You are on page 1of 46

LIGHT EMITTING DIODE

– Design Principles
4 Main Issues
1. The device configuration
2. Materials requirements
3. Materials selection
4. Material issues
What is LED?

to rs
du c
i c on ality
Sem i ng t!
q u
r
b i gh
to l

LED are semiconductor p-n junctions that under forward bias
conditions can emit radiation by electroluminescence in the
UV, visible or infrared regions of the electromagnetic
spectrum. The qaunta of light energy released is
Applications of LEDs
What Are Diodes Made Out Of?

• Silicon (Si) and Germanium (Ge) are the two most common Si Si Si
single elements that are used to make Diodes. A +4 +4 +4

compound that is commonly used is Gallium Arsenide
(GaAs), especially in the case of LEDs because of it’s large Si Si Si
bandgap. +4 +4 +4

• Silicon and Germanium are both group 4 elements,
meaning they have 4 valence electrons. Their structure Si Si Si
allows them to grow in a shape called the diamond lattice. +4 +4 +4

• Gallium is a group 3 element while Arsenide is a group 5
element. When put together as a compound, GaAs creates
a zincblend lattice structure. The diagram above shows the
• In both the diamond lattice and zincblend lattice, each atom 2D structure of the Si crystal.
shares its valence electrons with its four closest neighbors. The light green lines represent
the electronic bonds made
This sharing of electrons is what ultimately allows diodes to
when the valence electrons are
be build. When dopants from groups 3 or 5 (in most cases) shared. Each Si atom shares
are added to Si, Ge or GaAs it changes the properties of one electron with each of its
the material so we are able to make the P- and N-type four closest neighbors so that
materials that become the diode. its valence band will have a full
8 electrons.

Kristin Ackerson, Virginia Tech EE
Spring 2002
N-Type Material

N-Type Material: When extra valence electrons are introduced into a
material such as silicon an n-type material is
produced. The extra valence electrons are
introduced by putting impurities or dopants into the
silicon. The dopants used to create an n-type
+4 +4 +4 material are Group V elements. The most commonly
used dopants from Group V are arsenic, antimony
and phosphorus.
+4 +5 +4 The 2D diagram to the left shows the extra electron
that will be present when a Group V dopant is
introduced to a material such as silicon. This extra
electron is very mobile.
+4 +4 +4

Kristin Ackerson, Virginia Tech EE
Spring 2002
P-Type Material

P-Type Material: P-type material is produced when the dopant that is
introduced is from Group III. Group III elements
have only 3 valence electrons and therefore there is
an electron missing. This creates a hole (h+), or a
positive charge that can move around in the material.
+4 +4 +4 Commonly used Group III dopants are aluminum,
boron, and gallium.
The 2D diagram to the left shows the hole that will be
+4 +3 +4 present when a Group III dopant is introduced to a
material such as silicon. This hole is quite mobile in
the same way the extra electron is mobile in a n-type
material.
+4 +4 +4

Kristin Ackerson, Virginia Tech EE
Spring 2002
The PN Junction
Steady State1

Metallurgical Junction
Na Nd

- - - - - - + + + + +
- - - - - - +
- - - - - -
P - - - - - -
+
+
+ + + +
n
- - - - - -
+ + + + +
+

+ +
Space Charge + + +
ionized Region + ionized
acceptors + + + + + donors
+
E-Field
_ _
+ +
h+ drift = h+ diffusion e- diffusion = e- drift

Kristin Ackerson, Virginia Tech EE
Spring 2002
The PN Junction
Steady State
Metallurgical
Na Junction Nd

+ + + + +
- - - - -
+ + + + + When no external source is
P - - - - - +
+
+
+
+
+
+
+
+
+
n connected to the pn junction,
- - - - - diffusion and drift balance
- - - -Space
- Charge each other out for both the
ionized Region ionized
acceptors donors holes and electrons
E-Field
_ _
+ +
h+ drift == h+ diffusion e- diffusion == e- drift

Space Charge Region: Also called the depletion region. This region includes the
net positively and negatively charged regions. The space charge region does not
have any free carriers. The width of the space charge region is denoted by W in pn
junction formula’s.

