Semiconductor Culture

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
In This Lecture:
Semiconductor Culture: History, Trends

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
Basic Semiconductor Culture
1900-1940: Incubation period
History of Semiconductor Physics
Discovery of the electron(1899), photon (1905), QM (1925-28)
F. Bloch – Ph.D. Thesis (1928), introduces band theory of metals
R. Peierls (1929) conceives the theory of positive carriers to explain the thermal and
electrical conductiveness of semiconductors.
A.H. Wilson provides a theoretical understanding of se/c based on energy gap,
further clarifying the holes/electrons, the role of impurities
Fermi introduces pseudopotentials (vital for the development of band structure)
se/c devices on the stage: thyristors & photodetectors
1939-1945: WW-II
1940-1945: Post-war disclosure; Microwave detection using se/c, QED, Lamb shift
1947: Bardeen & Brattain demonstrate first bipolar transistor action
1947: Bardeen points the importance of surface states in Schottky contacts
1949: Shockley describes the physics of the p-n junction

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
History of Semiconductor Physics (cont’d)
1950’s: se/c recognized as being versatile mat’ls; new devices follow
1951: Shockley introduces heterojunctions
1952: Shockley introduces junction FET
1954: Townes invents maser after Dumke (1952) se/c laser proposal
1958: Esaki’s tunnel diode (a marriage of QM & se/c physics)
1958: Krömer’s NDR idea based on the effects of bandstructure on charge xport
1960’s: Boom in theory (Keldysh, Kane, Kohn, Phillips, Cohen, Harrison,…)
1962: Stimulated emission in GaAs diodes at GE & IBM labs (20 years more needed for
an industrial maturity)
1963: Gunn oscillations measured by Gunn; theory by Ridley-Watkins (1961)
1970’s: Mainly large-size integration, first microprocessor, RTDs
1970: Esaki & Tsui introduce superlattices & Bloch oscillations
Early 1970’s: First Molecular Beam Epitaxy (MBE) at Bell Lab’s

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
History of Semiconductor Physics (cont’d)
1980’s: Thanks to MBE, clean se/c samples; fundamental discoveries follow
Weak localization and the rise of the mesoscopic physics
Low-dimensional heterostructures (quantum wells & wires)
1980: von Klitzing discovered QHE in pure MOSFETs (at mK); resistance standard
1982: Tsui & Strömer discovered FQHE using very pure GaAs samples
1987: Conductance quantization in quantum point contacts (Cambridge & Delft groups); Coulomb
blockade
1990’s: More advances in growth…
Growth of Group-III Nitrides – Nakamura (1989-93)
Synthesis of Carbon nanotubes – Ijima et al. (1991)
Growth of Dilute Nitrides – Kondow et al.(1994)
Growth of Self assembled quantum dots – Several groups (such as P. Petroff of UCSB)
2000’s: The rise and rise of se/c physics…
Silicon photonics – Pavesi
Se/c quantum electrodynamics
Quantum computing utilizing se/c QDs
Spintronics – Spin Hall Effect experimentally demonstrated by UCSB group in 2004
Molecular electronics & organic se/c
Japanese Experimental
Breakthrough

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
Trends in Semiconductor Research
The period 1940-2000 was under the rule of information
systems (high speed devices...)
In 2000’s this monopoly is being weakened by other needs
such as energy (solar cells, energy storage, etc.) and health
(diagnostic and therapeutic tools)
Some References on the History of Semiconductor Physics
L. Hoddeson and G. Baym, ‘The development of the quntum-mechanical
electron theory of metals: 1928-1933’, Reviews of Moden Physics, 59 287
(1987).
R. Peierls, ‘Early work on solids, mainly thirties’, Reviews of Moden
Physics, 65 251 (1993).
A. B. Fowler, ‘A semicentury of semiconductors’, Physics Today, p. 59
(October, 1993)
J. W. Orton, The story of semiconductors, (Oxford University Press, 2004)
This book is available at our library: TK7871.85.O78 2004

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
How about the Future?
Moore’s Law
Source: J. Singh

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
Se/c industry roadmap
Visit: International Technology Roadmap for Semiconductors
http://www.itrs.org
NB: Main activity around CMOS technology
Severe Challenges ahead
Interconnect bottleneck
Thermal Inferno
In chip communication:
Do it with Photons!
Source: J. Singh

