© 1997-2004 R.LevinePage 1
Electronics Review B
EETS8304/TC715-N
SMU/NTU
Lecture Scheduled Jan. 27, 2004
Electronic Devices(print slides only, no notes pages)
© 1997-2004, R.LevinePage 2
Junctions• When two materials are in contact
– In general, some electrons transfers from one material to the other
– Materials with a higher atomic number have more positive charge on the atomic nuclei in the atoms, and thus they attract negative-charged electrons with greater force. Electrons move into that material from the other.
• In a mixture of atoms (an alloy or an almost pure material “doped*” with a small amount of a second material) the average positive atomic charge is used, based on a large number of atoms
– Materials are classified in reference books according to their “electro-negativity” or “contact potential” or “ionization potential” measured in volts
• This affects other situations when electrons leave or enter a piece of material
– Electrodes in electric batteries (flashlight, automobile, etc.)– Photoelectric emission of electrons from metals (“electric eye”)
*“Doping” is alloying using very small amounts of minor materials
© 1997-2004, R.LevinePage 3
“Static” Electricity Example• When you rub two dissimilar objects together and
then separate them quickly:– hard rubbing removes any contamination on
the surface, permitting good contact– electrons transfer to the material with higher
“average” atomic number, producing negative net movable electric charge
– The other material is left with a deficit of electrons and a net positive charge
• Also occurs when you:– quickly break solid objects (e.g., a sugar cube
or mint candy) into pieces– pull adhesive tape from a roll
© 1997-2004, R.LevinePage 4
Safe to Try This at Home!• Take roll of “Scotch” brand or similar sticky tape:
– Wait in a darkened room until the pupils of your eyes accommodate to the darkness
– Rapidly pull about 50 cm (20 inches) of tape off the roll while looking at the point where the adhesive side separates from the layer below it
– You will see a line of electric sparks...Due to electrons which cling to one of the separated materials, and then jump back through the air
– Safe “experiment” to do with/for children!• don’t bump into anything in the dark!• don’t waste too much tape!
pull
Pencil usedas axle
© 1997-2004, R.LevinePage 5
Static Electric Effects• When you brush your hair, rub your shoes on a carpet, rub
a glass rod with fur, etc. etc., you produce so-called “tribo-electricity” (electricity due to rubbing)– If you separate the two dissimilar touching objects
quickly, each object becomes oppositely electrically charged (some extra electrons stay with one object).
– Best done in dry, low atmospheric humidity conditions (winter months, dry climate area, etc.)
• High humidity (water vapor in air) causes surface condensation, producing an electrically conductive surface condition, which allows electric charge to move to other areas and thus neutralize a local charged area
• Anti-static sprays for clothing, etc., produce an electrically conductive surface
• Good conductors (like metals) don’t retain charge at one spot, but spread it over the surface of the entire object
© 1997-2004, R.LevinePage 6
Semiconductors and Insulators
• In a good insulator, surface charge stays put for a very long time
• Semiconductor spot surface charge very slowly moves (diffuses) away– slow movement is due to thermal diffusion (random
motion due to thermal energy) of electrons– electrons are always in some random motion, which we
perceive as motion (kinetic) energy of “heat”– Somewhat like a “neat” pile of leaves eventually
spreading out over the whole lawn due to random motions from changing wind directions, etc.
© 1997-2004, R.LevinePage 7
Controlled Charge Layers• The operation of “active” semi-conductor devices
depends on producing and controlling layers of electric charge
• These usually occur at the interface between two kinds of semi-conductor materials, or between a semi-conductor and a metal conductor
• “Favorite” semiconductor is silicon (Si), which is abundantly available (purified from beach sand SiO2, for example) and which forms an excellent protective layer of SiO2 on the surface of integrated circuits, transistors, etc.
• Other semiconductors are germanium (Ge) which is scarcer, and gallium arsenide (GaAs) 50-50% alloy
© 1997-2004, R.LevinePage 8
Purified Semiconductors• To produce a controlled result, semiconductors are first highly
purified– Typically only one “impure” atom in 100,000,000 silicon atoms!
