Reliability in the More than Moore Landscape

Reliability in the More than Moore Landscape

Loading
Loading Social Plug-ins...
Language: English
Save to myLibrary Download PDF
Go to Page # Page of 44
Important Tags: 

MEMS Reliability

Description: Paper about Reliability in the More than More Landscape, NXP Semiconductors, An R&D powerhouse, NXP Process Technology Research, Society Trends, Intelligent systems: Moore’s Law and Morethan-Moore combined, Sensor and Actuators technologies will move along More than Moore Arrow, Evolution in assembly & packaging, Increased technology integration poses new reliability challenges, Product reliability trend, Acceleration in typical product reliability tests, Classical Failures of Semiconductor.

 
Author: John Schmitz (Fellow) | Visits: 2144 | Page Views: 2162
Domain:  High Tech Category: Semiconductors Subcategory: MEMs 
Upload Date:
Short URL: https://www.wesrch.com/electronics/pdfEL1SE1EZDDMSE
Loading
Loading...



px *        px *

* Default width and height in pixels. Change it to your required dimensions.

 
Contents:
Reliability in the More than Moore Landscape
Dr. John Schmitz, VP Process Technology Research NXP Semiconductors IEEE International Integrated Reliability Workshop October 18 � 2009 S. Lake Tahoe, California

Outline
Introduction NXP Society Trends
� Society Trends and Applications � Changing landscape: More than Moore

More than Moore devices Changing Reliability requirements in More-than-Moore devices Conclusions Acknowledgments

IIRW 2009, John Schmitz, 2

NXP Semiconductors
President & CEO: Rick Clemmer Headquarters: Eindhoven, The Netherlands Net sales: $5.4 billion in 2008 *) Established in 2006 (formerly a division of Philips) 50+ years of experience in semiconductors

Businesses: � Home � Automotive & Identification � Multimarket Semiconductors Owner of NXP Software, a fully independent software solutions company
*) These figures include the Mobile & Personal business which was largely part of the ST-NXP Wireless JV in 2008

IIRW 2009, John Schmitz, 3

An R&D powerhouse
NXP employs around 6,000 engineers to ensure optimal product creation Investment of about $1.2 billion in R&D in 2008 *) 5,700+ patent families 20 R&D centers located in 14 countries

*) These figures include the Mobile & Personal business which was largely part of the ST-NXP Wireless JV in 2008

IIRW 2009, John Schmitz, 4

NXP Process Technology Research
Leuven Belgium Eindhoven Netherlands

Sector Process Technology John Schmitz John Schmitz Applied CMOS Techn. Juffermans RF-ID Low power techn. SSL Si process options Hijzen Embedded memories High voltage/Power RF technology Micro-systems Woltjer MEMS technology Integrated passives Sensor Technology Device modeling & char. Huizing Compact Modelling Characterization Device physics

IMEC

MiPlaza
IIRW 2009, John Schmitz, 5

Outline
Introduction NXP Society Trends
� Society Trends and Applications � Changing landscape: More than Moore

More than Moore devices Changing Reliability requirements in More-than-Moore devices Conclusions Acknowledgments

IIRW 2009, John Schmitz, 6

Society Trends
(from Shell Game Changer) World around us "Infrastructure and environment" 2010 2015
Innovation Business partnering Death of commuting Decentralized power supply

World between us

World within us

"Communication/Data" "Health"
Pervasive connectivity Low cost sensing Voice driven devices Information on demand Secure identity Preventative healthcare Super human performance Travel only for pleasure Perfect sight Healthier mind Home testing/diagnosis Smart drugs Slowing of aging process Average life >90 Self healing/repair Reversal of Alzheimer

2020 2025

Clean energy No domestic waste Intelligent vehicles Cars don't crash

IIRW 2009, John Schmitz, 7

Trends of Potential Importance for Chip Makers
Aging and Health Energy consumption and generation Mobility and transportation Always connected Anywhere Trust and Reliability Robotics in industry and lifestyle Ease, Simplicity and Relevance Personalisation

IIRW 2009, John Schmitz, 8

From "living faster" to "living better"
Productivity era
Making life faster

Value add era
Making life better
(safety, health, environment)

Mainstream markets Many functions in one chip (SoC) Digital, Moore's Law

Niche markets Dedicated functionality "More than Moore"
Analog, power, MEMS, memory, fluidics...