Metallurgical Junction: The interface where the p- and n-type materials meet.

Na & Nd: Represent the amount of negative and positive doping in number of
carriers per centimeter cubed. Usually in the range of 1015 to 1020 .
Kristin Ackerson, Virginia Tech EE
Spring 2002
The Biased PN Junction
Metal
Contact
“Ohmic
Contact” _
(Rs~0) +
Applied Electric
P Field n

I

_
+

Vapplied
The pn junction is considered biased when an external voltage is applied. There are two
types of biasing: Forward bias and Reverse bias.
These are described on then next slide.
Kristin Ackerson, Virginia Tech EE
Spring 2002
The Biased PN Junction

Forward Bias: In forward bias the depletion region shrinks slightly in width. With
this shrinking the energy required for charge carriers to cross the
depletion region decreases exponentially. Therefore, as the
Vapplied > 0 applied voltage increases, current starts to flow across the
junction. The barrier potential of the diode is the voltage at which
appreciable current starts to flow through the diode. The barrier
potential varies for different materials.

Reverse Bias: Under reverse bias the depletion region widens. This causes the
electric field produced by the ions to cancel out the applied reverse
bias voltage. A small leakage current, Is (saturation current) flows
Vapplied < 0 under reverse bias conditions. This saturation current is made up
of electron-hole pairs being produced in the depletion region.
Saturation current is sometimes referred to as scale current
because of it’s relationship to junction temperature.

Kristin Ackerson, Virginia Tech EE
Spring 2002
Properties of Diodes
Figure 1.10 – The Diode Transconductance Curve2

ID (mA) • VD = Bias Voltage
• ID = Current through
Diode. ID is Negative for
Reverse Bias and
Positive for Forward Bias
• IS = Saturation Current
IS • VBR = Breakdown Voltage
VBR • Vφ = Barrier Potential
Voltage
~V VD
φ

(nA)
Kristin Ackerson, Virginia Tech EE
Spring 2002
Properties of Diodes
The Shockley Equation
• The transconductance curve on the previous slide is characterized by the following
equation:
I = I (e
D S
VD/η VT
– 1)
• As described in the last slide, ID is the current through the diode, IS is the saturation
current and VD is the applied biasing voltage.
• VT is the thermal equivalent voltage and is approximately 26 mV at room temperature.
The equation to find VT at various temperatures is:
VT = kT
q
k = 1.38 x 10-23 J/K T = temperature in Kelvin q = 1.6 x 10-19 C
∀ η is the emission coefficient for the diode. It is determined by the way the diode is
constructed. It somewhat varies with diode current. For a silicon diode η is around 2
for low currents and goes down to about 1 at higher currents

Kristin Ackerson, Virginia Tech EE
Spring 2002
Getting to know LED
Advantages of Light Emitting Diodes
(LEDs)
Longevity:
The light emitting element in a diode is a small
conductor chip rather than a filament which
greatly extends the diode’s life in comparison
to an incandescent bulb (10 000 hours life time
compared to ~1000 hours for incandescence
light bulb)
Efficiency:
Diodes emit almost no heat and run at very
low amperes.
Greater Light Intensity:
Since each diode emits its own light
Cost:
Not too bad
Robustness:
Solid state component, not as fragile as
incandescence light bulb
Luminescence is the process
behind light emission

• Luminescence is a term used to describe the
emission of radiation from a solid when the
solid is supplied with some form of energy.
• Electroluminescence  excitation results
from the application of an electric field
• In a p-n junction diode injection
electroluminescence occurs resulting in light
emission when the junction is forward biased
Excitation
E

Electron (excited by the
biased forward voltage) is
in the conduction band

k

Normally the recombination takes
place between transition of electrons
between the bottom of the conduction
band and the top of the valance band
(band exterma).
The emission of light is therefore;
Hole is in valance
hc/λ = Ec-Ev = Eg(only direct band gap
band
How does it work?

