Lecture 1: Introduction
Gary , PhD
https://www.xjtlu.edu.cn/en/departments/academic-departments/electrical-and-electronic-engineering/staff/chun-zhao
Copyright By PowCoder代写 加微信 powcoder
On all slides of this Module
• All slides were used by the lecturer without owners’ permission ☺. The copyright of these materials which are used for educational purpose, remains as the properties of owners. The lecturer does not claim any ownership.
The Module Leaders:
– Dr. ZHAO
• BEng (Southeast University)
• MSc ( University of Science and Technology)
• PhD (University of Liverpool)
• After PhD ( University of Science and Technology)
• Contact details: Room SC342 (8816-1402)
– Prof. Zhao (Honorary Leader)
• Teaching/Research in Microelectronics > 30 years • VLSI: technology, design and reliability
• Semiconductor integrated optics and solar cells
• Contact details: Room EE516A (8816-1408)
Teaching Assistants
• On the Module
• History of semiconductor devices and ICs
Moore’s law
Moore’s law
– Transistor scaling
Teaching Plan: Part 1
Teaching Plan: Part 2
Assessment System
Distribution
Assignment
(homework, but class check)
Mid-term test
IC Design Project
Final Exam
Assessment System
• Semiconductor fundamentals (Week 1-4)
– Electrons in solids, energy bands, intrinsic and extrinsic silicon
• PN junction and diodes (Week 5)
– IV characteristics, depletion regions, capacitance, and multiplication
• MOS capacitors (Week 6)
– Accumulation, depletion and inversion, capacitance
• Fabrications and layout (Week 9-10)
– Resistors, diodes, bipolar transistors, and MOSFET
• MOSFETs and IC design (Week 8 & 11) – Operation, equation and nMOS IC design
References
• The most important one
– The handouts!
Understand them from cover to cover
– Textbook (
.) – Reference books (see next slide)
No need for reading from cover to cover
An Introduction to
Microelectronics, by C.Z. ZHAO et al
Reference books
• D. A. Neamen (2006). An introduction to semiconductor devices, 清华大学出版社. ISBN 978 7 302 12451 1.
• H. (1998). Devices for Integrated Circuits:
Silicon and III-V Compound Semiconductors, Wiley & So., ISBN 0471171344.
• S. M. Sze (2001). Semiconductor Devices: Physics and Technology, Wiley & So., ISBN 0471333727.
• D. A. Neamen (2003). Semiconductor physics and devices: Basic principles, McGraw-Hill, ISBN 0072321075.
• History of semiconductor devices and ICs • Moore’s law
– Transistor scaling • Yield
• On the module
Why I am so proud of what I am doing everyday: “Semiconductor”? -by
What is the size of a transistor nowadays?
The size of a human hair
(the size of a skin cell)
(in 1990s, the size of a transistor)
2019 nCoV droplet: 1-2 um
The Nanometer Size Scale
Carbon nanotube
– Vacuum tubes were used for radios, television, telephone equipment, and computers
… but they were
expensive, bulky, fragile, and energy- hungry
: Vacuum tube era
: First electronic computer
ENIAC: Electronic Numerical Integrator And Calculator
Real size: 3000 cubic foot
(1 foot =0.3048m) 19
• ENIAC filled an entire room!
17,468 vacuum tubes, 70,000 resistors, and 10,000 capacitors
6,000 manual switches and many blinking lights!
• could add 5,000 numbers in a single second
The first electronic computer
: First transistor
Picture shows a point- contact transistor structure comprising the plate of n-type germanium and two line- contacts of gold supported on a plastic wedge.
W. Shockley,
“The path to the conception of the junction transistor”,
IEEE Tr. on Electron Devices ED-23, 597 (1976).
“A device called a transistor, which has several applications in radio where a vacuum tube ordinarily is employed, was demonstrated for the first time yesterday at Bell Telephone Laboratories, 463 West Street, where it was invented.”
