Failure Mechanisms of Insulated Gate Bipolar Transistors (Igbts)

Failure Mechanisms of Insulated Gate
Bipolar Transistors (IGBTs)
Nathan Valentine, Dr. Diganta Das, and Prof. Michael Pecht
www.calce.umd.edu
Center for Advanced Life Cycle Engineering (CALCE)
2015 NREL Photovoltaic Reliability Workshop
diganta@umd.edu
calce Prognostics and Health Management Consortium 1 University of Maryland

TM
Copyright © 2015 CALCE

CALCE Introduction
•   The Center for Advanced Life Cycle Engineering (CALCE)

formally started in 1984, as a NSF Center of Excellence in systems
reliability.
•   One of the world’s most advanced and comprehensive testing and
failure analysis laboratories
•   Funded at $6M by over 150 of the world’s leading companies
•   Supported by over 100 faculty, visiting scientists and research
assistants
•   Received NSF Innovation Award and NDIA Systems Engineering
Excellence Award
in 2009 and IEEE
Standards Education
Award in 2013.

TM

IGBT Applications
•   Need for more compact power converters achieved through faster device
switching
•   IGBTs are the ideal choice with switching frequencies of 1kHz-150kHz and
current handling of up to 1500A
Induction Heating Units Power Converters Electric Cars
Uninterruptible
Electric Trains Wind Turbines
Power Supplies
TM

IGBT Technologies
Source:
Infineon

TM

Failed Wind Turbine IGBT Module
Unused IGBT Failed IGBT which experienced a

thermal runaway, burning the module

TM

Steps in Reliability Evaluation
•  Quantify the life cycle conditions

•  Failure Modes, Mechanisms, and Effects Analysis
(FMMEA) > reliability analysis, assess design
tradeoffs and revise/update design
•  Part, material and supplier selection
•  Virtual qualification (VQ), including stress and
thermal analysis

TM

FMMEA Methodology
Define system and identify
elements and functions to be analyzed
Identify potential failure modes
Identify life cycle profile Identify potential failure causes
Identify potential failure mechanisms
Identify failure models
Prioritize failure mechanisms
Document the process

TM

IGBT Failure Modes and Mechanisms
•  Failure modes in an IGBT are simple at top level:
–  Short circuit
–  Open circuit
–  Parameter drift
•  Parameter drift occurs as a part degrades and the
electrical characteristics such as VCE(ON) or ICE drift
from the acceptable operating range due to the
accumulation of damage within a device or module

TM

Failure Modes and Mechanisms
Potential Failure
Potential Failure Modes (Sites) Potential Failure Causes Mechanisms (Parameters
affected)
Short circuit, loss of gate control, High temperature, high Time dependent dielectric
increased leakage current (Oxide) electric field, overvoltage breakdown (Vth, gm)
Loss of gate control, device High electric field,

burn-out (Silicon die) overvoltage, ionizing Latch-up (VCE(ON))
radiation
High leakage currents

Overvoltage, high
(Oxide, Oxide/Substrate Hot electrons (Vth, gm)
current densities
Interface)
High temperature, high

Open Circuit (Bond Wire) current densities Bond Wire Cracking,
Lift Off (VCE(ON))
High temperature, high Voiding,

Open Circuit (Die Attach) Delamination of Die
current densities
Attach (VCE(ON))

TM

Examples of Failure Models
Failure Mechanism Failure Sites Failure Causes Failure Models
Die attach, Wirebond/TAB, Nonlinear Power
Fatigue Solder leads, Bond pads, Cyclic Deformations Law (Coffin-Manson)
Traces, Vias/PTHs, (Δ T, Δ H, Δ V)
Interfaces
Corrosion Metallizations M, ΔV, T, chemical Eyring (Howard)
Electromigration Metallizations T, J Eyring (Black)
Conductive Filament Between Metallizations M, ΛV Power Law (Rudra)

Formation
Stress Driven Metal Traces σ, T Eyring (Okabayashi)
Diffusion Voiding
Time Dependent Dielectric layers V, T Arrhenius (Fowler-
Dielectric Breakdown Nordheim)
Δ: Cyclic range V: Voltage