• A thick silicon rod is “zone refined”– Melted and then cooled slowly, starting from one end, to form a very pure
solid
• This “zone refining” process is similar to freezing pure water ice out of salty ocean water
– Icebergs near the earth’s poles consist of pure water (no salt) – “silicon ice” (solid) which is slowly frozen from melted silicon is very pure
• The impurities are mostly trapped in the end of the rod which solidifies last
• That end is cut off and used for other purposes where the silicon does not need to be so pure
– Example: making “Varistors” for telephone sets (discussed in EETS8302)
• Purified silicon (or a Group III-V alloy – described p.9) is then “doped” to produce a slight (1 part in 106) fraction of nuclei with either higher or lower electric charge than the average nuclear charge
© 1997-2004, R.LevinePage 9
Materials Used• Most semiconductor materials are in Group 4(a) of the
Mendeleyev Periodic Table of the elements– Doping materials are taken from Groups 3a and 5a
• Similar atomic size and electron bonding• Fits into the crystal structure of the solid semiconductor
• In some cases a 50-50 mixture (alloy) of materials from Groups 3a and 5a is the base material– Called III-V (Roman numerals 3-5) materials– Gallium Arsenide (GaAs) is used extensively because of
higher electron mobility (electron waves move further-- on average --before interacting with nuclei). Consequently transistors have better high frequency or fast switching performance
– Doping achieved by using slightly more/less than 50% of the Group 3 or 5 material
© 1997-2004, R.LevinePage 10
Periodic Table (in part)
Yellow-highlighted names are elements used in practical room-temperature semiconductor devices.
• Chemical abbreviation names are underlined.• C and Sn have multiple crystal structures, only one of
which (diamond structure) is a semiconductor• Elements in groups 3, 5 are used as dopants• Germanium is used only rarely for special applications.
Group 3a Group 4a Group 5aBoron Carbon Nitrogen
Aluminum Silicon Phosphorus
Gallium Germanium Arsenic (As)Indium Tin (Sn) Antimony (Sb)
© 1997-2004, R.LevinePage 11
Alloying or Doping• When Group 5 material is added, the average atomic number is
higher. This is called N (negative) type material– The average nuclear positive charge per unit volume is greater than
“intrinsic” (pure) silicon, but there are also more electrons as well– Of course, a piece of material as a whole is electrically neutral – When Group 3 material is added, the average atomic number is lower.
This is called P (positive) type material– The average nuclear positive charge per unit volume is less than
“intrinsic” (pure) silicon
• A semiconductor diode is made by joining two pieces of silicon: P and N material respectively, and outer electrodes– By welding two pieces in historically early transistors– Depositing built-up layers from vapor in a vacuum chamber– Implanting ions from the surface using an electric “ion gun” in
a vacuum chamber to produce doping in layers
© 1997-2004, R.LevinePage 12
PN Junction Diode
P NAnode Cathode
P-type N-type
Graph shows net electric chargedensity vs. distance right or left of junction
Electricallyneutralregion
Electricallyneutralregion
Region of extra electrons,represented by green color.
Region of missing or “depleted” electrons,represented by red color.
}So-called “depletion layers”
Electrode
Graphic Symbol
+
-
© 1997-2004, R.LevinePage 13
When P and N Pieces First Touch...
• (Touching surfaces must be microscopically clean in this example...)
• First, electrons spill over from the border side of the N material into the P material, because they are attracted by the greater nuclear positive electric charge of the P material
• This leaves a layer just inside of the left surface of the N material (red color) which has less electrons per unit volume than the neutral parts of the N piece
© 1997-2004, R.LevinePage 14
Depletion Layers• The width of these two layers increases until they reach an
equilibrium condition in which just enough electrons are on the left side to repel any more electrons spilling over.
• If we could mechanically break the P and N pieces suddenly apart at this time, we would leave some negative charge trapped on the P side, and a net positive charge trapped on the N side. (The charge may jump back creating a spark!)
