore, Sah) and Noyce.69 Hoerni’s 1959
motivation for his invention was based primarily on the stability of the npn
bipolar junction transistors, a fact gleaned from the leakage current via
the ambient-exposed-surface channels of the n-collector/p-base chemically
etched mesa junctions. Nevertheless, while no claims were filed by Noyce,
the p–n junction isolation was mentioned but not described properly in
Noyce’s patent to “teach one of ordinary skill in the art how to construct
the device”. However, Noyce did acknowledge Lehovec’s isolation patent
later in his 1977 paper.43
4.8. Why the key inventors other than Kilby and Noyce who had
contributed to the invention and advancement of ICs did not speak up
earlier, and write, clarify and publish to document the facts of the IC
inventions and claim their indispensable roles so far?
been known that Kilby had invented only a hybrid-IC, and Noyce had
invented the real monolithic-IC. Both could have been jointly credited
before Noyce’s death in 1990 as the inventors of IC. This is akin to the
invention of the transistor as discussed in Section 3.14 above. Nevertheless,
awarding the Nobel Prize in Physics in 2000 to Kilby as a co-recipient with
Alferov and Kroemer, still raises three questions:
4.9.3. Why Kilby was given 1/2 of the prize money, and Alferov
and Kroemer were given 1/4 each in 2000, whereas Shockley,
Bardeen and Brattain had each received 1/3 of the prize money
equally in 1956? (Shockley was the director of the solid-state
electronics research group at Bell Telephone Laboratories, even
though substantially younger than the rest who “reported” to him.
Nevertheless, Shockley’s two transistor articles were indeed based on
basic physics and used sound mathematical equations. They led to
the invention of the two transistors, that meets the strictest and the
February 19, 2009 8:52 B732 9in x 6in b732-ch15 FA
5. Epilogue
not just the popularity, because of the truths uncovered no matter when or
how late, their basic initiatives and contributions cannot be ignored. Most
of the readers must be familiar with the rise and fall of the icons in different
fields of scientific research, and in the history of mankind. Instead of giving
some esoteric example, such as that of a giant of a physicist like Einstein,
whose iconic stature will never be diminished despite his lack of noteworthy
success during the latter half of his illustrious career and all those tabloid
accounts on his personal life, one will be given which should hit the right
chord with most of the readers.
A caveat that I would like to mention here is about the hero worship
of the icons. It is almost cult like allegiance to ranging all the way from the
heavenly icons to the mortal humans, may they be scientists, entertainers,
sports stars, political leaders, etc. While the majority of the people will
accept the truths uncovered no matter when, and adjust and modify their
beliefs of their icons accordingly, a minority cannot be swayed from their
erroneous and blind worship. They represent a prejudiced mind which
stubbornly refuses to accept anything different to learn, thus the neural
interconnections remain unchanged in its brain, akin to a stone under
which nothing grows. Regarding the sensitive and personal area of religion,
which has been itemized and commented upon briefly by me in the QHL
discussions in Chapter 14, even the majority of the people have inflexible
views on their respective beliefs. Unfortunately this is the reality of mankind
from the early times to even today. Nevertheless, I hope that with consistent
diligent efforts to get the history right, proper education, and with the
advent and explosive growth of the internet, myths can be dispelled and
everybody will accept the truths and learn to live with each other in peace
February 19, 2009 8:52 B732 9in x 6in b732-ch15 FA
and harmony eventually. Most importantly, I hope that we all will learn
from the truth and its history, and to advance even faster to a higher
pinnacle. In my opinion, this will be perhaps the greatest contribution of
the invention of ICs to mankind.
Like Einstein, Kilby and Noyce certainly belonged to the first category.
Getting history right in this book and documenting the truth about the
contributions of Kilby, Noyce and the others must not be misinterpreted.
Its aim is not to take away or diminish the glory of Kilby and Noyce as
icons, or to promote others. Certainly it is also not to chastise and cast
aspersions on the actions as well as inactions of all the responsible people
worldwide in industry, academia and patent law. The fact that I have been
able to give the documented facts to get history right of such an important
invention as that of IC which has changed mankind forever, and raise the
unanswered questions, is a rare privilege indeed. Unfortunately both Kilby
and Noyce are no longer with us. Perhaps this book will stimulate the other
living icons in all the different fields of ICs, in particular the semiconductor
equipment industry and the chip industry of the microprocessors, to add to
the facts documented by me. The awareness in entire mankind reinforced by
the increasing impact of the modern internet revolution, is that the entire
world is becoming more secular in a very broad sense, and that the truths
in many areas will continue to manifest and will be disseminated worldwide
inevitably. It is already happening at an ever accelerated pace. This gives
us hope for a better future by knowing and learning from the truth, such
as in the present topic we have discussed, of who invented the IC. On
this last issue, I hope that I have written the last chapter to document
the truth.
Both Noyce and Kilby had acknowledged that it was a stroke of good
luck for them to have invented the ICs. Good luck did play a greater role
February 19, 2009 8:52 B732 9in x 6in b732-ch15 FA
for them and their respective versions of the invention of ICs, than it did for
several others who had also worked very hard, on known and/or predesigned
approaches and plans, rather than just shear luck, to make the monolithic-
ICs a reality in the marketplace from day one. There are many other
scientists, engineers, marketing and sales personnel all over the world who
also deserve the recognition for their respective invaluable contributions
to monolithic-ICs, and for advancing them to the ULSICs and the super-
chips of today. However, to select a few other than Noyce and Kilby for
whatever their contributions to the invention of ICs might be, and singling
out Moore for his contributions to take the entire industry beyond the
IC invention to what it is today, the names (listed alphabetically) of
Hoerni,8 Lehovec,9 Kooi,25 and Moore45,46,47 should also be on top of
the list.
The clap of the thunder of invention of ICs may be gone and belong
only to a few. However, the thunder usually lasts momentarily or for a short
duration only. But the resulting rains, akin to the invaluable contributions
of many, bear the fruits and the crops for a long time to come, whether
it is in the Silicon Valley and/or in the other global valleys. Certainly
the invention of the ICs has borne, and continues to bear, the fruits and
crops like the ULSICs to benefit all mankind. It is almost certain that the
additional fruits like the 3D-ICs, UPICs, etc. will also become realities in
the future and benefit everybody. Whether it will be in my and readers’
lifetimes or not is of little or no consequence. What is satisfying to note
is that the future generations will reap the harvest of the creativity and
hard work of all the contributors so far who themselves have stood on the
shoulders of the giants, recognized or not. Further not the least, I am sure
that the future generations will develop newer, more advanced and powerful
inventions to serve mankind. They shall be based on the prior knowledges
and experiences, which I hope my book here will provide the lead-in
description to them on just one of the contributions, perhaps the most
dominant one that is responsible for all the other advances, viz., the silicon
semiconductor monolithic integrated circuit, which we call it as IC and
the chip.
January 20, 2009 9:59 B732 9in x 6in b732-app01 FA
Appendix 1
379
January 20, 2009 9:59
Appendix 1. Listed in chronological order; only a few publications given.
380
Then Now
by Arjun N. Saxena
INVENTION OF INTEGRATED CIRCUITS: UNTOLD IMPORTANT FACTS
No. Field/Subject Year(s) Contributions Status in 2008
1. Solid State 1953–54 Published my first paper in electronics on 55 years later, it is a ULSIC (Ultra Large
Devices the improved design and performance Scale Integrated Circuits)
(Diodes/ of a pulse generator circuit, and microelectronics business whose
Transistors/ICs) I reduced it to practice in 1953 using volume is currently multi-hundred-
vacuum tubes. (See next Section billion dollars per year, affecting
no. 2.) Gave my first lecture on almost every industry in the world,
transistors; point-contact and alloyed with a total electronics systems volume
B732
junction transistor technologies in 1954 of multi-trillion dollars yearly.