Λ: gradient M: Moisture
T: Temperature J: Current density
H: Humidity σ: Stress

TM

CALCE Simulation Assisted Reliability
Assessment (SARA®) Software
http://www.calce.umd.edu/software
calceEP
calcePWA Device andPackage
Circuit Card Assemblies Failure Analysis
Thermal Analysis
Vibrational Analysis
Shock Analysis
Failure Analysis
Conductor II
Whisker
Spacing (ls)
calceFAST
Conductor I
Failure Assessment
calceTinWhisker FailureRiskCalculator Software Toolkit

TM

Thermally-Induced Stresses in IGBT

Material CTE (10-6 K-1) Conductivity
(W m-1 K-1)
A12O3 6.8 24
AlN 4.7 170
Si3N4 2.7 60
BeO 9 250
Al
23.5 237
Cu
17.5 394
Mo
5.1 138 - Bond Wire Fatigue
Si
2.6 148
AlSiC 7.5 200 - Solder Joint Fatigue

TM

IGBT Power Cycling Experiment
•  IGBT samples were power Switching at 1 or 5 kHz

cycled between specified
temperatures TMin and TMax.
The devices were switched
at 1 or 5 kHz. Cooling was
carried out passively by
exposure to ambient Cooling
temperature. TMax
•  This ‘power’ (thermal)
cycling was repeated until
failure occurred by latchup TMin
or by failure to “turn on”. Heating
Time
Power cycling illustration
TM

Parasitic Thyristor in IGBT Structure
Internal PNP Bipolar Transistor

TM

Parasitic Thyristor in IGBT Structure
Parasitic NPN Bipolar Transistor

TM

Die Attach Acoustic Scan Images
Delaminated surface Melted die attach
New IGBT sample. Failure to turn on Failure by latchup after

after 3126 power 1010 power cycles, ΔT
cycles, ΔT = 75°C. = 100°C. Melting T of
Die attach shows die attach = 233°C*.
delamination.
*Specification sheet for Sn65Ag25Sb10 solder from Indium Corp. Indalloy 209.

TM

Bond Wire Failures
Bond Wire Cracking Bond Wire Liftoff

TM

Lifetime Statistics of Experimental
Results
2P-Weibull with
95% confidence bounds
125-225C Data
β = 2.60
η = 1191 150-200C Data
ρ = 0.96 β = 2.26
MTTF = 1058 η = 7134
ρ = 0.96
MTTF = 6320
ANOVA p-value = 7.6E-6 1 kHz

∴ Different distributions 5 kHz
60% duty cycle

TM

Prediction of Other Reliability Metrics
Temperature Range MTTF (Cycles) [B5%Life; B95%Life]
(Cycles)
150-200°C 6320 cycles [1922; 11,582]
125-225°C 1058 cycles [381; 1815]
MTTF varies with loading conditions and from part to part.

Predicting service life of an IGBT based on a population MTTF
results in a high uncertainty.

TM

Physics of Failure Based Lifetime
Prediction
Temperature PoF Lifetime Prediction Experiment MTTF
150-200°C 15,300 cycles 6320 cycles
125-225°C 10,800 cycles 1058 cycles
•  Thermo-mechanical fatigue due to variations of power dissipation has

been identified as a failure mechanism of IGBT.
•  Die attach fatigue failure model was used in the CalceFAST software.
The model was based on the Suhir’s interface stress equation coupled
with the Coffin Manson equation.
–  Model inputs were: ∆T, cycling period, materials, and dimensions.
–  Failure criteria were based on separation of die attach material.
•  This model does not represent latchup failures and the actual
degradation involves intermetallic growth which changes the crack
propagation due to brittle fracture.

TM

Limitations of the Die Attach Method
•  Die attach area reduction may not be linear as assumed since
thermal stress is highest in the perimeter and reduces as cracks
move toward the center of the die. Crack growth in the brittle
intermetallic is not the same as the original material.
•  Power dissipation changes with time as efficiency degrades.
•  The latchup Tj is not always 255C due to difference in current
density between operating conditions, metallization
degradation, and chip manufacturing variations.
•  The developed thermal stack model does not represent the
actual thermal resistance network due to unknown spreading
resistance, dissipation through the encapsulant and bond wires,
and changing conductivity through the growing intermetallic.