• Because this is a semiconductor instead of a good conductor, these layers (called depletion layers, although only one of them is actually “depleted” below the normal number of electrons) stay in place at the two sides of the interface. (In a metal, the extra electrons would move quickly away from the interface and go all over the surface of piece of metal.)
• This double layer of two opposite net electric charges (+ and -) is also called a “dipole”
© 1997-2004, R.LevinePage 15
Current-Voltage Measurement• The Diode is a “non-linear” electrical device. This setup
(shown schematically) measures current, i, at various voltage v values
V
A
+-
Ammeter, measurescurrent
Ideal voltmeter measures diode’s voltage, but no current flows through the voltmeter. Real voltmeters allow very small current flow. Anode (top) of diode symbol is the conventional positive voltage terminal.
Adjustable or variable voltage source,can produce both positive and negativevoltages.
i
© 1997-2004, R.LevinePage 16
Typical Diode i-v Curve• Several distinct regions of operation
A
B
C
D
Vz 1 2 v (volts)
i(mA)
Note: a section of negativevoltage axis is not shown. Origin of graph,
v=0, i=0
1
Region Description
A-B Approximately linear increase incurrent vs. voltage.
B-C Accurate theoretical formula:
i = Io (e qv/kT -1)where Io is temperature dependent,but is typically ~10 µA at roomtemperature. Also kT/q 0.2 voltsat room temperature.
C-D Zener or avalanche breakdownregion. Approximately constantvoltage. Vz can be 3 to 600 V,depending on design of diode.
mA=milliAmperesof current;1 mA=0.001 A.
B is approx.boundarybetween exponentialand linear parts.
© 1997-2004, R.LevinePage 17
Forward Current Regions• In region A-B, the voltage across the depletion layer is very
small, and we mainly see the ordinary electrical resistance of the two neutral parts of the diode, resistance Rf.– The depletion layer is very thin.
• In region B-C, the depletion layer gets thicker or thinner, adding or removing electrons at their outer edges, when the voltage changes.
• When the applied voltage is positive, the depletion layer is very narrow, and most electrons can go across the junction (right to left flow of electrons makes a positive current left-to-right, since positive current flow is opposite negative electron flow*)
• The number of electrons which have enough energy to get across the depletion layer is dependent on temperature (more about this later)
• The theoretical prediction of this formula (stated without proof), based on electron thermal energy level, is very accurate in this region
* Blame Benjamin Franklin for using negative numbers for one kind of static electricity. If he knew then that electric current is mainly from electrons, he would have made the opposite choice, I’m sure! Before Franklin’s suggestion “positive” electricity was classified as vitreous (from rubbing glass) and “negative” electricity was classified as resinous (from rubbing amber). Franklin realized that they were two polarities of the same qualitative type, instead of two qualitatively different things.
© 1997-2004, R.LevinePage 18
Two-segment Approximation
• In some situations, a two segment straight line approximation can be reasonably accurate for some mathematical analysis purposes
1 2v (volts)
i(mA)
mA=milliAmperesof current;1 mA= 0.001 A
Origin of graph,v=0, i=0
Forward region is describedas a resistance Rf.
Special case is Rf=0 ohms.
Reverse current is described as zero. No description ofbreakdown voltage regionin this example.
© 1997-2004, R.LevinePage 19
Reverse Current• Reverse Current
– The depletion layer is very wide when reverse voltage is high. Very few electrons get into the depletion layer from the neutral parts of the diode. Only electrons “produced” inside the depletion layer will move through it. These electrons are “produced” by giving more energy to valence band electrons so that they become conduction band electrons (a change of electron wavelength). Two methods for giving electrons more energy:
• Higher temperature• Shine light (infra-red, visible, ultra-violet) on the junction
– The reverse “leakage” current (from origin to point C) is almost constant over most of the negative voltage range. Reverse current depends on the number of electrons per second which “appear” in the depletion layer, and not upon the voltage. Mainly temperature dependent.
