The Story of Semiconductors, John Orton, c. 2004

The Story of Semiconductors, John Orton, c. 2004.

This page is all messed up because I'm taking notes on the book at various times, various places. I doubt I will ever have time to clean up this page.

Impossible to take exhaustive notes on this book, so may just note some "cocktail chatter" trivia.  

Chapter 1: Perspectives

1.1: The "Information Age"

1.2. Early materials technology -- no fewer than 14 Nobel prizes for individuals in each of the following:

• quick overview of the pace of development of materials
• wireless telegraphy
• semiconductor rectification
• transistor
• tunneling in semiconductors
• discoveries in amorphous semicoductors
• the quantum Hall effect in a metal oxide silicon structure
• work on the fractional quantum Hall effect in a gallium arsenid "low dimensional structure"
• various contributions to the fields of electronics and optoelectronics
• time periods and heat:
• 4000 BC: copper, 1000°; jewelry; possibly
• 3000 BC: copper --> artificial form of lapis lazuli Egyptian facience;
• then, 1100°
• 2000 BC: tin to copper --> bronze; 1000 BC:
• Assyrians -- three major improvements with copper
• 1200 BC: end of Bronze Age --> Iron Age
• 
   

1.3. What makes a semiconductor?
slices from a single crystal "boule"
impurities
steel: 1% (one part in a hundred; one part in 10^2)
crystal perfection: one part in a billion (one part in 10^9)
resistivities measured in ohms (Ω -- option-z)
most metals: 10^-7 to 10^-8
insulating materials such as certain oxide films, mica, glass, plastics, etc: 10^10 to 10^14
range: 10^-8 to 10^14 = 8+14 = 22 zeros
semiconductors, that we're most interest in: also, quite a range
silicon: 10^-6 - 10^2
range: -6 to +2 = 8
semi-cinsulating gallium arsenide: 10^7 -- extends the range upwards by a further five orders of magnitude
so, semiconductors are key, but note the range, even of semiconductors: 13 orders of magnitude (-6 to +7 =13)

band theory (page 7): 
an important application of the recently developed (and highly exciting) quantum theory of atomic structure
first major success of quantum theory was its explanation of atomic spectra, particularly that of the simplest atom, hydrogen
electrons could only occupy well-defined energy states -- again, page 7; defined by this equation:
hv - E2 - E1
v: frequency of the light emitted

λ = 1.240/ΔE
λ: wavelength measured in microns (10^-6 M)
ΔE = energy difference in electron volts

n

p

 

1.4 Semiconductor doping

1.5 How many semiconductors are there? Many:
C, SiC, Si, GE, AIN, GaN, InN
A1P, GaN, InN
A1P, GaP, InP
A1As, GaAs, InAs
A1Sb, GaSb, InSb
ZnO, CdO
ZnS, CdS
AnSe, CDSe
ZnTe, CdTe

Band gaps:
C: 5.5 -- widest gap
SiC: 2.86
Si: 1.12
Ge: 0.664 -- narrowest gap

Aluminum nitride (AlN) is a wide bandgap semiconductor with unique properties making it suitable for high-power, high-temperature electronic devices and other applications. It's considered a "third-generation" semiconductor, following SiC and GaN. While currently used in some specialized applications, full commercialization of AlN-based devices is still some years away, potentially around the 2030s.

Chapter 2: The cat's whiskers

The cat whisker:
a fine wire was nicknamed a cat's whisker because it looked like one
the fine wire (steel or phosphor bronze) delicately touched a crystal -- typically galena -- a naturally occuring lead sulfide semiconductor
the contact point formed a p-n junction, which allowed current to flow in only one direction, like a modern diode
it detected radio signals by rectifying the tine AC radio signal into a detectable DC signal for headphones
this was before vacuum tubes became widespread, and long before transistors
first use of semiconductors -- demonstarted rectification with natural semiconductors like galena
foundation for diodes: direct precursor to modern silicon diodes and point-contact transistors
experimental roots: users had to "hunt" for a sensitive spot on the crystal -- making it an early form of tuning semiconductors

Greenleaf Whittier Pickard (not Packard): the person credited with one fo the earliest cat's whisker detectors in 1906.

Know what I find most amazing: galena was a natural substance
galena: it's a mineral -- specifically lead sulfide; the most important ore of lead
can also be a significant source of silver
widely distributed and commonly found in various geological settings
it's almost as if GOD did this to help mankind along!

2.1 Early days

2.2 First applications

2.3. Commercial semiconductor rectifiers

• rectifiers:
• AC (the grid) --> DC (end product) -- essential
• current can only run one direction
• one of the oldest but most essential applications of semiconductors
• uses p-n junctions -- fundamental
• enables modern energy-saving devices

From the book:

Section 2.3. Commercial semiconductor rectifiers

Then, re-read this section, the importance and development of rectifiers. And comparison of various rectifiers. Again, engineers were flying blind, not understanding why one semiconductor performed better than another. A lot of trial and error.

Page 25: historical developments in physics with regard to semiconductor research. Again, a lot of basic research and a lot of applied research.