(before the planar technology was
introduced in 1960). I had used the
above circuit as an example for
miniaturization by using solid state
devices in place of vacuum tubes. I
also gave the concept of Integrated
9in x 6in
Circuits (ICs — this term did not even
exist then) to fabricate and
interconnect transistors, diodes,
resistors and capacitors for a desired
circuit on a single substrate (nowadays
referred to as a “chip”). Even though
I had given the concept of ICs in 1953,
and described it in my lecture in 1954,
I neither documented it in a legal
b732-app01
manner nor applied for a patent. I was
unaware of those procedures at that
time. However, a few of the attendees
of my lecture in 1954 who are still alive
(Continued)
FA
January 20, 2009 9:59
Appendix 1. (Continued)
B732
awarded the Nobel Prize in Physics in
2000 “. . . for his part in the invention
of integrated circuit”. But what was
his part in the invention of which kind
of IC has never been clarified.
2. Electronic 1952–54 Introduced the diode clamp to improve Much improved electronic circuits now
Circuits the pulse generator; I constructed it available with ICs and computer
9in x 6in
and published my first paper (see controlled data logging and analyses.
Section 1 above); {cf: “A Modification
of Model 50 Pulse generator”, Current
Science, 22, 199 (1953)}; built highly
regulated high-voltage power supplies
for scintillation counters; single
channel pulse height analyzer.
3. Nuclear Physics 1954–56, and Co-authored a paper published in 1955, Many more elementary particles have
1960–61 and gave independent evidence for the been discovered (post quark era) since
b732-app01
existence of nuclear shell structure my time when I had produced π and µ
from beta decay. This was the first mesons in my PhD work; affecting
paper to give the shell structure of many theories of various types of
nuclei from beta decay. Earlier papers nuclear forces and phenomena;
381
(Continued)
FA
January 20, 2009 9:59
Appendix 1. (Continued)
382
Then Now
by Arjun N. Saxena
INVENTION OF INTEGRATED CIRCUITS: UNTOLD IMPORTANT FACTS
No. Field/Subject Year(s) Contributions Status in 2008
B732
interaction; discovered n–p pairing American Institute of Physics, January
interaction in odd-odd nuclei; gave a 26, 2007.
new method of calculating ZA in the
mass formula; advanced work in n–p
pairing interactions {cf: Proc. Phys.
Soc. London, 69, 293 (1956), & Phys.
Rev., 121, 595 (1961)}. I was qualified
9in x 6in
to receive my PhD in 1956 from The
Institute of Nuclear Physics, Calcutta.
But I took the advice from my
professors to get it from Stanford
instead. So I came to Stanford in 1956
and got it from there.
4. High-energy 1956–60 Investigated the enhancement of Pair production/annihilation for
Physics Bremsstrahlung produced by 575-MeV mono-energetic high-energy photons;
electrons in a single crystal of Si; storage rings.
b732-app01
produced π mesons to detect
mono-energetic photons by π → µ → e
decay; designed and constructed
several types of equipment; Si crystal
(Continued)
FA
January 20, 2009 9:59
Appendix 1. Developments in Physics, Microelectronics
Appendix 1. (Continued)
Then Now
No. Field/Subject Year(s) Contributions Status in 2008
B732
311 (1960); Phys. Rev., 125, 1720
(1962)}.
I had received the single crystals of Si at
Stanford for my experiments from
Texas Instruments in 1956–57 before
Jack Kilby had joined it in 1958. Also,
it was approximately the same time
9in x 6in
when Bob Noyce, Gordon Moore, etc.
had come to join Shockley in Mt.
View. They had started to use Si for
devices, while I was using Si for my
PhD research at Stanford to study the
coherent production of bremsstrahlung
by 575 MeV electrons. This electron
beam was generated by Mark III
accelerator which was then the largest
b732-app01
electron linear accelerator in the world.
I had also helped to assemble and
maintain it at Stanford.
(Continued)
383
FA
January 20, 2009 9:59
384
Appendix 1. (Continued)
by Arjun N. Saxena
INVENTION OF INTEGRATED CIRCUITS: UNTOLD IMPORTANT FACTS
Then Now
5. Invention of ICs, 1960–present Entered this field in 1960, the year Si Today’s technology is a far cry from when
Basic IC planar transistor invented by Jean I had started. The progress in the past
Patent, and IC Hoerni was announced. I had already 48 years has been mind-boggling. My
Technology given lectures on transistors, and had recent paper has documented and
given the concept of ICs earlier in 1953 corrected the misleading and
B732
and 1954 (see Sections 1 & 2 above). incomplete information of the
My technical inputs were sought by invention of ICs which had been given
the patent attorneys at Fairchild in by several other authors for the past
1960–61, regarding the choice and 48 years. It also gives a discussion of
adhesion of metal films to SiO2 for the incomplete citation in the Nobel
monolithic-IC fabrication. This was Prize awarded to Kilby in 2000 for his
one of the key issues which had purported invention of IC, which was
9in x 6in
resulted in the award of the basic IC not for the monolithic-IC, the only
patent to Bob Noyce (in 1961) ahead kind sold in the market from the very
of Jack Kilby (in 1964), even though beginning onwards. See my paper at:
Kilby had filed earlier than Noyce. I http://www.nsti.org/Nanotech20
had given the concept of ICs in 07/WCM2007/Saxena.pdf
1953–1954, but I did not document nor At present, I am completing a book on
apply for a patent. I was completely “Invention of Integrated Circuits”,
unaware of those procedures at that which documents all the information in
time. While I did build my improved greater detail than what I could give in
b732-app01
circuit in 1953 with the vacuum tubes my paper above.
available at that time to improve the
(Continued)
FA
January 20, 2009 9:59
Appendix 1. Developments in Physics, Microelectronics
Appendix 1. (Continued)
Then Now
B732
practice with solid state devices
to demonstrate my concepts for
ICs. Neither the transistors and
diodes nor the facilities for their
fabrication were available then
to me. I could only talk about
the solid state devices in India
9in x 6in
in 1954. Later, I obtained my
US patents on interconnects,
semiconductor equipment and
next generation ICs in 1972,
1984, 1995, 1998, 2000 & 2002
(see Section 26). I had obtained
other patents on solid state
devices earlier.
6. Si Epitaxy 1960–62 Developed early epitaxial process; Used routinely now in ICs.
b732-app01
first planar epitaxial bipolar
transistors fabricated on
material grown by me.
(Continued)
385
FA
January 20, 2009 9:59
Appendix 1. (Continued)
386
Then Now
by Arjun N. Saxena
INVENTION OF INTEGRATED CIRCUITS: UNTOLD IMPORTANT FACTS
No. Field/Subject Year(s) Contributions Status in 2008
7. Schottky 1962– Developed manufacturing processes and Used routinely in bipolar ICs; a good
design curves for Schottky diodes and Schottky barrier on p-Si still needed;
Schottky clamped T2 L ICs; published primary technology used in GaAs ICs
basic studies on deviations from ideal (MESFETS).
Schottky behavior; gave a universal
plot which allows one to determine at
B732
a glance the current flow mechanism in
any junction; Schottky-Gate FETs; I-V
behavior of Schottky barriers on
heavily doped Si; ohmic contacts {cf:
Appl. Phy. Letters, 14, 11 (1969)};
GaAs Schottky barriers; GaN Schottky
barriers {cf: Int. J. Appl. Rad. and
9in x 6in
Isotopes, 26, 33 (1975)}; introduced Hf
in IC technology and published papers
on Hf Schottky on P-Si {cf: Appl.
Phys. Letters, 19, 71 (1971)}; Schottky
barriers with various silicides; new
guard-ring structures; solar cells;
obtained patents on improved Schottky
barriers in 1969, 1971 and 1972.