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
Classification of Electronic Materials
Source: J. Singh
E
Semimetals
k
e.g., Bi, Sb, As,
HgTe, HgSe

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
A typical se/c (EPM) bandstructure: GaN & AlN
Conduction
Bands
Fermi
level
Band gap
for GaN
Valence
Bands

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
Periodic Table
Common se/c

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
A Classification of Se/c’s
Elemental Semiconductors
Group-IV: Si, Ge; Diamond structure, tetrahedrally coordinated
Group-V, VI: P, S, Se, Te are also se/c’s with several different crystal
structures. Good glass formers
Binary Semiconductors
III-V compounds are similiar to group IV
IV → III-V ionicity increases. Electronic charge transfer from III to V
atom: Coulomb interaction, changes in electronic band structure
II-VI (ZnS): more ionic. Mostly large bandgaps → displays and lasers.
Exception: HgTe zero bandgap → IR detectors.
I-VII (CuCl): have larger bandgaps. Some are regarded as insulators.
Increased cohesive energy. Rock salt struc.
IV-VI (PbS, PbTe, SnS): semiconductors. Large ionicity. 6-fold
coordination. Very small gaps. → IR detectors
Ref: Yu-Cardona

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
A Classification of Se/c’s (Cont’d)
Oxides
CuO, CuO2 :semiconductors ZnO →transducer
High Tc SC: Copper oxides: La2CuO4 → bandgap 2 eV.Dope with Ba or Sr.
P-type.
Layered Semiconductors
PbI2 , MoS2 , GaSe, GaS: Intraleyer bonding covalent, interlayer bonding van
de Waals.
Quasi 2-D.Intercalation.
Organic Semiconductors
Polyacetylene [(CH2 ) n ], polydiacetylene
LEDs, lasers? Displays. Cheap. Slow. Bandgap manipulation is easier.
Magnetic Semiconductors (gained further importance with spintronics)
Magnetic ions: Mn, Eu etc. EuS, Cd1-xMnxTe: ferromag, antiferromag
possible. Dilute mag se/c’s. Large Faraday rotation → optical modulators.
Others
SbSI: Ferroelectricy at low T,
I-III-VI2 , II-IV-V2 : AgGaS2 and ZnSiP2 → chalcopyrite struc. Tetrahedral
bonding. Analog to III-V and II-VI.
IV-VI with formula such as As2Se3 : se/c’s in crystalline or glassy states
Ref: Yu-Cardona

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
A Classification of Se/c’s (Cont’d)
Ternary and Quaternary Se/c’s
III x -III 1-x -V type ternaries:
AlxGa1-xN, AlxGa1-xAs,AlxIn1-xP, AlxIn1-xSb, etc.
III-V1-x-Vx type ternaries:
AlAs1-xPx , GaAs1-xPx , InSb1 xAsx , InSb1-xBix , etc.
IIIx-III1-x-Vy-V1-y , type Quaternaries:
GaxIn1-xAsyP1-y , GaxIn1-xAsySb1 y , etc.
III 1-x-y IIIx-IIIy-V type Quaternaries:
In 1-x-y Al x Ga y P, In 1-x-y Al x Ga y As, etc.

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
Band gap vs. Photon Wavelength
Surprise: dilute nitrides!
Bandgap not linear interpolation
Source: J. Singh

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
Comparison of se/s
Source: J. Singh

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
Why insist on se/c ?
Physical properties of se/c can be altered drastically by
Doping
Pressure
Electric or magnetic field
Light
Temperature
Response of se/c to external inputs can be tailored in a manner that allows
the devices to implement all necessary information processing operations
Digital & analog signal processing
Oscillators
Detectors
Memories
…
Industrial conservatism; huge investments and acquired knowledge

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
Why insist on se/c ?
Electrons have:
charge → interact strongly
Good for information processing/computation (digital/analog)
But they interact strongly among themselves and the environment, hence they
are prone to noise
mass → they suffer from propagation delays (drift velocity in se/c ~ 10 7 cm/s)
Photons have:
no mass, no charge → very weak interaction
Ideal for signal transmission as they are fast and hardly interact with each other
But, it is also harder to operate on them as in the case of electrons
Semiconductor structures are ideal for hosting both electrons & photons!

C. Bulutay Semiconductor Electronic & Optical Processes
Lecture 1
Demands on se/s
Source: J. Singh