© 1997-2004, R.LevinePage 20
Breakdown Current• In region C-D, the diode has a sudden increase in current.
This is called the “avalanche breakdown” or “Zener” region (named for physicist Clarence Zener)
• In this region, the high electric field in the middle of the depletion layers accelerates electrons produced there (by action of heat or light) so much that they can “dislodge” other electrons from the valence band (into the conduction band)
• When one energetic electron can “dislodge” two or more such electrons, we start a “chain reaction” in which these electrons can produce even more conduction electrons.
• This is like a geological avalanche, in which the first boulders rolling down a hill dislodge other boulders and so forth...
© 1997-2004, R.LevinePage 21
Depletion Layer Thickness• Depletion layer becomes narrow when positive voltage is
applied to the diode– Then more electrons spill over from N to P part of diode.
• Depletion layer becomes thicker when negative voltage is applied to the diode
• Thickness of depletion layer is main thing which controls how many electrons can cross the depletion layer “barrier”
distance right or left of junction
Vertical axis isnet electric chargedensity
Zero volts ondiode (blue)
Positive voltageon diode (red)
Negative voltage ondiode (green)
© 1997-2004, R.LevinePage 22
• The average number of electrons at each level of internal energy in a solid is given by the Fermi formula (stated here without proof). Non-integer values of n(E) indicate average of various integers.
Electron Energy
Ef (the Fermi energy level)
E
n(E)
1
0
Very low temperature (blue)Medium temperature (green)Very high temperature (red)
Eb, a typical “barrier” energyGrey shaded area on graph indicatesenergy levels with electrons at mediumtemperature. Gap surrounding Ef is dueto the band gap in a semiconductor.
1
n(E) = e ((E-Ef)/kT) +1
© 1997-2004, R.LevinePage 23
There are Either 1 or 0 Electrons at a Specific Energy Level (Pauli’s Exclusion
Principle)• Because of electron “spin” (intrinsic
magnetism and angular momentum) there are two wave arrangements at almost the same energy level
• Some documents describe the maximum number of electrons at each level as 2
• Some documents describe each level with different spin separately, and give the maximum number of electrons per level as 1
© 1997-2004, R.LevinePage 24
How Many Electrons Pass Over the Barrier?• The depletion layers in the diode act as an adjustable
energy level barrier to control electron flow across the two parts of the depletion layer– Positive applied battery voltage lowers the energy barrier, and
negative voltage raises the energy barrier
• The amount of current flow is related to the number of electrons which have enough (thermal) energy to naturally get past the barrier– This is shown on the previous graph by the shaded area under
a curve from the barrier energy, Eb, upward– Such a typical area is shaded under part of the medium
temperature (green) curve– You can see that the corresponding area would be greater
under the high temperature curve, although it is not marked
• For a positive voltage, the barrier is lowered so much that almost all the conduction electrons can pass through– Only the ordinary “ohmic” resistance of the neutral parts limits
the current when very high positive voltage is used!
© 1997-2004, R.LevinePage 25
Reverse Current• When the energy barrier is very high (large negative voltage)
almost no electrons have enough thermal energy to pass over the two parts of the depletion layer
• But the electric field in the junction, between the two parts of the depletion layer, is very strong:
– electrons in the left depletion layer repel any electron at the junction, pushing it to the right
– the right (positive) depletion region pulls any electrons at the junction to the right
• We have all the forces to move electrons an produce a large negative current…. except that there are almost no conduction electrons located at the junction!
• If a conduction electron is produced or created in the middle of the junction, it will immediately be moved by the strong electric field
• A few electrons “appear” in the junction each second because they have enough thermal energy to change from the valence to the conduction band just at the junction! (consider case without light on the junction)
• Thus the reverse current is dependent on the number of thermally produced conduction electrons, and not on the reverse voltage. It changes only due to temperature, not due to voltage changes.