P-n Electrical
junction Contacts

A typical LED needs a p-n
junction
There are a lot of electrons
and holes at the junction due
to excitations
Electrons from n need to be
injected to p to promote
recombination
Junction is biased to produce Recombinatio
even more e-h and to inject
electrons from n to p for
n produces
Injection Luminescence
in LED
 Under forward bias – majority carriers from both sides of the
junction can cross the depletion region and entering the
material at the other side.
 Upon entering, the majority carriers become minority carriers
 For example, electrons in n-type (majority carriers) enter the p-
type to become minority carriers
 The minority carriers will be larger  minority carrier injection
 Minority carriers will diffuse and recombine with the majority
carrier.
 For example, the electrons as minority carriers in the p-region
will recombine with the holes. Holes are the majority carrier in
the p-region.
 The recombination causes light to be emitted
 Such process is termed radiative recombination.
Recombination and Efficiency
(a) (b)
p n+ p n+

ECEg Eg
hν =Eg
EF eVo
EV

Electrons in CB

Holes in VB

◘Ideal LED will have all injection electrons to take part in
the recombination process
◘In real device not all electron will recombine with holes to
radiate light
◘Sometimes recombination occurs but no light is being
emitted (non-radiative)
Emission wavelength, λ g

◘ The number of radiative recombination is proportional to
the carrier injection rate
◘ Carrier injection rate is related to the current flowing in the
junction
◘ If the transition take place between states (conduction and
valance bands) the emission wavelength, λ g = hc/(EC-EV)

◘ EC-EV = Eg
◘ λ g = hc/Eg
Construction of Typical LED

Al
Light output
SiO2

p

n
Electrica
l
contacts
Substrate
LED Construction

 Efficient light emitter is also an efficient absorbers of
radiation therefore, a shallow p-n junction required.
 Active materials (n and p) will be grown on a lattice
matched substrate.
 The p-n junction will be forward biased with contacts
made by metallisation to the upper and lower surfaces.
 Ought to leave the upper part ‘clear’ so photon can
escape.
 The silica provides passivation/device isolation and
carrier confinement
Efficient LED
 Need a p-n junction (preferably the same semiconductor
material only different dopants)
 Recombination must occur  Radiative transmission to
give out the ‘right coloured LED’
 ‘Right coloured LED’  hc/λ = Ec-Ev = Eg
 so choose material with the right Eg
 Direct band gap semiconductors to allow efficient
recombination
 All photons created must be able to leave the
semiconductor
 Little or no reabsorption of photons
Correct band gap Direct band
gap
Materials
Requirements

Efficient radiative Material can be
pathways must made p and n-
exist type
 UV-ED λ ~0.5-400nm
Direct band gap
Eg > 3.25eV
materials
 LED - λ ~450-650nm
e.g. GaAs not Si
Eg = 3.1eV to 1.6eV
 IR-ED- λ ~750nm- 1nm
Eg = 1.65eV

Candidate
Materials

Materials with refractive Readily doped n or p-
index that could allow types
light to ‘get out’
Visible LED
Definition:
LED which could emit visible light, the band gap of the materials that
we use must be in the region of visible wavelength = 390- 770nm.
This coincides with the energy value of 3.18eV- 1.61eV which
corresponds to colours as stated below:

The band gap,
Violet ~ 3.17eVEg that the
Blue ~ 2.73eV semiconductor
Green ~ 2.52eV must posses to
Colour of Yellow ~ 2.15eV emit each light
an LED
should Orange ~ 2.08eV
emits Red ~ 1.62eV
Electromagnetic Spectrum