TM

MIL-217 Handbook: Reliability
Prediction of Electronic Equipment
•  MIL217 Handbook provides formulas to estimate failure rate of military
electronic equipment. Constant failure rate is assumed.
•  No formula was provided for IGBT, therefore a MOSFET and Bipolar
Junction Transistor (BJT) was modeled in series to represent an IGBT.
•  Failure rate is calculated by multiplying a base failure rate with several
conditional factors. For example:
λ p = λbπ T π Aπ Qπ E failures/106 hours

where λP = part failure rate
λb = base failure rate Temperature factor does not
πT = temperature factor account for temperature
λA = application factor cycling input.
λQ = quality factor
πE = environment factor

TM

Comparison of MTTFs
Temperature MIL- Die Attach Experimental Data
Profile HDBK-217 Fatigue Model 2P-Weibull
150-200°C 115,843 hours 15,300 cycles 18.7 hours (6320 cycles)
125-225°C 96,327 hours 10,800 cycles 12.2 hours (1058 cycles)
•  MIL-HDBK-217 method does not account for temperature

cycling loading and other relevant loading conditions.
•  Die attach fatigue model provides a better estimate than the
handbook. Improvement to the model includes obtaining
material fatigue properties, incorporating intermetallic growth
into the crack propagation, and estimation of junction
temperature.
•  Predicting lifetime using a population MTTF cannot account
for variability from part to part.
TM

Motivation for Health Monitoring
Approach (for IGBT and System)
•  Using MTTF to predict IGBT lifetime is not sufficient to avoid
unexpected failures in the field due to the variability in
prediction.
•  Handbook approach ignores relevant loading conditions, device
characteristics, and failure mechanisms leading to erroneous
lifetime predictions.
•  Physics-based lifetime prediction cannot avoid unexpected
failures in the field due to variations from part to part and field
loading conditions.
•  An alternative approach to avoid failures is to monitor IGBT
health individually under operation by using a data-driven
method to analyze the operating data and detect for faulty
conditions before failure occurs.

TM

What We Need to Do?
•  Relevant material properties for the critical failure
mechanisms
•  Ability to update the failure models quickly
•  Modeling platforms for the units and components
•  Life cycle condition information from monitoring

•  Use of data for determination of anomaly at the level
of interest
•  Remaining useful life assessment ability

TM

IGBT Prognostics
•  Patil et. al. [9] IGBTs were monitored for VCE and ICE during continuous
power cycling. Proposed a method to predict remaining useful life (RUL)
of IGBT under power cycling by extrapolating VCE curve to a failure
threshold using particle filter
•  Sutrisno et. al. [10] generated a K-Nearest Neighbor algorithm for fault
detection of IGBTs under continuous power cycling conditions using
monitored electrical characteristics such as VCE and ICE .
[9] N. Patil, “Prognostics of Insulated Gate Bipolar Transistors,” Ph. D. dissertation, Dept. Mech. Eng., University of
Maryland, College Park, MD, 2011.
[10] E. Sutrisno, “Fault Detection and Prognostics of Insulated Gate Bipolar Transistor (IGBT) Using K-Nearest Neighbor
Classification Algoritihm,” M.S. dissertation, Dept. Mech. Eng., University of Maryland, College Park, MD, 2013.

TM

IGBT Prognostics
•  Xiong et al. [11] proposed an online diagnostic and prognostic system to
predict the potential failure of an automotive IGBT power module. A
prognostic check-up routine was implemented that would be activated at a
preset frequency and current during vehicle turn-on and turn-off.
•  Ginart et al. [12] developed an online ringing characterization technique to
diagnose IGBT faults in power drives. Analysis of the damping
characteristic allowed the authors to identify failure mechanisms
[11] Y. Xiong, Xu. Cheng, Z. Shen, C. Mi, H. Wu, and V. Garg, ―Prognostic and Warning System for Power-Electronic
Modules in Electric, Hybrid Electric, and Fuel-Cell Vehicles,ǁ‖ IEEE Transactions on Industrial Electronics, Vol. 55, No.
6, pp. 2268-2276, 2008.
[12] A. Ginart, D. Brown, P. Kalgren and M. Roemer, ―Online Ringing Characterization as a Diagnostic Technique for
IGBTs in Power Drives,ǁ‖ IEEE Transactions on Instrumentation and Measurement, Vol.58, No.7, pp.2290-2299, 2009.
TM

Unclamped Inductive Switching (UIS)
Current Imbalance
•  IGBTs operated with inductive loads can experience voltages
well above their breakdown rating if no voltage clamp is
implemented
•  Voiding and delamination caused by either aging or voiding
leads to current imbalance within the IGBT cells, causing local
heating

TM

Heating within IGBT under UIS Conditions
Unstable behavior observed on die at nominal current localized

heating [13]
[13] M. Riccio, A. Irace, G. Breglio, P. Spirito, E. Napoli, and Y. Mizuno, “Electro-thermal instability in multi-
cellular Trench-IGBTs in avalanche condition: Experiments and simulations,” in Proc. IEEE 23rd Int. Symp.
Power Semiconductor Devices and ICs (ISPSD), May 23–26, 2011, pp. 124–127.