8. Ellipsometry 1963– Introduced ellipsometry for the Used routinely in process control in Si
b732-app01
measurement of film thicknesses; VLSI/ULSIC microelectronics,
refractive index, optical constants {cf: optoelectronic systems, and IC
J. Opt. Soc. Am., 55, 1061 (1965)}, manufacturing.
etch rate, surface cleanliness
(Continued)
FA
January 20, 2009 9:59
Appendix 1. (Continued)
B732
GaAs, Hf, HfO2 (now being considered
as one of the high-k dielectrics in
CMOS in 2008!) {cf: Appl. Phys.
Letters, 46, 2788 (1975); Appl. Phy.
19, 25 (1979)}.
9. Bipolar 1961– Planar epitaxial bipolar transistor; R-F Multibillion dollar business; solar cells
power transistors; overlay transistors; important as energy source in
9in x 6in
Schottky T2 L technology; high-speed commercial, defense and space
ECL technology; solar cells. applications.
10. MOS 1965– Independent MOS model; Qss ; annealing CMOS is the largest segment of multi-
effects; independent isoplanar process; hundred-billion dollar IC business;
oxidation in high-electric fields; O−2 highest device packing density and
model; radiation-hard MOST; early lowest power consumption.
work on self-aligned MOST and VT
adjustment by ion implantation;
NMOS; CMOS; polycides; salicides;
b732-app01
BI/CMOS; Schottky-Source-
Drain-MOS; LDD; SRAMS; 3D-ICs;
Chairman of First International
BI/CMOS Symposium;
387
(Continued)
FA
January 20, 2009 9:59
388
Appendix 1. (Continued)
Then Now
by Arjun N. Saxena
INVENTION OF INTEGRATED CIRCUITS: UNTOLD IMPORTANT FACTS
No. Field/Subject Year(s) Contributions Status in 2008
B732
Intel Corporation’s “Pentium” issued
by Stanford University Video
Communications.
11. Ion Implantation 1968– Taught the first device physics course on Used routinely now in microelectronics
Ion Implantation; range-energy curves; manufacturing; MeV implantation has
VT control; Si3 N4 , SiO2 , SiC unique applications, in particular in
formation, GaN; obtained patent on advanced CMOS.
9in x 6in
etch-less patterning; MeV implantation
use; RBS; invited papers in Brussels
and Strasbourg. {cf: J. Matls. Sc. and
Eng. B2, 1–13 (1989)}.
12. Multilevel 1969– Keynote addresses at the 1st and 3rd Most important technology for ULSICs;
Metallizations VMIC, and at the 1st International chip size and speed performance
Workshop on W and other refractory determined by multilevel
metals; contact metallurgy; ρc ; metallizations; dominates yield,
electromigration; stress voiding; effect reliability and profitability. The
b732-app01
of passivation; jn ; planarization; number of multilevel interconnects on
dielectrics; silicides/polycides; chips having more than billion devices
salicides; refractory metals; Al and is now >8.
(Continued)
FA
January 20, 2009 9:59
Appendix 1. (Continued)
B732
13. SiO2 1977– Gave the first measurement and More basic work needed on adhesion of
characterization of the chemical state films; important in ULSI and
of Phosphorus in SiO2 {Proceedings of packaging.
the International Topical Conference,
Yorktown Heights, NY, pp. 195–199,
March 22–24 (1978)}; kinetics of the
change of its chemical state on
9in x 6in
annealing and processing {cf: J.
Electrochem. Soc. 132, 932 (1985)};
effect on thermal reflow; thin-film
thermal conductivity; applied it for
quantitative characterization of
adhesion of films in ULSI. {cf: First
International Congress on Adhesion
Science and Technology; Mittal
Festschrifft, eds. W. J. van Ooij and H.
b732-app01
R. Anderson, Jr., VSP International
Science, Utrecht, The Netherlands,
ISBN 90-6764-291-6, pp. 137–146
(1998)}.
389
(Continued)
FA
January 20, 2009 9:59
390
Appendix 1. (Continued)
Then Now
by Arjun N. Saxena
INVENTION OF INTEGRATED CIRCUITS: UNTOLD IMPORTANT FACTS
No. Field/Subject Year(s) Contributions Status in 2008
14. Fiber Optics 1972 Company president and CEO. Turned the Most important for fiber optic
company around, and got it ready to go communications; now a reality in
public. Won against The Teamsters telephone, satellite; medical, defense,
Union. etc.; multi-billion dollar industry.
15. Medical 1973–74 Developed a new quantitative technique to Hundreds of millions of dollars have been
Diagnostics characterize cytologic specimens; spent to have a quantitative technique
B732
demonstrated feasibility to diagnose and to screen and diagnose cancer routinely
screen Pap smears for cancer in patients in Pap smears, but no such technique
quantitatively (as compared to the has come to fruition so far. My
subjective/qualitative evaluations by quantitative medical diagnostic
conventional pathology); also used in-vitro technique appears quite promising and
cells known to be normal and cancerous has applications to diagnose other type
from the National Cancer Research of cancers also, but it needs further
9in x 6in
Institute, and showed unambiguous and development.
quantitative characterization with my new
technique; it has applications in various
other applications in cytology and medical
diagnostics.
16. Optical Detectors 1979 Chapter in handbook. Good progress; used routinely now in
fiber optic communications.
17. Measurement 1979 Keynote address. Essential for ULSIC manufacturing.
Science
b732-app01
18. Lithography and 1969– Editor of fine line lithography issue of Solid Essential for ULSIC manufacturing.
Dry Etch State Technology, 1980; obtained patents;
wrote chapter on dry etch.
(Continued)
FA
January 20, 2009 9:59
Appendix 1. (Continued)
19. Aluminum 1980– “All you wanted to know about Al Still a basic metallization technology in
Technology metallizations but were afraid to ask”; ULSICs, although Cu-based multilevel
reliability studies. metallizations are used now in
advanced ULSICs.
20. Silicides and 1970– Earliest review talk on silicides in I.C. Used in production.
Salicides technology; TiSi2 by RTA.
21. Refractory 1971– Use of Hf in Schottky; W use in I.C. W technology met my predictions; used
B732
Metals technology; my predictions of the use of now in production.
W in manufacturing of ULSICs became a
reality; keynote speaker of 1st
International Conference.
22. Dielectrics in 1980– Inorganic and organic dielectrics. Used in ULSI.
ULSI
23. Recording: Audio 1981– Was awarded a key U.S. Patent # 4,280,148 This recording field has grown now to
9in x 6in
and Video on July 21, 1981, in the shortest time by billions of dollars.
the US Patent & Trademark Office.
24. Planarization 1983– Review paper; gave key methods and Indispensable in ULSI production.
development of technologies.
25. GaAs/Si 1987– Invited talk on process issues. Feasibility demonstrated; important for
UPICs.
26. UPICs and 1987– Introduced Ultra Performance ICs (UPICs), Feasibility shown; huge potential of new
3D-ICs which include 3D-ICs. UPICs are the multi-billion dollar business.
ultimate in the monolithic ICs needed by
b732-app01
mankind, and they allow integration levels
beyond ULSICs, thus extending the
validity of Moore’s Law. In addition,
UPICs will allow both electronic and
391
(Continued)
FA
January 20, 2009 9:59
392
Appendix 1. (Continued)
by Arjun N. Saxena
INVENTION OF INTEGRATED CIRCUITS: UNTOLD IMPORTANT FACTS
Then Now
B732
#6,392,253 (May 21, 2002).
27. Cooperative 1987– Invited talks in 1987 and 1990; key roles in Crucial for microelectronics in the USA.
Support SEMATECH; Member of Executive
Strategies in Committee; Director of SEMATECH
the USA Center of Excellence.
28. High Tc 1988– Invited talk in 1988. Promising new applications.
Superconductors
9in x 6in
in ULSI
29. Advanced 1991– Course development and teaching; Important in today’s global economy.
Microelectronics Advanced Microelectronics
Manufacturing Manufacturing; equipment technologies.
