Introduction to Diodes

Lecture notes: page 2-1 to 2-19

Sedra & Smith (6th Ed): Sec. 3.* and 4.1-4.4
Sedra & Smith (5th Ed): Sec. 3.7* and Sec. 3.1-3.4
* Includes details of pn junction operation which is not covered in this course
F. Najmabadi, ECE65, Winter 2012

Energy levels in an atom

The larger the energy level, the larger is
the spatial extent of electron orbital.

Discrete energy levels!

Each energy level can be filled with
a finite number of electrons.
Lowest energy levels are filled first.
Nucleus position

Electrons in the last filled energy level are called valance electrons
and are responsible for the chemical properties of the material.
F. Najmabadi, ECE65, Winter 2012

Energy Bands in Solids

For small inter-distance

between ions, energy levels
become energy bands.

Forbidden energy gaps

between energy bands.

Conduction band: the lowest energy band with electrons NOT tied to the atom.
Valance band: the highest energy band with electrons tied to the atom.
Band-Gap is the energy difference between the top of valance band and the
bottom of conduction band
F. Najmabadi, ECE65, Winter 2012

Difference between conductors,

semiconductors and insulators
In a metal, the conduction band is partially filled. These electron can move
easily in the material and conduct heat and electricity (Conductors).
In a semi-conductor at 0 k the conduction band is empty and valance band
is full. The band-gap is small enough that at room temperature some
electrons move to the conduction band and material conduct electricity.
An insulator is similar to a semiconductor but with a larger band-gap.
Thus, at room temperature very few electrons are in the conduction band.
Metal

F. Najmabadi, ECE65, Winter 2012

Semiconductor
at T = 0 k

Semiconductor
at T > 0 k

Insulator

Electric current in a semiconductor is due

to electrons and holes
At T > 0 k, some electrons are promoted
to the conduction bands.
A current flows when electrons in the
conduction band move across the
material (e.g., due to an applied electric
field).

Electrons

Conduction band
Valance band
holes or available slots
in the valance band

A current also flows when electrons in the

valance band jump between available
slots in the valance bands (or holes).
o An electron moving to the left = a hole
moving to the right!
o We call this is a hole current to
differentiate this current from that due to
conduction band electrons.
F. Najmabadi, ECE65, Winter 2012

Pure Si Crystal

Doping increases the number of charge carriers

Doped n-type Semiconductor

Doped p-type Semiconductor

Donor atom (P doping) has an extra

electron which is in the conduction band.

Acceptor atom (B doping) has one less

electrons: a hole in the valance band.

Charge Carriers:
o Electrons due to donor atoms
o Electron-hole pairs due to thermal
excitation
o e: majority carrier, h: minority carrier

Charge Carriers:
o Holes due to acceptor atoms
o Electron-hole pairs due to thermal
excitation
o h: majority carrier, e: minority carrier

F. Najmabadi, ECE65, Winter 2012

Electric current due to the motion of

charge carriers
Drift Current: An electric field forces charge carriers to move and
establishes a drift current:

I drift = AqnE
Diffusion Current: As charge carrier move randomly through the
material, they diffuse from the location of high concentration to
that of a lower concentration, setting up a diffusion current:

I diffusion

dn
= A| q | D
dx

Einstein Relationship:

kT
= VT =

|q|

F. Najmabadi, ECE65, Winter 2012

o VT is called the Thermal voltage or

volt-equivalent of temperature
o VT = 26 mV at room temperature

Junction diode

Simplified physical structure

F. Najmabadi, ECE65, Winter 2012

Construction on a CMOS chip

A pn junction with open terminals

(excluding minority carriers)
High concentration of e on the n side
Electrons diffuse towards the junction

High concentration of h on the p side

Holes diffuse towards the junction

Holes from the p side and electrons

from the n side combine at the
junction, forming a depletion region
Idif

p side is negatively
charged because it
has lost holes.

Idif

n side is positively
charged because it
has lost electrons.