© 1997-2004, R.LevinePage 26
Avalanche (Zener) Breakdown• Zener breakdown occurs due to high electric field in the
junction• High reverse electric fields are produced by:
– Heavily doped P and N materials to fabricate diode– High negative voltage
• They produce a larger charge density in the depletion layer, even at low reverse voltage
• Diodes made specifically to “break down” at low reverse voltages are called Zener diodes. They also are designed and made with cooling fins, etc. to keep them from melting under high voltage and high current (high power)
• Current is then limited only by the ohmic resistance (usually an external resistor designed to be used with the diode)
• Zener diodes are mostly used to produce an accurate reference voltage for measurement devices or analog-digital converters, etc.
© 1997-2004, R.LevinePage 27
Semiconductor Applications• One important use for diodes is to convert alternating current into
direct (unidirectional) current in power supply circuits.• Diodes are also useful in some logic devices, and we can make
some types of digital logic circuits using only diodes and resistors
• Transistors are more interesting and have more applications than diodes.
• Two types of transistors are widely used:– Bipolar Junction transistors1 (BJTs), which are physically like
two junction diodes back-to-back– Field Effect Transistors (FETs), consisting of a singlelarge
area junction diode, in which we use the voltage on a control (gate) electrode to modify the available current flow area outside the depletion layer for transverse current flow in the other part of the diode. This category includes metal oxide silicon (MOS) FETs
Note 1: The name “transistor” is a contraction of the two terms trans-resistor. Name due to John R. Pierce, Bell Laboratories scientist.
© 1997-2004, R.LevinePage 28
Transistor Properties• Transistors can “amplify” electrical signals
– In the normal amplification state, transistors actually control the flow of electric power, from a battery or other power source, usually in proportion to the input power from the signal
– The British name for the vacuum tube (the historical predecessor of the transistor) was a “valve,” which is a very good description of what a transistor does in the amplifying state
– It controls power flow from the power supply like a water valve controls water flow
• Transistors have 3 electrical terminals, and thus a separate input and output “port”– More convenient for processing analog or digital signals
© 1997-2004, R.LevinePage 29
Junction Transistor
N
P
N
Collector
Emitter
Base
C
B
E
Graphicsymbol
In the usual amplifying configuration,the base is more positive than the emitter, and the collector is at an even more positive voltage. The E-B junction is thus ON and the C-B junction is OFF (reverse biased). The thickarrow represents the magnitude of electron flow.Most of the electrons that pass from the Emitterto the Base are collected by the Collector.
NPN unit is shown.PNP units alsoare made, and useopposite voltagepolarities from NPN.The graphic symbolfor a PNP transistorhas the opposite arrowpoint direction.
© 1997-2004, R.LevinePage 30
Transistor Amplification• The voltage across the Emitter-Base junction
controls the Emitter current• A large and almost constant fraction () of the
emitter current is “collected” by the collector– The ratio iC/iE is traditionally called (alpha). It depends
mainly on the geometry of the transistor. Since the neutral region in the base is very narrow, most of the emitter-base electrons go into the base-collector junction, where the high electric field propels them out the collector electrode. A small fraction (1-) leaves via the base electrode. Typical value for is 0.99
– The ratio beta = /(1-) is the ratio of the collector current to base current. Typical value for is 99. The transistor therefore “amplifies” the base current by approximately 100 and produces a larger current at the collector.
© 1997-2004, R.LevinePage 31
One Computational Model• This simplified circuit model for a junction
transistor uses a current-controlled current source– The base current is viewed computationally as the thing
which controls the collector current
B
C
E
iC
iE= iB (1+)
iB
iB (A current-controlled current source.)
• This model only describes behavior when the collector junction is in reverse voltage state and emitter junction is in forward voltage state, typically for amplification purposes.
© 1997-2004, R.LevinePage 32
Field Effect Transistor
SourceElectrode
Drain Electrode
GateElectrode
P-gate, N-channel unit shown.
}
S
D
GGraphicSymbol
Notice the “blob” in theN-side depletion layer dueto electric field interactionwith Drain electrode.