30. Adhesion Science 1995– Gave the first and only invited talk on the Important impact on the reliability of the
and role of adhesion science and technology in ICs and many other fields.
Technology microelectronics at the 1st International
Congress on Adhesion Science and
Technology in 1995; (see Section 13
b732-app01
above); proposed new quantitative
diagnostic technique.
(Continued)
FA
January 20, 2009 9:59
Appendix 1. Developments in Physics, Microelectronics
Appendix 1. (Continued)
Then Now
No. Field/Subject Year(s) Contributions Status in 2008
31. Education, 1987– Emeritus Professor (1996-for life.) Important in today’s global economy.
B732
Research and Director of Center for Integrated A graduate fellowship “Veera and
Manufacturing Electronics (1987–89); established Arjun Saxena Fellowship in
microelectronic fabrication facility for Microelectronics” has been established
education and research; Professor (1987–) at Rensselaer Polytechnic Institute
gave invited papers/lectures because of his teaching and their
internationally (Belgium, Brazil, China, substantial donation. This Fellowship
Czech Republic, England, France, is awarded every year to outstanding
9in x 6in
Germany, India, Italy, Japan, PhD students in advanced research in
Netherlands, Romania, and USA); microelectronics. It is a real pleasure
Consulting Editor of a series of books on to receive letters from several students
and consultant in Microelectronics who have already benefited from this
Manufacturing; awarded four fundamental Fellowship. We will not be around
U.S. Patents recently on the next sooner or later, but this Fellowship will
generation ICs beyond ULSI/GSI which live forever.
will extend the validity of Moore’s Law.
(see Section no. 26)
b732-app01
393
FA
January 20, 2009 9:59 B732 9in x 6in b732-app01 FA
Appendix 2
In the main text of this book, I have given all the relevant documents
available for the invention of ICs by Kilby, Noyce and several others,
and analyzed them thoroughly. Regarding the concepts of ICs given by
me earlier than Kilby and Noyce that I shall write in this Appendix 2,
I have followed the sage advice given to me by Gordon Moore,55 “On the
part relating to your early insight you should do whatever you think is
appropriate. . . . I think that your job is to distinguish your insight from
what others saw as the potential.” My objective is to do just that when
I give my concepts of ICs in this Appendix. I think that it is appropriate
to at least document them for the sake of completeness of the historical
record without trying to claim any credit. I have also given the relevant
comparative information of Kilby and Noyce to distinguish my insight from
theirs.
395
January 20, 2009 9:59 B732 9in x 6in b732-app02 FA
Table App. 2.1. Summary of comparisons of Kilby, Noyce and Saxena’s IC inventions
Important facts
of IC invention Kilby Noyce Saxena
Important facts
of IC invention Kilby Noyce Saxena
As documented in the Table App. 2.1 above, my concepts for the ICs
predate Kilby’s and Noyce’s. The credibility of my documents has some
similarities with those of Kilby and Noyce. The reprint of my paper in a
journal is no match for the US patents filed by Kilby and Noyce. However,
the handwritten but un-witnessed notes are in the same category as those
of Kilby and Noyce, because theirs were also handwritten and un-witnessed.
The official announcement and the letters from 5 well qualified attendees
of my 1954 lecture attest to what I had said. Unfortunately the details of
my concepts for ICs were not documented when I had given them in 1954.
I could not file for a patent in 1954 because I had no knowledge of doing
so then.
January 20, 2009 9:59 B732 9in x 6in b732-app02 FA
It is known from publications that Dummer4 also did not file and obtain
a patent for his disclosure even though he had lived in England and given his
paper in USA where the procedures for patents were known. However, I did
receive legally recognized US patents in devices, USLICs and beyond later as
listed in Table 5.2 in Chapter 5, and also in Table App. 2.2 below. Therefore,
because of the weakness in the documentation of the details of the concept
for ICs that I had given in 1954, I have taken myself out from the list with
the others given in Section 13 of Chapter 1. This makes it clear that I am
not trying to claim any credit for inventing ICs earlier than the others.
However I am giving all the available documents and what I had disclosed
in my lecture in 1954 after publishing my paper in 1953, for the sake of
historical record in this Appendix 2.
Full copies of these original documents are given in the Sections 9–12
of this appendix. Even though their list is given above, they are referred to
and my patents in devices, ULSICs and beyond are listed in Table App. 2.2
below. The patents and earlier unpatented but documented work are
listed chronologically with patent filing dates and public disclosures as
publications. This procedure is similar to that I have followed for Kilby,
Noyce and the other contributors (see Chapters 5 and 9). To emphasize,
listing patent filing dates and public disclosures chronologically is important
because it documents the origin and sequence of conception of an invention,
January 20, 2009 9:59 B732 9in x 6in b732-app02 FA
Table App. 2.2. Documents and patents of Saxena on and beyond the invention of ICs
•Saxena (ICs) Gave concepts of ICs in 1953–54, but patent not filed.
Reprint of 1953 paper; original handwritten notes and
announcement of the lecture given on Jan. 14, 1954.
(See Sections 9–11.) Documentation of Saxena’s
comments in 1954 by several qualified personnel who
had attended his lecture. (See section 12.)
• Saxena 3,450,957 Jan. 10, 1967 Jun. 17, 1969 1 co-inventor
• Saxena 3,450,958 Jan. 10, 1967 Jun. 17, 1969 -----------
• Saxena 3,599,323 Nov. 25, 1968 Aug. 17, 1971 - - - - - - - - - - -
• Saxena36 3,687,722 Mar. 10, 1971 Aug. 29, 1972 - - - - - - - - - - -
(Interconnect)
• Saxena 3,694,719 Nov. 27, 1970 Sept. 26, 1972 -----------
• Saxena 3,700,979 Apr. 7, 1971 Oct. 24, 1972 -----------
• Saxena 4,056,642 May 14, 1976 Nov. 1, 1977 1 co-inventor
• Saxena 4,057,460 Nov. 22, 1976 Nov. 8, 1977 1 co-inventor
• Saxena 4,243,865 May 14, 1976 Jan. 6, 1981 -----------
• Saxena 4,436,582 Oct. 28, 1980 Mar. 13, 1984 -----------
• Saxena 5,212,118 Aug. 9, 1991 May 18, 1993 -----------
• Saxena 5,472,508 Jan. 14, 1993 Dec. 5, 1995 -----------
• Saxena 5,792,270 Oct. 21, 1993 Aug. 11, 1998 -----------
• Saxena 6,103,019 Feb. 18, 1998 Aug. 15, 2000 -----------
• Saxena37 6,110,278 Aug. 10, 1998 Aug. 29, 2000 Next
(3D-ICs & generation
UPICs) technologies
• Saxena38 6,392,253 Aug. 6, 1999 May 21, 2002 Next generation
(3D-ICs & ICs; extends
UPICs) Moore’s Law.
which is not reflected by the issue dates of the patents. The process in
between the filing and the issue dates of the patent, as well as what each
inventor did beyond his respective invention to advance its technology to
what it is today, are also important to acknowledge and critique each
contribution.
diagrams and cross sectional structures of these devices, and explained their
electrical performance during my lecture. No photographs were taken of
these drawings. However, I did state that the semiconductor transistors and
diodes would replace the vacuum tubes (as documented in my handwritten
notes; see Section 11) in the future electronic circuits, and would enable
their miniaturization.
5. Choice of semiconductor
7. Interconnects
Like everybody else in the world, I was unaware of the planar technology
in 1953–54. It was invented much later in 1959 by Hoerni8 at Fairchild
Semiconductor; although Hoerni had combined the important work done
by several others before him and claimed his planar invention in two patents.
I had specified in my concepts of miniaturization that the devices could be
interconnected by alloying leads or wire-bonding, which was earlier than but
similar to what Kilby1 had done independently a few years later in 1958, and
described in Kilby’s23 Application Serial No. 791,602. I had also suggested
that techniques such as painted metal lines or silk-screening through stencils
instead of wire-bonding could be used also to connect the devices. The
former would have given a hybrid-UC-IC and the latter a monolithic-IC.