A potential is formed which inhibits further diffusion

of electron and holes (called junction built-in voltage)
F. Najmabadi, ECE65, Winter 2012

A pn junction with open terminals

(including minority carriers)
Thermally-generated minority carriers on the n side
(holes) move toward the depletion region, and are
swept into the p side by the potential where the
combine with electrons. (similar process for
minority carriers on the p side). This sets up a drift
current, IS.
To preserve charge neutrality, a non-zero Idif = IS
should flow (height of potential is slightly lower).
Idif scales exponentially with changes in the
voltage barrier.
IS is independent of the voltage barrier but is a
sensitive function of temperature.
F. Najmabadi, ECE65, Winter 2012

Idif
IS

pn Junction with an applied voltage

Reverse-Bias:
Height of the barrier is increased, reducing Idif
Idif approaches zero rapidly, with iD IS
A very small negative iD !

Forward-Bias:
Height of the barrier is decreased, increasing Idif

Idif increases rapidly with vD leading to iD Idif

A very large positive iD !

F. Najmabadi, ECE65, Winter 2012

Diode iv characteristics equation

iD = I S e vD / nVT 1

IS : Reverse Saturation Current

(10-9 to 10-18 A)
VT : Volt-equivalent temperature
(= 26 mV at room temperature)
n: Emission coefficient
(1 n 2 for Si ICs)

For | vD | 3nVT
Forward bias : iD I S e vD / nVT
Reverse bias : iD I S
Sensitive to temperature:
IS doubles for every 7oC increase
VT = T /11,600

F. Najmabadi, ECE65, Winter 2012

For derivation of diode iv equation, see Sedra & Smith Sec. 3

Diode Limitations
Thermal load, P = iD vD
(typically specified as
maximum iD )
Reverse Breakdown at
Zener voltage (VZ)
(due to Zener or avalanche
effects)
Zener diodes are made specially
to operate in this region!

F. Najmabadi, ECE65, Winter 2012

How to solve diode circuits

F. Najmabadi, ECE65, Winter 2012

Diode circuit equations are nonlinear

KCL : current iD in all elements
KVL : vi = RiD +vD

iD = I S e vD / nVT 1

Two equation in two-unknowns to solve for iD and vD

Non-linear equation: cannot be solved analytically
Solution methods:
o Numerical (PSpice)
o Graphical (load-line)
o Approximation to get linear equations (diode piece-linear model)

F. Najmabadi, ECE65, Winter 2012

Graphical Solution (Load Line)

KCL : current iD in all elements
KVL : vi = RiD +vD

iD = I S e vD / nVT 1

iD = I S e vD / nVT 1

Intersection of two curves

satisfies both equations
and is the solution

vi/R
iDQ

vi = RiD +vD
Load Line

vDQ
F. Najmabadi, ECE65, Winter 2012

vi

Diode piecewise-linear model:

Diode iv is approximated by two lines
Constant Voltage Model

Diode ON

Diode OFF

VD0

Diode ON :

vD = VD 0

Diode OFF : iD = 0

and

iD 0

and

vD < VD 0

Circuit Models:
ON:

" cut - in" voltage, VD 0 = 0.6 0.7 V for Si

F. Najmabadi, ECE65, Winter 2012

OFF:

Recipe for solving diode circuits

(State of diode is unknown before solving the circuit)

1. Write down all circuit equations and simplify as much as
possible
2. Assume diode is one state (either ON or OFF). Use the diode
equation for that state to solve the circuit equations and find
iD and vD
3. Check the inequality associated with that state (range of
validity). If iD or vD satisfy the inequality, assumption is
correct. If not, go to step 2 and start with the other state.
NOTE:
o This method works only if we know the values of all elements so
that we can find numerical values of iD and vD .
o For complicated circuits use diode circuit models.
F. Najmabadi, ECE65, Winter 2012

Example 1: Find iD and vD for R = 1k, vi = 5 V, and Si Diode (VD0 = 0.7 V).
KCL : current iD in all elements
KVL : vi = RiD +vD
5 = 103 iD +vD
Assume diode is OFF : iD = 0 and vD < VD 0
5 = 103 0 +vD vD = 5 V
vD = 5 V > VD 0 = 0.7 V Assumption incorrect
Assume diode is ON : vD = VD 0 = 0.7 V

and

iD 0

5 = 103 iD +0.7 iD = 4.3 mA

iD = 4.3 mA > 0 Assumption correct

Diode is ON with iD = 4.3 mA and vD = 0.7 V).