The arrow indicatesdirection of electronflow. Narrowingof arrow suggestscurrent “strangling”effect from negativegate voltage, whichnarrows the neutral N channel.
Depletion layers.
The words “source”and “drain” are basedon the concept of positive charge flow.
© 1997-2004, R.LevinePage 33
Two FET Analysis Models
1. Variable resistor between Source and Drain– Resistance increases when Gate voltage is more
negative– Physically a good model
• Represents the narrowing of the N-channel• But computationally non-linear, leads to products of
independent variables like current and voltage
2. Current source between Source and Drain, controlled by gate voltage– Not as accurate physically for signals with large voltage
ranges– But computationally leads to linear equations, which are
easier to solve
© 1997-2004, R.LevinePage 34
Model 2• This circuit model uses a voltage-controlled
current source– The gate voltage is viewed computationally as the thing
which controls the source-drain current
Gate
Source
Drain
iS-D+ vG -
gvG
The parameter g is the so-called trans-conductance ofthe FET. It is the ratio of a change iniS-D to the causative change in vG.
Note that there is no current pathin this model between the gate and other parts of the FET. This is dueto assuming that the reverse currentof the gate-body junction is zero. Infact it is typically a few microamperes.
© 1997-2004, R.LevinePage 35
Metal Oxide Silicon (MOS) Transistor
SourceElectrode
Drain Electrode
GateElectrode(metal)
Also called insulated gate FET(IGFET). A layer of SiO2 (equi-valent to beach sand, shown inblue on the drawing) electricinsulation here is actually much thinner than the illustration. No P-type layer! This still produces a positive (red) depletion layer in the N-type part and channelwidth is controlled by the gatevoltage. No steady (dc) gate current flows for either positiveor negative gate voltage.
© 1997-2004, R.LevinePage 36
Multiple Gate Transistor - I
SourceElectrode
Drain Electrode
GateElectrode 1
Multiple gate electrodes areused to implement digitallogic functions (to be discussedmore in a later lecture). This form with side-by-side gatesallows some source-drain current to flow when either gate 1 ORgate 2 has a positive voltage. This implements the inclusive OR logical function with a minimumnumber of components, particularlywhen implemented in an integratedcircuit.
GateElectrode 2
© 1997-2004, R.LevinePage 37
Multiple Gate Transistor - II
SourceElectrode
Drain Electrode
GateElectrode 1
Current from source to drain flows only when both gate 1AND gate 2 are positive. Notethat there are two places wherenegative gate voltage couldpinch off the channel. Thisimplements the digital logic AND function.
All of these configurations canbe implemented in integratedcircuits, although these picturesshow source and drain electrodeson the edges.
GateElectrode 2
© 1997-2004, R.LevinePage 38
Generic Amplifier• All these 3-electrode transistor types can be used to build an amplifier• Digitally interesting things happen in the two extreme output voltage
regions of operation, aside from use of such devices for amplifying sound or radio signals
• More details on the operation of the amplifier in the next lecture.
output voltage point
+ -
inputvoltagesource
Generic “ground” or “earth”graphic symbol. Actually representsthe frame or cabinet in most modernequipment.
Fixed voltage powersupply, shown hereas a battery symbol.Often the power supplydevice and the wire frompower source to ground is omitted from drawings.
RL , “load” resistor (orLoudspeaker, etc.)
Generic “cathode”:(drain or emitter)
Generic control electrode: (gateor base)
Generic “anode”:(source or collector) Vpower
This box is replacedby a particular transis-tor in a real amplifier.
© 1997-2004, R.LevinePage 39
Radiation and Semiconductor Junctions
• Several important interactions between absorption and radiation of light and electromagnetic waves occur in semiconductor junctions
• These interactions relate to:– Temperature dependence of Io (“leakage current”)– Photo-voltaic cells (“electric eyes”)– Light Emitting Diodes– Laser Diodes
• Important to systems reliability and use of diodes in optical systems
© 1997-2004, R.LevinePage 40
Temperature Dependence of Io
• The term Io in the formula for diode current:
i = Io (e (qv/kT) -1), is itself temperature dependent• There is a very high electric field at the very
center of the junction, but usually almost no conduction electrons are present– High electric field is due to a combination of excess
electron repulsion and net positive depletion layer attraction, which both act in the same direction on any moveable electron which may exist at the junction center
– A conduction electron can be “created” (electron-hole “pair” production) in that location when a valence electron absorbs enough energy so that it reconfigures its wave function as a conduction electron there.