The appearance of the
Visible lights V ~ 3.17eV
visible light will be the results
B ~ 2.73eV of the overlap integral
between the eye response
G ~ 2.52eV curve and the spectral power
Y ~ 2.15eV of the device  the peak of
the luminous curve will not in
O ~ 2.08eV general be the same as the
R ~ 1.62eV peak of the spectral power
curve
Candidate Materials for LED’s
Candidate Materials
Group III-V & Group II-VI
Group II Group III Group IV Group V

iii iv v
N
ii P
Al
Ga As

In

Periodic Table to show group III-V and II-V
binaries
Group III-V (1950)

The era of III–V compound semiconductors
started in the early 1950s when this class of
materials was postulated and demonstrated by
Welker (1952, 1953). The class of III–V
compounds had been an unknown substance
prior to the 1950s that does not occur naturally.
The novel man-made III–V compounds proved
to be optically very active and thus instrumental
to modern LED technology.
Group III-V LED materials
Al N AlN, AlP,AlAs
Ga P Binary
GaN, GaP,
GaAs compounds
In As
InN, InP, InAs
GaP GaAsP Ternary
GaAs
GaA GaAsAl compound
l s
Questions to ask when choosing the right material:
1. Can it be doped or not?
2. What wavelength it can emit?
3. Would the material able to allow radiative recombiation?
4. Direct or indirect semiconductor?
LED-Electrical Properties-PN junctions
• PN junction diode in
forward bias, the
electron-hole
recombination leads to
photon emission

• I = Is(eeV/kT -1)

• Threshold voltage Vth =
Eg/e

• I = IseeV/ ηkT
where η is the ideality
factor

Double Heterostructure is used to confine the carriers, improving
the radiative recombination rate
From Light-Emitting Diodes, Fred Schubert.
LED-Optical Properties-Light Escape Cone

• Total internal reflection at the semiconductor air interface
reduces the external quantum efficiency.
• The angle of total internal reflection defines the light escape
cone.
sinθc = nair /ns
• Area of the escape cone = 2πr2(1-cosθc)
• Pescape / Psource = (1-cosθc)/2 = θc2/4 = (nair 2/ns2)/4

From Light-Emitting Diodes, Fred Schubert.
LED-Optical Properties-Emission Spectrum

• Light intensity in air
(Lambertian emission
pattern) is given by
Iair = (Psource/4πr2) X
(nair2/ns2) cosΦ

• Index contrast
between the light
emitting material and
the surrounding
region leads to non-
isotropic emission
pattern

From Light-Emitting Diodes, Fred Schubert.
LED Radiation Patterns

• LED:Directional light source, maximum emitted
power in the direction perpendicular to the emitting
surface.
• typical radiation pattern shows that most of the
energy is emitted within 20° of the direction of
maximum light.
• Some packages for LEDs include plastic lenses to
spread the light for a greater angle of visibility.

38
Structure and
electroluminescence

39
Band Gap

• Various band gaps different photon energies
Ultra violet :GaN 3.4 ev –infra-red: InSb 0.18 ev
• Ternary&quarternaryincreasing number of
available energies

40
Colors
• III-V materials
• Before II-VI (hard to have p-n junction)
• Solution: Nitrogen ZnSe (MBE grown)
• Progress: using multilayer hetero structures by MBE(Mulecular Beam
Epitaxy) & OMVPE
(organometalic vapor-phase epiaxy)
• (AlGaAs) - red and infrared
• (AlGaP) - green
• (AlGaInP) - high-brightness orange-red, orange, yellow, and green
• (GaAsP) - red, orange-red, orange, and yellow
• (GaP) - red, yellow and green
• (GaN) - green, pure green (or emerald green), and blue
• (InGaN) - near ultraviolet, bluish-green and blue
• (SiC) as substrate - blue
• (Si) as substrate - blue (under development)
• (Al2O3) as substrate - blue
• (ZnSe),(GaN) - blue
• (C) - ultraviolet
• (AlN), (AlGaN) - near to far ultraviolet
• New colors: pink and purple :2 layers of phosphors on Blue LED chip

41
LED Characteristics

• Forward biased fast increase in
current(control needed)

44
45