F. Najmabadi, ECE65, Winter 2012

Example 1: Find iD and vD for R = 1k, vi = 5 V, and Si Diode (VD0 = 0.7 V).
Solution with diode circuit models:

Diode OFF : iD = 0 and vD < VD 0

Diode ON :

vD = VD 0 and iD 0

5 = 103 0 +vD vD = 5 V

5 = 103 iD +0.7 iD = 4.3 mA

vD = 5 V > VD 0 = 0.7 V Incorrect!

iD = 4.3 mA 0 Correct!

Diode is ON with iD = 4.3 mA and , vD = 0.7 V.

F. Najmabadi, ECE65, Winter 2012

Parametric solution of diode circuits is desirable!

Recipe:
1. Draw a circuit for each state of diode(s).
2. Solve each circuit with its corresponding diode equation.
3. Use the inequality for that diode state (range of validity) to
find the range of circuit variable which leads to that state.

F. Najmabadi, ECE65, Winter 2012

Example 2: Find vD in the circuit below for all vi .

Diode OFF : iD = 0 and vD < VD 0

vi = R 0 +vD vD = vi
vD < VD 0

vi < VD 0

Diode ON :

Solution
Inequality

For vi VD 0 , Diode ON

and

iD 0

vD = VD 0
vi = R iD + VD 0 iD = (vi VD 0 ) / R
iD 0

and

For vi < VD 0 , Diode OFF and

F. Najmabadi, ECE65, Winter 2012

vD = VD 0

vi VD 0

vD = VD 0
vD = vi

Other types of diodes

Schottky Barrier Diode

Zener Diode

Large IS and VD0 0.3 V

Light-emitting diode (LED)

VD0 = 1.7 1.9 V

F. Najmabadi, ECE65, Winter 2012

Made specially to operate in the

reverse breakdown region.
Useful as a reference voltage in
many circuits.

Zener Diode piecewise-linear model

Diode ON :

vD = VD 0

Diode OFF : iD = 0

Diode ON

Zener :

vD = VZ

Circuit Models:
Diode OFF

ON:
VD0

OFF:

Zener
Zener:
F. Najmabadi, ECE65, Winter 2012

and

iD 0

and

vD < VD 0

and

iD < 0

Zener diodes are useful in providing

reference voltages
Example 3: Find the iv characteristics of the two-terminal circuit below
(for vo > 0)
1) Assume diode in Zener region :
vD = VZ and iD < 0
KVL : vo = vD = VZ = constant

(Independent of io !)

Diode in Zener region for iD < 0

KCL : iD = io i

Acts as independent
voltage sources
even if vs changes!

KVL : vs = Ri + VZ
iD = io

vs VZ
R

iD < 0 io <
F. Najmabadi, ECE65, Winter 2012

vs VZ
= io ,max
R

Example 3 (contd)

2) Assume diode in reverse bias region :

iD = 0 and VZ < vD < VDo
KCL :

i = io

KVL : vs = Rio + vo
vo = vs Rio

(vo drops as io increases)

Diode in reverse - bias region for VZ < vD < VDo

vo = vD
VZ < vD < VDo + VZ > vo > VDo
0 vo = vs Rio < VZ

F. Najmabadi, ECE65, Winter 2012

vs VZ
v
< io s
R
R

Other piecewise linear models for diode

Diode iv characteristics can be
modeled with a sloped line:
vD = VD0 + RDiD
(instead of vD = VD0 )
Not used often:
o Model needs two parameters:
(RD and VD0 ) and the choice is
somewhat arbitrary.
o Extra work does not justify
increased accuracy
o Useful only changes in vD are
important

F. Najmabadi, ECE65, Winter 2012

Other piecewise linear models for diode

Diode Zener region can also be
modeled with a sloped line:
vD = VZ0 + RZ iD
(instead of vD = VZ0 )
Useful when changes in vD is
important.
o For example, If we use this
model for Example 2, we find*:

vo VZ 0 RZ io
instead of
vo = VZ = constant

F. Najmabadi, ECE65, Winter 2012

* See lecture notes, page 2-18