© 1997-2004, R.LevinePage 41
Conduction Electron Production• Energy could come from:
– Thermal kinetic energy • Interaction with thermal vibration of nuclear cores of atoms • More thermal energy transfer at higher temperature, leads to
greater Io reverse “leakage” current, exponentially increasing with temperature
– Direct electron absorption of radiation• Infra-red, visible or ultraviolet light, or x-rays, cosmic rays,
etc.• Frequency of radiation must be high enough so E=h•f is
greater than energy gap (where h is Planck’s constant). Radio frequency radiation is usually too low
– Due to “avalanche” chain reaction• Secondary effect of thermally created conduction electrons
at Zener breakdown voltage• Direct electron-electron interactions create even more
conduction electrons via “chain reaction”
© 1997-2004, R.LevinePage 42
Diode PhotoElectric Devices• These effects allow reverse-voltage diode
to generate current due to radiation– Photo-voltaic direct power conversion from
sunlight• Io proportional to incident light intensity
– opto-electric detector for fiber optic system receiver
• Avalanche diode is sensitive to very low radiation, due to “multiplication” of current by the avalanche effect
• Similar phenomena of electron avalanche was used in historical vacuum tube technology. Electron-multiplier photocells were used to detect very low levels of light and in the early Farnsworth “image dissector” TV camera
© 1997-2004, R.LevinePage 43
Undesired Radiation Effects• Devices which use semiconductor junctions
(transistors, etc.) for digital logic and memory purposes are adversely affected by low-level ionizing background radiation, cosmic rays, etc.
• Computer memory chips appeared to have random infrequent but mysterious data errors until this cause was identified in the 1970s
• Radiation-induced current pulses cause OFF transistors to suddenly go ON
• Integrated circuit packaging must shield the silicon chip from external radiation, and must not itself contain radioactive isotopes
• High purity levels required in plastic encapsulation as well as interior silicon!
© 1997-2004, R.LevinePage 44
Light Emission During Forward Current Flow
• When an electron crosses the junction from N to P side, its energy changes due to difference in interior average atomic number of the atom cores
• The electron “cloud” experiences oscillations during the transition from higher to the lower energy level– The frequency f of this oscillation is given by E2-E1=h•f– Some diodes are made with opaque enclosures so
emitted light is not noticeable. Light may also be in the Infra-red spectrum and not perceptible to the human eye! However, radiation is produced by forward current in a diode.
© 1997-2004, R.LevinePage 45
Light Emitting Diodes (LEDs)
• Greater difference in energy levels in the P and N sides of the diode (due to high dopant amounts) produce greater energy change, higher frequency light, shorter wavelength
• Earliest light emitting diodes produced infra-red or visible red light– LEDs in yellow, green and recently blue visible light
colors are now available
• LEDs are used extensively as indicator lamps, and as picture elements in color matrix displays for lap-top computers, etc.
• LEDs are used as electro-optic converters for multi-mode and graded index fiber optics
© 1997-2004, R.LevinePage 46
Laser Diodes (LDs)• Fabrication of light emitting junction surrounded by
partially reflecting surfaces which produce a standing wave electromagnetic field, thus causing intense emission of approximately mono-chromatic light (LASER=light amplification by stimulated emission of radiation)
• More efficient light output than LED• Narrower, monochromatic, focused beam
– Couples better into small core of single mode optical fiber than LED
– Less chromatic dispersion (pulse time spreading) in the fiber, so higher data bit rate is permitted
– Used also for reading/writing reflective spots on CD-ROM disks