I wish to acknowledge that I have been very lucky in many ways too.
Having had the opportunities to be a part of the epoch of the birth and the
growth of the ICs, and several other fields of physics and microelectronics
and make my contributions (see Appendices 1 and 2), has been a rare
privilege indeed. To quote from Frank Sinatra’s famous song, one important
January 20, 2009 9:59 B732 9in x 6in b732-app02 FA
The clap of the thunder of invention of ICs may be gone and belong
only to a few. However, the thunder usually lasts momentarily or for a short
duration only. But the resulting rains, akin to the invaluable contributions
of many, bear the fruits and the crops for a long time to come, whether
it is in the Silicon Valley and/or in the other global valleys. Certainly
the invention of the ICs has borne, and continues to bear, the fruits and
crops like the ULSICs to benefit all mankind. It is almost certain that the
additional fruits like the 3D-ICs, UPICs, etc will also become realities in
the future and benefit everybody. Whether it will be in my and readers’
lifetimes or not is of little or no consequence. What is satisfying to note
is that the future generations will reap the harvest of the creativity and
hard work of all the contributors so far who themselves have stood on the
shoulders of the giants, recognized or not.
This notice was issued in 1954 by Brig. SVS Chowdhry when he was
a student and the Secretary of the Physical Society, Lucknow University,
Lucknow, India, dated January 12, 1954, for the lecture on “Transistors”
January 20, 2009 9:59 B732 9in x 6in b732-app02 FA
I could find only one sheet so far of my original handwritten notes for
the talk I had given on “Transistors” in 1954. Another file of my notes is
still missing or lost. Since my talk was only on “Transistors”, this original
sheet contains comments only on transistors. The very first sentence of the
handwritten notes states, “Transistors will replace Vacuum Tubes”. These
notes do not have any of the comments that I had given verbally on the
January 20, 2009 9:59 B732 9in x 6in b732-app02 FA
Appendix 3
419
January 20, 2009 9:59 B732 9in x 6in b732-app03 FA
Appendix 3 421
January 20, 2009 9:59 B732 9in x 6in b732-app03 FA
Appendix 3 423
January 20, 2009 9:59 B732 9in x 6in b732-app03 FA
Appendix 3 425
January 20, 2009 9:59 B732 9in x 6in b732-app03 FA
Appendix 3 427
January 20, 2009 9:59 B732 9in x 6in b732-app03 FA
Appendix 3 429
January 20, 2009 9:59 B732 9in x 6in b732-app03 FA
Appendix 3 431
January 20, 2009 9:59 B732 9in x 6in b732-app03 FA
Appendix 3 433
January 20, 2009 9:59 B732 9in x 6in b732-app03 FA
Appendix 4
435
January 20, 2009 9:59 B732 9in x 6in b732-app04 FA
Appendix 4 437
January 20, 2009 9:59 B732 9in x 6in b732-app04 FA
Appendix 4 439
January 20, 2009 9:59 B732 9in x 6in b732-app04 FA
Appendix 4 441
January 20, 2009 9:59 B732 9in x 6in b732-app04 FA
Appendix 4 443
January 20, 2009 9:59 B732 9in x 6in b732-app04 FA
Appendix 4 445
January 20, 2009 9:59 B732 9in x 6in b732-app04 FA
Appendix 4 447
January 20, 2009 9:59 B732 9in x 6in b732-app04 FA
Appendix 4 449
January 20, 2009 9:59 B732 9in x 6in b732-app04 FA
Appendix 5
451
January 20, 2009 9:59 B732 9in x 6in b732-app05 FA
Appendix 5 453
January 20, 2009 9:59 B732 9in x 6in b732-app05 FA
Appendix 5 455
January 20, 2009 9:59 B732 9in x 6in b732-app05 FA
Appendix 5 457
January 20, 2009 9:59 B732 9in x 6in b732-app05 FA
Appendix 5 459
January 20, 2009 9:59 B732 9in x 6in b732-app05 FA
Appendix 6
Å = Angstrom
Al = Aluminum
ALS = Amyotrophic Lateral Sclerosis
AMD = Advanced Micro Devices
ASAP = As soon as possible
Au = Gold
BiCMOS = Bipolar and CMOS on the same chip
BPI = Board of Patent Interference
C◦ = Centigrade
CCPA = Court of Customs and Patent Appeals
CDs = Critical dimensions
CMOS = Complementary MOS
CNFET = Carbon nanotube Field Effect Transistor
CS = Compound Semiconductor
Cu = Copper
2D-ICs = 2-dimensional ICs
3D-ICs = 3-dimensional ICs
DG-FET = Double-gate-field-effect-transistor
DOD = Department of Defense
DRAM = Dynamic Random Access Memory
ECS = The Electrochemical Society
Fab = Fabrication facility
GaAlAs = Gallium Aluminum Arsenide
GaAs = Gallium Arsenide
GaN = Gallium Nitride
461
January 20, 2009 9:59 B732 9in x 6in b732-app06 2nd reading
Appendix 6. Alphabetical List of Acronyms and Abbreviations Used in this Book 463
O = Oxygen
OA = Original Application
PCBs = Printed circuit boards
PECVD = Plasma enhanced chemical vapor deposition
PhD = Doctorate in Philosophy
PN junction = In situ junction between p-type doped and n-type
doped semiconductor
P2 O5 = Phosphorus-pent-oxide
Proc. Phys. Soc. = Proceedings of Physical Society
QHL = Quality of human life
R&D = Research and Development
RCA = Radio Corporation of America
RIE = Reactive ion etching
RRE = Royal Radar Establishment
Sb = Antimony
Si = Silicon
SiGe = Silicon-germanium alloy
SiO2 = Silicon-di-oxide
SOCs = Systems-on-a-chip
SOI = Silicon-on-insulator
Stacked FET = Stacked Field Effect Transistor
Strained-Si FET = Strained-Si Field Effect Transistor
TI = Texas Instruments
ULSICs = Ultra Large Scale Integrated Circuits
UPICs = Ultra Performance ICs
USA = United States of America
USC = United States Code ; also University of Southern
California
USPTO = United States Patent and Trademark Office
V. P. = Vice President
W = Tungsten
January 20, 2009 9:59 B732 9in x 6in b732-app06 2nd reading
Appendix 7
1. Introduction
The Nobel Prize is regarded as the most prestigious and highly visible
award in the whole world. Winning it is like getting the stamp of approval to
be the par excellence elite in one of the six fields in which it is awarded. The
whole world stands in awe of and reveres the winner. All the information
given by me in this Appendix 7 can be obtained in the original website,
www.nobelprize.org. If all the information is public knowledge, a natural
question that could be posed to me by a reader, “Why are you re-writing
some of the information from this document if it is already available to the
public?” I would regard this as an excellent question indeed.
1. As you have read in this book, almost all the people in the world have
known of Kilby and Noyce as the only two co-inventors of the integrated
circuit (IC), an invention that has revolutionized mankind forever. The
latter part of the statement, i.e., it has revolutionized mankind forever, is
absolutely correct, but the former, i.e., Kilby and Noyce were the only two
co-inventors, is not entirely correct. If this recognition accorded to Kilby
and Noyce was entirely wrong or completely correct, then it would have
been easy to prove it either way as completely false or as being the gospel
truth respectively. Since it is neither, it is like half-truth in which it is not
465
January 20, 2009 9:59 B732 9in x 6in b732-app07 FA
always easy to discern the fact from fiction. However, sorting this out and
solving the mystery is important to get the history right and let the truth
prevail for the sake of mankind, especially for the younger generation. This
task gets even harder if the misinformation has been entrenched in the
society for a very long time, which is now for about 50 years in the case of
the invention of ICs.