Still tracking history of thermionic valve. A thermionic valve is known in the US as a vacuum tube.
 

2.4. Early semiconductor physics.

2.5 The cat's whisker reborn

2.6. Postcript -- how things happen.

Again, this will be an enjoyable “read.” 

Chapter 3: Minority rule

*************************
From chapter 3:

3.1. The transistor .

Vannevar Bush, Chairman of the National Defense Research Committee, in 1945, President Truman. Critical to the beginning of large scale government support for basic science

Bell Labs already ahead of the government. Bell Labs, 1943, Mervin Kelly, Director of Research — emphasized importance of semiconductor research for Bell; resulted in a strong solid state group in the Murray Hill laboratory. Headed by William Shockley, a physicist, and Stanley Morgan, a chemist, contained an important semiconductor group continuing not only physicists but also circuit engineers and chemists, both experimentalists and theoreticians, a truly interdisciplinary team such as had been successful I wartime research projects. It included two future Nobel Prize winners, John Bardeen and Walter Brattain …. rhe very frontiers of semiconductor understanding.

Bell Labs wanted their research to be valuable to everyone … licensed their inventions … result …. the integrated circuit was invented NOT by Bell but by a small new company known as Texas Instruments.

Next: understanding the requirement for purity; minimizing impurities in semiconductor material.

Ge was initially preferred to Si was due to melting points (937°C vs 1412°C).


Again, William Shockley mentioned: the idea of a solid state amplifier to rival the triode valve.

Shockley had an obsession with the idea for what later became known as a “field effect transistor” (FET), a device in which the conductivity of a control voltage to a “field electrode” or “gate” placed alongside it. The “gate” could be likened to the grid electron tin the triode valve in that it controlled the flow of current through the device without any significant amount of power being dissipated in the gate circuit….. and continue on page 51.

Minority carrier injection: p. 51.

Long full paragraph, p. 51. Why the Bell group was well ahead of the rest of the world in terms of their basic understanding: it was Bardeen’s theory of “surface states.”  Huge frustration for Shockley. Couldn’t figure out why his transistor didn’t work.

P. 52: the importance of serendipity by Brattain — first paragraph, p. 52.

Second paragraph, p. 52 —- serendipity again … Brattain — even more bizarre than Brattain’s first serendipitous breakthrough.

The story of all this in the last paragraph, page 52, and first paragraph, page 53, and finally, on Christmas Eve, December 24, 1947, the “transistor” was finally born — though the name was invented somewhat later, when Bell unveiled the new device to a still uncomprehending world.

Continue story on page 53, again, the first long paragraph.

So, again, basic research ... and then voila, using Ge, things came together. Things came together so well, the recognition of this new gain mechanism led to the immediate renamig of the various contacts as "emitter," "base," and "collector."

But look at this, again serendipity. This all came together because they had selected Ge to use as the semiconductor, ad not because of its electrical / semiconductor properties which were perfect, but not yet know, but because of the melting point of Ge -- p. 53. The author uses the word fortuitous. Perhaps he did not want to use serendipity for the third time. The suitability of Ge had to do with it s rather small band gap of this material (0.66 eV) -- but again, it was the melting point that led these researchers to use Ge. 

The melting point of Germanium (Ge) is 938.25 °C (1720.85 °F). It is a metalloid with properties between silicon and tin.

At the top of page 54, the author says, "the whole process appears to have owed as much to serendipity as to foresight, I should emphasize that  the very essence of good research frequently comes down to just this -- the ability to ride one's lucky breaks and capitalize on them is paramount -- the invention of the transistor was a wonderful example and surely no one (least of all the present author) could possibly wish to belittle it." The author continues with great writing, page 54. 

The importance of all this was reflected in an equally important book by Shockley, Electrons and Holes in Semiconductors which explained what they they did and the significant of what they did and the science of what they discovered. The book came out just three years later, 1950. Remember, the transistor was discovered Christmas Eve in the year of so many great political events, 1947, not least of which, the seeds for the creation of Israel: 

The State of Israel was established on May 14, 1948, when David Ben-Gurion proclaimed its independence. This declaration fulfilled the long-standing Zionist goal of creating a Jewish homeland in Palestine. The United Nations had previously recommended the partition of Palestine into separate Jewish and Arab states in 1947.

Transcribe 2nd paragraph, page 47.

From the book:

3.1 The transistor

If the author had to choose a single event which truly put the semiconductors on the international map, it would surely be the invention of the transistor at Bell Telephone Laboratories in late 1947. Without this dramatic step, the leap into “information technology” which has so radically changes all orulives may never have occurred. The successful development of the transistor can be seen, to only as an enabling technology in its own right but an once-and-for-all justification for the huge financial investment in semi-conductor research which was a noteworthy feature of the second half of the twentieth century.

The story of the semiconductor — semiconductors exactly contemporary with my life — I was born in 1951. When the author says “a noteworthy feature of the second half of the twentieth century” it exactly corresponds with the fifty years of my life. Exactly.

This chapter, then examines in detail, what happened in the second half of the 20th century to move the semiconductor story along.

