Document

advertisement
Network for Computational Nanotechnology (NCN)
UC Berkeley, Univ.of Illinois, Norfolk State, Northwestern, Purdue, UTEP
First Time User Guide to
PN Junction V1.31
Saumitra R Mehrotra*, Ben Haley &
Gerhard Klimeck
Network for Computational Nanotechnology (NCN)
Electrical and Computer Engineering
http://nanohub.org/resources/229
*smehrotr@purdue.edu
Outline
 Introduction
 What is a PN Junction?
 Working of a PN Junction.
 What can be simulated by in PN Junction Lab?
 What if you just hit “Simulate”?
 Examples
 What if the doping is changed?
 What if the intrinsic region is included (e.g. PIN diode)?
 Limitation/Comments
 References
Saumitra R Mehrotra
What is a PN Junction?
• A PN junction is a device formed by combining p-type ( doped
with B,Al) and n-type (doped with P,As,Sb) semiconductors
together in close contact.
• PN junction can basically work in two modes,
» forward bias mode (as shown below: positive terminal connected to pregion and negative terminal connected to n region)
» reverse bias mode ( negative terminal connected to p-region and
positive terminal connected to n region)
PN junction device
Saumitra R Mehrotra
Working of a PN junction
Current
I-V characteristic of
a PN junction diode.
Reverse Bias
Forward Bias
Zener or
Avalanche
Breakdown
Voltage
• PN junction diode acts as a rectifier as seen in the IV characteristic.
• Certain current flows in forward bias mode.
• Negligible current flows in reverse bias mode until zener or
avalanche breakdown happens.
Refer https://nanohub.org/resources/68 for a detailed discussion on operation of PN junction.
Saumitra R Mehrotra
What can be simulated in PN junction lab?
PN junction device structure
Specify P-type region length. (More
number of nodes lead to higher
resolution but also more compute
time)
Specify intrinsic region length
Specify N-type region length
Specify doping level for Ptype and N-type region.
Saumitra R Mehrotra
What can be simulated in PN junction lab?
PN junction material definition
Specify the material to be
simulated (Si,Ge,GaAs)
Specify intrinsic minority
carrier lifetime (s).
Saumitra R Mehrotra
What can be simulated in PN junction lab?
PN junction environment definition
Specify temperature (K).
Specify applied voltage and
number of steps.
Saumitra R Mehrotra
What can be simulated in PN junction lab?
PN junction output plots
Default settings will simulate :
PN junction diode in forward bias mode with,
1m long P-type and N-type regions doped at 1e17 cm3.
Saumitra R Mehrotra

What if you just hit simulate?
• IV characteristic for PN junction in
forward bias mode in default
settings.
Knee voltage
•Current increases slightly till
Knee Voltage*.
•Beyond it current rises
exponentially.
*Refer https://nanohub.org/resources/68 for a detailed discussion on operation of PN
junction.
Saumitra R Mehrotra
What if you just hit simulate?
C-V characteristic for PN junction
in forward bias mode in default
settings.
• C  d –1
[1]&[2]
where,
C : capacitance across PN junction
&
d : depletion width* (insulating region
at the junction where carriers have
diffused away or have been swept by
the electric field.)
• Increasing bias  decreasing
depletion width  Increasing
Capacitance
*Refer https://nanohub.org/resources/68 for a detailed discussion on operation of PN
junction.
Saumitra R Mehrotra
What if you just hit simulate?
P-region
N-region
Built in charge, electric field
and potential at equilibrium.
Built in charge
Built in
potential,
Vbi= 0.834V
P-region
Built in electric field
Saumitra R Mehrotra
N-region
What if you just hit simulate?
P-region
N-region
Decreasing charge with applied bias due to
thinning of depletion width.
Built in charge, electric field and
potential at forward bias Va=0.6V
Potential
difference
Vbi-Va= 0.234V
Increased diffusion of
electrons across the
barrier lowered by Va.
P-region
N-region
Decreasing electric field with applied bias
due to thinning of depletion width.
Saumitra R Mehrotra
Positive bias at P side reduces
the barrier leading to increase in
diode current.
What if doping is changed?
On changing doping for both n-type and p-type
regions from 1e16 cm3 to 1e18 cm3.
• Increasing doping leads to increasing
built in potential, Vbi [1],[2].
Doping= 1e18 cm3
 N a .N d
Vbi  K bT log 
 ni
=V
Doping= 1e16
cm3
bi



Na : P region doping level (cm-3).
Nd : N region doping level (cm-3).
ni : Intrinsic carrier density (cm-3).
KbT : Thermal voltage (= 0.0259 V).
Saumitra R Mehrotra
What if intrinsic region is included i.e PIN diode?
On introducing an intrinsic region of length 0.2 m with default setting.
PIN diode (bold)
PN diode (light)
Increased depletion width (d) due to
addition of intrinsic region as seen
in energy band diagram.
Saumitra R Mehrotra
Junction Capacitance, Cj  d-1
shows a decrease as seen in CV
characteristic.
Limitations/Comments
• Large physical dimensions (>10um) might lead to non
convergence or large compute time.
• More nodes might be required for better convergence in
some cases i.e. high doping in PN junction.
• PN junction currently performs steady state simulations only,
no time dependent simulations are possible.
• Contacts during the simulation are considered to be ohmic
(i.e. Current-Voltage,I-V curve is linear and symmetric).
Saumitra R Mehrotra
References
PN junction theory
• [1] PN junction OPERATION : https://nanohub.org/resources/68
• [2] “Semiconductor Device Fundamentals”, by R.F. Pierret
PADRE
• [3] Dragica Vasileska; Gerhard Klimeck (2006), "Padre," DOI: 10254/
nanohub-r941.3.
• [4] PADRE MANUAL : http://nanohub.org/resource_files/tools/padre/doc/
index.html
Please report any comment/review at the following link,
• https://nanohub.org/resources/pntoy/reviews
If you reference this work in a publication, please cite as follows:
• Matteo Mannino; Dragica Vasileska; Michael McLennan; Xufeng Wang; Gerhard Klimeck
(2005), "PN Junction Lab," DOI: 10254/nanohub-r229.9.
Saumitra R Mehrotra
Download