The transistor is the defining
invention of the 20th Century
• First transistor (by Bardeen, Shockley and Brattain 1947): The Transistor Revolution
William Bardeen Walter Brattain
– reproducibility was an issue, however. 24
in Physics
• Invention of the bipolar junction
transistor (BJT)
▪ , Bell Labs, 1950
– more stable and reliable; easier and cheaper to make
• In 1954, Texas Instruments produced the first commercial silicon transistor.
• Discrete Electronic Circuits ~$2.50 each
• Before the invention of the integrated circuit, electronic equipment was composed of discrete components such as transistors, resistors, and capacitors. These components, often simply called “discretes”, were manufactured separately and were wired or soldered together onto circuit boards. Discretes took up a lot of room and were expensive and cumbersome to assemble, so engineers began, in the mid- 1950s, to search for a simpler approach…
: First commercial
1)Discrete Electronic Circuits
– PCB circuits will not be discussed.
2) Integrated Circuits
– this is the module’s issue of a transistor)
(in 1990s, the size
– A British idea: 1952 G. Dummer
– American fabrication: 1958 – No. of components: 12
(1923-2005), received the
in 2000 in Physics
: 1st Integrated Circuit
Integrated Circuit (IC)
• AnICconsistsofinterconnectedelectroniccomponentsin a single piece (“chip”) of semiconductor material.
• In 1958, . Kilby (Texas Instruments) showed that it was possible to fabricate a simple IC in germanium.
• In 1959, ( ) demonstrated an IC made in silicon using SiO2 as the insulator and Al for the metallic interconnects.
The first planar IC
(actual size: 0.06 in. diameter) 29
Eight talent engineers, “traitorous eight”, from the Laboratory in Palo Alto, California
(from left to right), C. , , , , , and .
Corporation, American electronics company that shares credit with Texas Instruments Incorporated for the invention of the integrated circuit. Founded in 1957 in Santa Clara, California, Fairchild was among the earliest firms to successfully manufacture transistors and integrated circuits. 30
Intel: & AMD, National Semiconductor, SDA Systems (Cadence), Xilinx, Atmel…
Sequoia Capital (红杉资本), KPCB-> Apple, Google, Amazon, Alibaba，
今日头条，京东，爱奇艺，饿了么
: First Monolithic IC
Picture shows a flip- flop circuit containing 6 devices, produced in planar technology.
R. N. Neyce, “Semiconductor device-and-lead structure”, U.S.Patent 2,981,877
http://www.computerhistory .org/semiconductor/timeli ne/1960-FirstIC.html
A digital (DTL) IC from 1964
• Images courtesy of . Used without permission. An early bipolar integrated circuit.
D: the Feature Size 2D
“D” represents IC technology level. The names of “D”, nowadays:
“D technology”, or “Generation D”, or “D node”
: first microprocessor
Picture shows a
four-bit microprocessor Intel 4004.
• 10 μm technology
• 3 mm 4 mm
• 2300 MOSFETs
• 108 kHz clock frequency
Intel Corporation
4004 Micro
History: The 1st microprocessor
• Invented by Intel in 1971, 4004 • The basic structure for ‘computing’
ENIAC —1946
Real size: 3000 cubic foot
1 foot =0.3048m
“Honey, I shrank the computer”
4004 — 1971
Real size: Thumbnail
Intel Pentium II
➢3 million transistors 163mm2 chip
0.6 μm technology 8W power consumption
: Pentium II processor
: Pentium IV processor
Picture shows a ULSI chip with 32-bit processor
Intel Pentium 4.
• 0.18μm CMOS technology
• 17.5 mm 19 mm
• 42 000 000 components • 1.6 GHz clock frequency
Intel Corporation
• Pentium 4: Intel—April 2, 2002
– 0.13um; 2.4GHz; 55million transistors
: Pentium IV processor
: Intel processor
Picture shows a ULSI chip with single core processor
• 45nm CMOS technology
Intel Corporation
45 nm feature size
Over 1 billion transistors
8 core processor
• 3.2 GHz clock frequency
Intel shows
processor technology
• 32nm entered full scale production in 2010 and
made 22nm in
2.5 billion transistors
• IC Technology Roadmap -> D Technology
• D=180nm or 130nm or … or 10nm or 7nm or 5nm
• On the module
• History of semiconductor devices and ICs
• Moore’s law
– Transistor scaling
Moore’s Law
: co-founder of Intel. “Density of IC devices is doubling with each new generation” (in 1965)
In 1965, noted that the number of transistors on a chip doubled every 18 to 24 months.