TM

Dynamic Avalanche at Turn-off
•  Similar to UIS conditions, dynamic avalanche can cause
current imbalance between the cells of the IGBT
•  Dynamic Avalanche can be self-induced if the gate resistance
is too low causing high gate currents
Burned emitter contact

pad for discrete IGBT

TM

Failure Mechanisms of Insulated Gate Bipolar Transistors (Igbts)

Uploaded by

Copyright:

Available Formats

Failure Mechanisms of Insulated Gate Bipolar Transistors (Igbts)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Failure Mechanisms of Insulated Gate Bipolar Transistors (Igbts)

Uploaded by

Copyright:

Available Formats

Failure Mechanisms of Insulated Gate

Bipolar Transistors (IGBTs)

Nathan Valentine, Dr. Diganta Das, and Prof. Michael Pecht

Center for Advanced Life Cycle Engineering (CALCE)

2015 NREL Photovoltaic Reliability Workshop

calce Prognostics and Health Management Consortium 1 University of Maryland

Copyright © 2015 CALCE

• The Center for Advanced Life Cycle Engineering (CALCE)

calce Prognostics and Health Management Consortium 2 University of Maryland

Copyright © 2015 CALCE

Induction Heating Units Power Converters Electric Cars

Copyright © 2015 CALCE

calce Prognostics and Health Management Consortium 4 University of Maryland

Copyright © 2015 CALCE

Unused IGBT Failed IGBT which experienced a

calce Prognostics and Health Management Consortium 5 University of Maryland

Copyright © 2015 CALCE

• Quantify the life cycle conditions

calce Prognostics and Health Management Consortium 6 University of Maryland

Copyright © 2015 CALCE

Identify potential failure modes

Identify life cycle profile Identify potential failure causes

Identify potential failure mechanisms

Identify failure models

Prioritize failure mechanisms

Document the process

calce Prognostics and Health Management Consortium 7 University of Maryland

Copyright © 2015 CALCE

calce Prognostics and Health Management Consortium 8 University of Maryland

Copyright © 2015 CALCE

Loss of gate control, device High electric field,

High leakage currents

High temperature, high

High temperature, high Voiding,

calce Prognostics and Health Management Consortium 9 University of Maryland

Copyright © 2015 CALCE

Electromigration Metallizations T, J Eyring (Black)

Conductive Filament Between Metallizations M, ΛV Power Law (Rudra)

Δ: Cyclic range V: Voltage

calce Prognostics and Health Management Consortium 10 University of Maryland

Copyright © 2015 CALCE

calce Prognostics and Health Management Consortium 11 University of Maryland

Copyright © 2015 CALCE

calce Prognostics and Health Management Consortium 12 University of Maryland

Copyright © 2015 CALCE

• IGBT samples were power Switching at 1 or 5 kHz

Copyright © 2015 CALCE

Internal PNP Bipolar Transistor

calce Prognostics and Health Management Consortium 14 University of Maryland

Copyright © 2015 CALCE

Parasitic NPN Bipolar Transistor

calce Prognostics and Health Management Consortium 15 University of Maryland

Copyright © 2015 CALCE

New IGBT sample. Failure to turn on Failure by latchup after

calce Prognostics and Health Management Consortium 16 University of Maryland

Copyright © 2015 CALCE

Bond Wire Cracking Bond Wire Liftoff

calce Prognostics and Health Management Consortium 17 University of Maryland

Copyright © 2015 CALCE

•   The Center for Advanced Life Cycle Engineering (CALCE)

•  Quantify the life cycle conditions

•  IGBT samples were power Switching at 1 or 5 kHz

•  Thermo-mechanical fatigue due to variations of power dissipation has

•  MIL-HDBK-217 method does not account for temperature

•  Life cycle condition information from monitoring