Starts with the need to improve upon the thermionic valve if ATT were to remain the leader in global communications.

So, before we go further, we need to go back and explore the therm - ionic valve.

The therm-ionic valve.

p. 23: the crystal rectifier made commercial radio communication possible but it failed in terms of convenience and reliability.  The thermionic valve can trace its origins to 1883 when Thomas Edison took out a patent for a vacuum diode but it was not until 1904 that Fleming began experiments to employ a similar device as a radio detector (a radio detector: a gizmo to detect radio waves).

The thermionic valve had superior stability and reproducibility which soon make it first choice for its application as a radio detector and the crystal rectifier rapidly faded from the picture — though, as shall see, it enjoyed a resurgence as a radar detector in the early years of WWII.

The detector diode was soon followed in 1906 by the triode valve, the ‘audion’ invented by Lee de Forest which, because of its ability to amplify electronic signals, completely revolutionized radio technology.

A few years later, in 1912, the device was taken up by Bell laboratories and in the following year Bell demonstrated its use in repeater stations which formed. Vital part of their first long distance telephone transmission.

At this point, semiconductors probably seemed totally redundant, though this was, as we know, only a temporary setback. Valve technology had certainly won an important battle but semiconductors, with the invention of the transistor in 1947, were surely set to win the war. In fact, even before this, semiconductor rectifiers still had an important part to play, as we shall see in the next section.

*********************

From the book:

p. 39: early WWII; importance of radar; pressure of war; organized science developed.

The concept of radar — could be long range but resolution lousy; needed microwave techniques to get better resolution of targets

Two major improvements were essential:
the development of a high power microwave source and a sensitive and reliable detecting device.

The cavity magnetron was to satisfy the former and our (now discredited) old friend, the cat’s whisker rectifier, the latter.  The author then turns his attention to the rejuvenation of the cat’s whisker.

Re-read this section beginning on second half of page 39.

Lots of great stuff, needs to be re-read.

3.2. Ge and Si technology. Page 54.
Ge and Si seemed the perfect semiconductors with which to start
purity: foremost; a good quality single crystal
so, next two problems to be solved: "a single crystal" and "purity."

How they solved this. Page 56. 

"Zone refining" -- to achieve purity;
"Crystal pulling -- to achieve the desired structural quality

"fractional crystallization" -- only one end of the ingot, only a fraction of the ingot is satisfactorily "pure"

Last paragraph, page 56, again tells of the was the broth was slowly frozen from the top downwards in an effort to produce a single crystal ...

... again, basic research. Some of it seemed to be trial and error.

And this goes on through page 60.

Germanium, atomic number, 32; lustrous, hard-brittle, grayish-white and similar in appearance to silicon; a metalloid and semiconductor; not typically considered a mieral in its pure elemental form
Silicon, atomic number, 14, metalloid

3.3. The physics of Ge and Si, page 60. 

3.4. The junction transistor

made from semicondutor materials (like silicon)
three terminals:
emitter
base
collector (in a bipolar junction transistor (BJT) or source, gate, drain (in a field effect transistor, FET)

controls the flow of current through one pat (collector-to-emitter or drain-to-source) using a small iput as another termial (base or gate). This lets the transistor act as:
an amplifier: small signal in --> bigger signal out
a switch: turning current on or off;

Transistor: transfer + resistor = transistor. When it was invented at Bell Labs in 1947, the term was chosen to reflect how the device transfers an input signal across a resistive load -- essentially describing how it modulates or controls electrical current. Coined by John R. Pierce, one of the Bell Lab engineers, who wanted something catchy and descriptive, much like "resistor" or "condenser" (an old word for capacitor).

************************
Chapter 4: Silicon, Silicon, and yet more Silicon

4.1. Precursor to the revolution. Entire section re-typed (transcribed) and placed here at this blog.

4.2 The metal oxide silicon transistor. It begins:

The Metal Oxide Silicon (MOS) transistor was yet another product of the fertile ground cultivated by Bell Telephone Laboratories and, once again, it involved just a small element of good fortune. The critical step in its evnention was the (accidental!) discovery that the Si surface can be oxidized to forma highly stable insulating film which possesses excellent interface qualities (i.e. the interface between teh oxide layer and the underlying silicon). We have already commented on the important of this interface  in passivating Si planar transistors which, in turn, led to the practical realization of integrated circuits. The further applicatio in the metal oxide silicon field effect transistor (MOSFET) turned out to be singularly important bonus.

[Comment: so, now in the very beginning of the transistor story four discoveries that were serendipitous (serendipity), fortuitous, or accidental. Absolutely amazing.] Passivating:

Passivating is a chemical process that enhances a material's corrosion resistance by creating a protective oxide layer on its surface. It's commonly used on stainless steel to improve its resistance to rust and other forms of corrosion. The process removes free iron and other contaminants from the surface, allowing a passive oxide layer (often chromium oxide) to form.

4.3. Semiconductor technology.

4.4. Wise men from the East.

4.5. Power and energy -- sometimes size is important

4.6. Silicon is good for physics, too.