He made a prediction that semiconductor technology will double its effectiveness every 18 months
Die Single die
(single chip)
Intel Pentium®4 Processor
From http://www.amd.com
300mm Si wafer, inch??
Why the success:
• Moore’s Law:
Components (transistor) per chip double every 1.5 – 2 years
Cost per Transistor
Moore’s law in
cost: transistor
0.1 0.01 0.001 0.0001 0.00001 0.000001
Fabrication capital cost per transistor (Moore’s law)
1988 1991 1994
2003 2006 2009
Processing is performed at
The largest silicon wafer
• Not only small chips, but also large wafers
• 12 inch — 300mm
• With 90nm CMOS: 120b per wafer
• On the module
• History of semiconductor devices and ICs
• Moore’s law
– Transistor scaling • Yield
Why Scaling?
• Technology shrinks by
0.7 per generation
• With every generation can integrate 2x more functions per chip; chip cost does not increase significantly
• Cost of a function decreases by 2x
1) Cheaper
(faster and more functional)
Why is small beautiful?
• Price reduction: A million to one
• “Nothing had ever done that for anything before”
1.E+06 1.E+04 1.E+02 1.E+00
1970 1980 1990 2000
, a car can only
— J. Kilby
“If this happens in
auto industry
worth less than a
No. Transistor per $
Not only but also
• Binary: “0” and “1”
• “0”=Empty “1”=Full
– Filling the tank
• Capacitors between gate and silicon
• Capacitor Electronic tank
• Electronic charges Water
Delay = RC
• Smaller devices Smaller tank
Metal Oxide
Benefit of Transistor Scaling
Generation: Intel386TM DX
Intel486TM DX Processor
Pentium® Processor
Pentium® II Processor
0.8μ 0.6μ 0.35μ 0.25μ
smaller chip area → lower cost 2D
more functionality on a chip →better system performance
TSMC President
• 0.13m: 1000 trans. per hair
2001 2003 2005
0.18 um→ 0.13um → 0.09 (90nm) → 65nm →
→ 45nm → 32nm → 22nm → 16nm → 14nm → 7nm
2007 2009 2011 2013 2014 2017
art CMOS process
2020: 5nm -> 2022: ???
• On the module
• History of semiconductor devices and IC
• Moore’s law
– Transistor scaling
Die yield =
Y = No. of good chips per wafer 100%
Total number of chips per wafer
Die cost = Wafer cost
Intel Pentium®4 Processor Dies per wafer Die yield
2 Diesperwafer=(waferdiameter/2) −waferdiameter
die area 2die area
• Each ‘chip’ consists of a large number transistors interconnected to make a circuit. Some of the silicon chips will contain defects such as those marked with D.
• Of 16 circuits in the example, 14 are working, so the yield is 87%. So if this slice were to cost £500 to make, the net cost of each chip would be 500/14=£36.
• But much depends on how small each transistor is, and there is a relationship between this size and the size of the chip.
• Only two chips are working so that the yield is 50% and the cost would be £250.
• The smallest dimension that can be produced is called the minimum feature size(). It has changed from 10m to 0.3m in 25 years a factor of 35. The above example, which corresponds to a minimum feature size reduction of two leads to a reduction of cost from £250 to £36.
Next Lecture
• Atomic Structures • Crystal structure • Energy Bands
• Read: Chapter 1 – 2.1, 2.2 & 2.3 Textbook (
An Introduction to Microelectronics,
by Cezhou ZHAO, ANG and Qifeng
程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com