2. While solving the mystery, the reader(s) should keep in mind that
the only kind of ICs sold from the very beginning in the chip business of
the microelectronics industry have been the Si-monolithic-ICs, commonly
referred to as just ICs. Kilby did not invent the Si-monolithic-IC, but Noyce
did. At best, Kilby’s invention1 was for a hybrid-IC. Also a less known fact
is that Noyce’s invention2 used others’ (Hoerni8 and Lehovec9) inventions
and did not acknowledge them in his patent.
5. The citation for the above Nobel award to the co-winners Alferov,
Kroemer and Kilby was published as “for basic work on information and
communication technology”. The part of the citation specific to Alferov and
January 20, 2009 9:59 B732 9in x 6in b732-app07 FA
6. The sequence of listing the names, i.e., the order in which the names
were listed in the Nobel Award in 2000 was “Alferov, Kroemer and Kilby”.
Before we worry about the sequence and the distribution of the Nobel
Prize money to ALL the other 3-person awardees in Physics, we must
recall the facts about the Nobel Prize in Physics awarded in 1956. The
sequence of listing the names of the Nobel Prize in physics in 1956 was
Shockley, Bardeen and Brattain in this order, and each had been given
equal amounts of 1/3rd of the Nobel Prize money. Even though the prize
money was distributed equally to all three, their names were not listed
in alphabetical order. Shockley was listed as the senior recipient followed
by Bardeen and Brattain in alphabetical order. There is a reason behind
this non-alphabetical listing of their names in 1956 which is not generally
known. The Nobel Committee never divulged this reason. (See Section 8.1)
8. As you have read in this book, and will read it again in section 6
below, a few facts about Kilby’s invention and the Nobel Prize are as
follows.
8.1. The citation of the Nobel Award to Kilby was vague and
inconsistent with the purported invention of the monolithic-IC by Kilby.
of prior art, but he did not acknowledge it. Another fact is that Kilby
did not describe his invention as succinctly in his other patents as he
had done in the patent referred to above.
“Since 1901, the Nobel Prize has been honoring men and women from
all corners of the globe for outstanding achievements in physics, chemistry,
medicine, literature, and for work in peace. The foundations for the prize
were laid in 1895 when Alfred Nobel wrote his last will, leaving much of his
wealth to the establishment of the Nobel Prize.”
“On November 27, 1895, Alfred Nobel signed his last will in Paris. When
it was opened and read after his death, the will caused a lot of controversy
both in Sweden and internationally, as Nobel had left much of his wealth
for the establishment of a prize! His family opposed the establishment of
the Nobel Prize, and the prize awarders he named refused to do what he
had requested in his will. It was five years before the first Nobel Prize could
be awarded in 1901.”
Mrs. Sofie Kapy von Kapivar, whose address is known to the Anglo-
Oesterreichische Bank in Vienna, is hereby entitled to an annuity of 6000
Florins Ö.W. which is paid to her by the said Bank, and to this end I have
deposited in this Bank the amount of 150,000 Fl. in Hungarian State Bonds;
January 20, 2009 9:59 B732 9in x 6in b732-app07 FA
the greatest benefit on mankind. The said interest shall be divided into
five equal parts, which shall be apportioned as follows: one part to the
person who shall have made the most important discovery or invention
within the field of physics; one part to the person who shall have made
the most important chemical discovery or improvement; one part to the
person who shall have made the most important discovery within the
domain of physiology or medicine; one part to the person who shall have
produced in the field of literature the most outstanding work in an ideal
direction; and one part to the person who shall have done the most or the
best work for fraternity between nations, for the abolition or reduction of
standing armies and for the holding and promotion of peace congresses.
The prizes for physics and chemistry shall be awarded by the Swedish
Academy of Sciences; that for physiological or medical work by the Caroline
Institute in Stockholm; that for literature by the Academy in Stockholm,
and that for champions of peace by a committee of five persons to be
elected by the Norwegian Storting. It is my express wish that in awarding
the prizes no consideration whatever shall be given to the nationality of the
candidates, but that the most worthy shall receive the prize, whether he be
a Scandinavian or not.
This Will and Testament is up to now the only one valid, and revokes
all my previous testamentary dispositions, should any such exist after my
death.
That Mr. Alfred Bernhard Nobel, being of sound mind, has of his own
free will declared the above to be his last Will and Testament, and that
he has signed the same, we have, in his presence and the presence of each
other, hereunto subscribed our names as witnesses:
Sigurd Ehrenborg
former Lieutenant
Paris: 84 Boulevard Haussmann
R. W. Strehlenert
Civil Engineer
4, Passage Caroline
Thos Nordenfelt
Constructor
8, Rue Auber, Paris
Leonard Hwass
Civil Engineer
4, Passage Caroline”
form of prizes to those who, during the preceding year, shall have conferred
the greatest benefit on mankind. The said interest shall be divided into
five equal parts, which shall be apportioned as follows: one part to the
person who shall have made the most important discovery or invention
within the field of physics; one part to the person who shall have made the
most important chemical discovery or improvement; one part to the person
who shall have made the most important discovery within the domain of
physiology or medicine; one part to the person who shall have produced
in the field of literature the most outstanding work in an ideal direction;
and one part to the person who shall have done the most or the best work
for fraternity between nations, for the abolition or reduction of standing
armies and for the holding and promotion of peace congresses. The prizes
for physics and chemistry shall be awarded by the Swedish Academy of
Sciences; that for physiology or medical works by the Karolinska Institute
in Stockholm; that for literature by the Academy in Stockholm, and that
for champions of peace by a committee of five persons to be elected
by the Norwegian Storting. It is my express wish that in awarding the
prizes no consideration be given to the nationality of the candidates, but
that the most worthy shall receive the prize, whether he be Scandinavian
or not.”
3.1. Physics.
3.2. Chemistry.
3.3. Medicine.
3.4. Literature.
3.5. Peace.
3.6. Economics.
Thus, in accordance with Nobel’s Will, the Nobel Prizes shall be given
annually in the above 6 fields to person(s) irrespective of nationality who,
during the preceding year, shall have conferred the greatest benefit on
mankind.
For whatever it is worth, humbly I wish to state that I have had the
distinguished privilege and the great honor to have met personally many
of the Nobel Laureates in my career. The first whom I had met in 1950
was Sir C.V. Raman who had won the Nobel Prize in Physics in 1930. My
meetings with him and several others were very interesting indeed. Since
they are not the subject of this book, I shall not discuss them.
I shall copy from the Nobel website ALL the Nobel Awards in Physics
given from the very beginning in 1901 to 2007 in Table App. 7.1. As the
readers will note, the number of awardees in each year of the Nobel award
have ranged from one to at most three persons. The sequence of the names of
January 20, 2009 9:59 B732 9in x 6in b732-app07 FA
The following entire list of the Nobel Awards in Physics from the very
beginning in 1901 to 2007 has been copied verbatim from the original Nobel
website.
“The Nobel Prize in Physics has been awarded to 180 individuals since
1901. (John Bardeen was awarded the prize in both 1956 and 1972.)
• 1916 — The prize money was allocated to the Special Fund of this
prize section
• 1915 — William Bragg, Lawrence Bragg
• 1914 — Max von Laue
• 1913 — Heike Kamerlingh Onnes
• 1912 — Gustaf Dalén
• 1911 — Wilhelm Wien
• 1910 — Johannes Diderik van der Waals
• 1909 — Guglielmo Marconi, Ferdinand Braun
• 1908 — Gabriel Lippmann
• 1907 — Albert A. Michelson
• 1906 — J.J. Thomson
• 1905 — Philipp Lenard
• 1904 — Lord Rayleigh
1903 — Henri Becquerel, Pierre Curie, Marie Curie
1902 — Hendrik A. Lorentz, Pieter Zeeman
• 1901 — Wilhelm Conrad Röntgen
Table App. 7.2. The entire list of all Nobel Prize in Physics winners to 3-person
awardees from the very beginning with their citations and distribution of award money.
Nobel Prize, the list will be obvious to most of the readers to understand
what the awards were for in each year. Therefore for brevity, I shall give only
overall comments on the 3-person awardees below. However, I shall discuss
the details of only the Nobel Prize awarded in Physics in 2000 to Alferov,
Kroemer and Kilby in the next Section no. 9, since it is the subject of my
book. First, my overall comments on 3-person awardees of Nobel Prizes are
as follows.
8.2. The prize distribution among the 3-person awardees with a single
subject citation is usually equal, i.e., 1/3, 1/3, 1/3 of the total amount. But
here again, evaluations of the importance and seniority adjudged by the
Nobel Committee including influence peddling as described in Section 8.1
above, the prize money distribution can be unequal, i.e., 1/2, 1/4, 1/4, and
its order can be reversed.
For the sake of continuity, I shall repeat that the Nobel Prize in
Physics was awarded in 2000 to Alferov, Kroemer and Kilby, listed in
this order. The citations for their Nobel award read as “for basic work on
information and communication technology”; the part of the citation listed
January 20, 2009 9:59 B732 9in x 6in b732-app07 FA
first specific to Alferov and Kroemer read as, “for developing semiconductor
heterostructures used in high-speed- and opto-electronics”; the part of the
citation listed next which was specific to Kilby read as, “for his part in the
invention of the integrated circuit”. The Nobel Prize money was distributed
as 1/4 each to Alferov and Kroemer, and 1/2 to Kilby. For additional
discussions, see Chapters 1 and 12.
The Nobel Committee members were either not aware or had ignored
the key facts given in Section 3.2 of Chapter 1 when they had included
Kilby as a co-winner of the Nobel Prize7 in Physics in 2000. There was
hardly any physics involved in Kilby’s invention of the IC. If any field
to which Kilby’s invention could have been ascribed to, would have been
Engineering. But as discussed above in section 4, Engineering is not one of
the designated fields in Nobel’s Will in which the award could have been
given. The citation specific to Kilby’s award in the Prize, “for his part
in the invention of the integrated circuit”, did not state precisely what
was Kilby’s part in the invention of which kind of integrated circuit? It
was incomplete and inconsistent with his contribution to the purported
invention of monolithic-ICs for which he was given the Nobel award. I was
the first to document in my paper12 that Kilby’s invention was not for the
monolithic-IC (see also Chapter 7). This key fact had been neglected in the
entire literature until my paper12 was published. A member of the Nobel
Committee (Dr. MNC-1) in his written communications with me confirmed
recently that I was correct, i.e., Kilby indeed did not invent the monolithic-
IC. Despite acknowledging this crucial fact, Dr. MNC-1 refused to clarify
why then Kilby was chosen to receive the Nobel Award in 2000 if he did
not invent the monolithic-IC, and why the language of the citation for his
award was incomplete and inconsistent with the award?
Another fact that Kilby was given twice the amount of financial award
than to each of the other two co-recipients (Alferov and Kroemer) whose
fundamental contributions did involve physics, struck me as unusual. This
unequal financial award to Kilby enhanced the odd feeling, in particular
when I had remembered that equal financial amounts had been given to
Shockley, Bardeen and Brattain for their invention of the transistor when
they were awarded the Nobel Prize in Physics in 1956. The refusal of
Dr. MNC-1 to clarify this also added to the mystery of the Nobel Award
during my written communications with him. Referring to the original Will
January 20, 2009 9:59 B732 9in x 6in b732-app07 FA
of Alfred Nobel that no explanations can be given until after 50 years of the
Award, Dr. MNC-1 gave it as the reason for his inability to explain all this
now. This reply did not satisfy me, because I did not find this restrictive
clause of 50 years in Nobel’s Will. Similar and additional comments were
also made by another member of the Nobel Committee, Dr. MNC-2 in
his written communications with me. Therefore I did personal research
into this matter using the Nobel website itself as the key source of my
documented information, which resulted finally in this Appendix 7. I came
to the conclusion that the imprecise citation of the Award to Kilby cannot
still be explained, but perhaps I can offer explanations for the sequence of
non-alphabetical listing of the names and the unequal financial Award as
follows.
As stated above and to repeat, the citation of the Nobel Prize in Physics
in 2000 common to all three, viz, Alferov, Kroemer and Kilby was, “for
basic work on information and communication technology”. The part of
the sub-citation specific to Alferov and Kroemer read as “for developing
semiconductor heterostructures used in high-speed- and opto-electronics”.
The part of the sub-citation specific to Kilby read as “for his part in the
invention of the integrated circuit”. What was Kilby’s part to invent what
kind of the integrated circuit was never spelled out? Kilby had invented at
best only a hybrid-IC whose use is very small as compared to the use of Si
monolithic-ICs invented a la Noyce.
I assume that all the members of the Nobel Committee from the early
years, even before Dr. MNC-1 and Dr. MNC-2 had come on board, would
have known the above facts. Even though the Nobel Committee has not,
and will not divulge the reasons at this time for giving the award the way
they did in 2000, my interpretation is that the whole world had come to
the realization that a Nobel Prize was long overdue for recognizing the
invention of the IC. The Nobel Committee apparently could not focus on a
clear winner for this invention for a very long time, and unfortunately Bob
Noyce had died in 1990. Most probably the Committee was also aware of
the key contributions from some of the others to Noyce’s invention. But the
Committee was just unable to come to a clear cut decision in 1980s–1990s.
9.1. The subject chosen for the Nobel Prize in physics in 2000 shall
be “the information and communication technology” which has “conferred
the greatest benefit on mankind” at least from and “during the preceding
year”, the criteria specified in Nobel’s Will. The name of the subject chosen,
viz., information and communication technology (ICT) for the Nobel Prize
clearly identifies it to be in Engineering rather than in Physics. But the
Committee made an ad hoc decision for awarding the Nobel Prize in 2000
to assign ICT to Physics.
9.3. The work of Alferov and Kroemer had made basic contributions to
physics. Hence it will be considered more important than the IC invention
by Kilby which did not make basic contribution to physics. Therefore
they shall be listed first in the sequence of the names jointly. Alferov
and Kroemer will be cited jointly and listed alphabetically, i.e., Alferov
first and Kroemer second. Kilby’s citation and name will be listed after
Alferov and Kroemer (as explained above in 9.2).
9.4. From the financial aspect of the Nobel Award, even though the
impact of the invention of IC is orders of magnitude larger than the
semiconductor heterostructures, and has “conferred the greatest benefit on
mankind” at least from and “during the preceding year”, the award money
shall be divided equally between the two fields being recognized in the
Nobel Prize in Physics in 2000. This will reflect the unbiased and egalitarian
principles of the Nobel Committee by giving 1/2 the prize money to each
field. This meant that since there were two winners in the semiconductor
heterostructures field, they shall divide equally and Alferov and Kroemer
shall each receive 1/4 of the prize money. Since Kilby is the only person
being recognized in the IC invention, he will receive the 1/2 of the prize
money.
Thus, the items 9.1 to 9.4 above explain the most likely scenario that
may have occurred during the difficult deliberations of the Nobel Committee
to arrive at a consensus decision for the Nobel Prize in Physics in 2000. Why
the citation specific to Kilby’s award was worded as given above, because it
was inconsistent and incomplete to clearly characterize Kilby’s IC invention
purported to be of monolithic-IC, is inexplicable indeed. The “50 year”
clause invoked by the Nobel Committee for not divulging its deliberations
for 50 years for an award does not seem to exist in Nobel’s original Will.
Since it has been invoked by Dr. MNC-1, it is fair to say that neither the
Committee members, nor most of the readers, and for sure I shall be long
gone to receive a satisfactory answer for Kilby’s inclusion and the wording
of citation in his award. Hopefully the younger generations of that era in
January 20, 2009 9:59 B732 9in x 6in b732-app07 FA
the future will be able to compare notes with this book, if it survives its
existence until then.
10. Conclusion
10.1. The Nobel Committee is a closed society which does not feel
obligated to offer any explanation to anybody else for their decisions and
actions.
10.2. The details of the nomination process can be obtained from the
Nobel website. Even though the awardees may be eminently qualified to
receive the Nobel Prize, the overriding influence in the selection process
comes also from the campaigning and the votes of the previous Nobel
Laureates and other senior people in the field(s).
10.5. I shall not offer the critique in detail on whether the majority of
the awardees did or did not meet the original criterion for the Nobel award.
Nevertheless I shall offer my humble opinion as follows. In majority of the
cases, the selection of the awardees and their respective Nobel citations
have not met the original criterion explicitly stated in Nobel’s Will. What
to say about the work of Nobel Prize winners to have “conferred the greatest
benefit on mankind”, most of the work is so restrictive and esoteric that it is
hard to judge that each awardees’ work has “conferred the greatest benefit
even to the field of physics”. The excitement generated by a new discovery
is quite short lived especially in the recent years, sometimes akin as an
example to the half-life of the new elementary particles being discovered by
the ever more powerful accelerators. What is their utility to have “conferred
the greatest benefit on mankind”? Even their benefit to physics is not
accepted universally. Nevertheless it is fair to say that the Nobel Committee
has done a reasonably good job to select the awardees some of the time, if
we keep the quintessential criterion of Nobel’s Will in the forefront.
than facts. Sometimes irreversible errors of judgment are made, and the
Nobel Prize is awarded to a less worthy person, and/or not awarded to an
eminently qualified person.
References
4. G. W. A. Dummer:
495
February 2, 2009 9:53 B732 9in x 6in b732-refs FA
15. George Rostky, “The 30th Anniversary of the Integrated Circuit”, VLSI
Systems Design, CMP Publications Inc., Vol. IX, No. 9A, September,
1988.
17. Leslie Berlin, “The Man Behind the Microchip: Robert Noyce and the
Invention of Silicon Valley”, p. 97–127; p. 141; Oxford University Press,
ISBN–13: 978-0-19-516343-8, 2005.
26. K. Lehovec
49. Marvin J. Moss, “Present for the Birth of the Integrated Circuit”, IEEE
Life Member Newsletter, p. 4, April, 2007.
52. Homer O. Blair, “Famous United States Patents”, Pierce Law Center,
2 White Street, Concord, NH 03.01; www.PierceLaw.edu; also personal
communication (2006).
59. Tekla S. Perry, “Gordon Moore’s Next Act”, IEEE Spectrum, 45, 40–43
(2008).
60. Sally Adee, “Thirty-seven Years of Moore’s Law”, IEEE Spectrum, 45,
56 (2008).
62. Arjun Saxena: ”Law of the Famous” - The famous are given most if
not all the credit, and a large number of others who also made key
contributions to the success are often ignored.
63. Chih-Tang Sah and Bin B. Jie, monthly articles on the bipolar theory
of the nanometer field-effect transistor, with the double-gate thin-base
MOSFET or FinFET; complete theoretical analyses reported in the
monthly issues of the Chinese Journal of Semiconductors, from October
2007 to December 2008; further articles to continue to be published.
71. Joel N. Shurkin, “Broken Genius: The rise and fall of William Shockley,
creator of the electronic age,” Macmillan, N. Y. (2008).
List of Tables
8. Table 10.1: List of patents of Jack Kilby on and beyond the 314
invention of ICs.
9. Table 10.2: List of patents of Bob Noyce on and beyond the 318
invention of ICs.
10. Table 13.1: Total world market of all Si products including 344
the ICs.
13. Table App. 2.2: Documents and patents of Saxena on and 400
beyond the invention of ICs.
14. Table App. 7.1: The entire list of the Nobel Awards in 477
Physics from the very beginning in 1901 to 2007.
15. Table App. 7.2: The entire list of all Nobel Prize in Physics 481
winners to 3-person awardees from the very beginning with
their citations and distribution of award money.
February 2, 2009 9:53 B732 9in x 6in b732-refs FA
List of Figures
7. Fig. 5.1: Figs. 1-4 copied from Kilby’s1 patent no. 3,138,744. 86
Note the mesa structures in Kilby’s Figs. 3 & 4 (instead of
the planar structures which are necessary in
monolithic-ICs).
10. Fig. 5.4: Figs. 3, 4 & 5 copied from Noyce’s2 patent no. 89
2,981,877. Note the planar structures and monolithic
interconnects in Noyce’s Figs. 3 and 4 (instead of the mesa
structures and wire-bonding). Fig. 5 shows the circuit
diagram of Noyce’s invention.
13. Fig. 13.1: A semi-log plot of transistors per die (chip) versus 343
time for the early data of 1965 until the recent years and
beyond. The linear behavior according to Moore’s Law is
followed in various segments which were influenced by the
associated technologies used for the chips.
10. Copy of the first page of the final hearing on March 16, 124
1966, of Kilby vs. Lehovec. See reference 8 of Lehovec’s
paper,26 given as “Decision of the Board of Interferences in
the patent interference Kilby vs. Lehovec,” No. 93612, April
5, 1966. [Given at the end of Chapter 6.]
15. Jay Last,33 “Solid State Circuitry Having Discrete Regions 305
of Semi-Conductor Material Isolated by an Insulating
Material”, US Patent no. 3,158,788, filed Aug. 15, 1960, Ser.
No. 49,717, issued Nov. 24, 1964. [Given at the end of
Chapter 9.]
17. Marvin J. Moss,49 “Present for the Birth of the Integrated 332
Circuit”, IEEE Life Member Newsletter, p. 4, April, 2007.
[Given at the end of Chapter 11.]
Index
513
January 20, 2009 10:0 B732 9in x 6in b732-Index 2nd reading
Index 515
Index 517
Additional approval to publish by Dr. Kurt Lehovec was obtained from him
on October 28, 2006.
Copy of the first page of the final hearing on March 16, 1966, of Kilby vs.
Lehovec. See reference 8 of Lehovec’s paper26 , given as “Decision of the
Board of Interferences in the patent interference Kilby vs. Lehovec,” No.
93612, April 5, 1966.
519
January 20, 2009 10:0 B732 9in x 6in b732-ack 2nd Reading
“
c 2007 IEEE. Reprinted, with permission, G. W. A. Dummer, “Integrated
Electronics Development in the United Kingdom and Western Europe”,
Proc. IEEE, P. 1412–1425, December, 1964.
2.1 Copy of the original article14 by Jeffrey Marque, “Getting History Right
is an Important Matter”, F O R U M O N P H Y S I C S &
S O C I E T Y of The American Physical Society, Vol. 36, No. 3, July
2007.
2.2 Copy of Marvin J. Moss49 , “Present for the Birth of the Integrated
Circuit”, IEEE Life Member Newsletter, p. 4, April, 2007.
“c 2007 IEEE. Reprinted, with permission, Marvin J. Moss, “Present for
the Birth of the Integrated Circuit”, IEEE Life Member Newsletter, p. 4,
April, 2007.
2007
c IEEE. Reprinted, with permission, Arjun N. Saxena, “Fundamentals
of the Invention and Impact on Future Developments of Integrated
Circuits and Nano-optoelectronic Devices” — Proceedings of IEEE EDS
Mini-Colloquium NADE-Nanoelectronic Devices-Present and Perspectives-
Sinaia, Romania, 14th of October 2007.
“
c 2003, with the permission from the IETE ‘Golden Jubilee Compendium:
Evolution and Perspective — Electronics, Telecommunications, IT and
Broadcasting’ by The Institution of Electronics and Telecommunications
Engineers (IETE) bearing ISBN 81-901477-2-2; p. 23–31 (2003).”
January 20, 2009 10:0 B732 9in x 6in b732-ack 2nd Reading
523