Today: both More Moore and More than Moore are needed
IIRW 2009, John Schmitz, 9

Intelligent systems: Moore's Law and Morethan-Moore combined

Processor Storage

Radio

Power

Sensor Actuator

Moore's Law More than Moore

IIRW 2009, John Schmitz, 10

Sensor and Actuators technologies will move along More than Moore Arrow
More than Moore: Enrichment
Analog/RF Passives HV Power Sensors Actuators Biochips

Baseline CMOS: CPU, Memory, Logic

More Moore: Miniaturization

130nm 90nm 65nm 45nm 32nm 22nm . . . V
Co m bi
Information Processing Digital content System-on-chip (SoC)

ni ng

Sense, Interact, Empower Heterogeneous content System-in-package (SiP)

So C

an d

Si

P:

Hi gh

Va lu eS

ys t

em s

IIRW 2009, John Schmitz, 11

Outline
Introduction NXP Society Trends
� Society Trends and Applications � Changing landscape: More than Moore

More than Moore devices Changing Reliability requirements in More-than-Moore devices Conclusions Acknowledgments

IIRW 2009, John Schmitz, 12

Examples of More than Moore Devices

MEMS devices Biosensor MR sensor RF-ID sensors e-Pill SSL Hybrid car power/battery mgt In Vehicle Network devices

IIRW 2009, John Schmitz, 13

Evolution in assembly & packaging
QFN RF-ID tag Flexible Silicon

Package Diversity

Stacked die Flex Carrier CSP LED lamp

Flip Chip SSOP SO MCM SOT TO DIP DBS QFP HTQFP L/TBGA BGA

Lab-on-a-Chip

Camera Modules
Source: Thesis Willem van Driel, NXP, 2007

1965

1970

1975

1980

1985

1990

2000

2010

IIRW 2009, John Schmitz, 14

Outline
Introduction NXP Society Trends
� Society Trends and Applications � Changing landscape: More than Moore

More than Moore devices Changing Reliability requirements in More-than-Moore devices Conclusions Acknowledgments

IIRW 2009, John Schmitz, 15

Increased technology integration poses new reliability challenges
SiP

Diversity

ip + Biological domain ab-on-a-ch L S) gy (�TA hnolo Tec ystem -S + Chemical domain M ic r o S MEM

+ Mechanical domain IC

ni c s roelectro M ic

Transistor

SoC
1970 1980 1990 2000 2010
Source: Hugo de Man / IMEC

1960

IIRW 2009, John Schmitz, 16

Reliability
Definition: Reliability is the probability that a product will perform according to its specifications during a specified time. Degradation mechanism:
� � � � � Material-interaction-induced Stress-induced Mechanically induced failure Environmentally induced failure etc.....

IIRW 2009, John Schmitz, 17

Bathtub curve
Early failure rate
Failure rate

Intrinsic failure rate Infant mortality Wear-out Useful life Stress related fails

"quality" fails

time

Product reliability trend
Reliability requirements tightened by two decades in 30 years. Number of transistors in an IC increased by 6 to 7 decades (tens to hundreds of millions) over the same time. Reliability per semiconductor element increased by 8 decades in 30 years by:
� 1: reduction of early fails. � 2. Keep wear-out out of specified life time.

Failure rate

time

IIRW 2009, John Schmitz, 19

Acceleration in typical product reliability tests
DHTL Dynamic High Temperature Life SHTL Static High Temperature Life DLTL Dynamic Low Temperature Life HAST Highly-Accelerated Stress Test HTSL High Temperature Storage Life TMCL Temperature Cycling PPOT Pressure Pot THBS Temperature Humidity Bias Static T,V T,V T,V T,H20,V T deltaT T, H20 T, H20,V

Classical Failures of Semiconductor Devices
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. Phenomenon Gate oxide integrity Oxide stability IMD integrity ESD Hot Carrier injection NBTI Latchup Electromigration Stress voiding Junction spiking Passivation integrity Bond degradation Defects
13 4 13

Failure mode gate leakage to source/drain or substrate leakage through parasitic transistors metal to metal or metal to poly short open or short too slow or fast or leaky transistors too slow transistors short between supplies leading to local meltdown opens in metal lines and (less frequent) shorts opens in metal lines leaky substrate diodes exposed metal, contamination ingression, pattern shift disconnect at bondpad opens and shorts
12 8 1 10 5 2 3 6 9

Failure rate

7

11 13

time

Test and addressed mechanisms
test mechanism 1.Gate oxide breakdown 2.Oxide stability 3.IMD integrity 4.ESD 5.Hot Carrier injection 6.NBTI 7.Latchup 8.Electromigration 9.Stress voiding 10.Junction spiking 11.Passivation integrity 12.Bond degradation 13.Defects HAST DHTL SHTL DLTL THBS HTSL TMCL PPOT ESD Latchup

x x x x x x

x x x

x x x

x x x

x x

x

x x

x x

x x x x x x x x x

x

x

x

IIRW 2009, John Schmitz, 22

So Far So Good....... But is this know-how, built up in the last 30 years, 1-1 applicable in the MtM device field?

IIRW 2009, John Schmitz, 23

More Than Moore......
Will ask for quite different reliability requirements in terms of:
� � � � Life time (days � 20 years) Environment (chemicals, body fluids, automotive,..) Conditions (shock, high temperature, EMC, .. ) Mechanical (MEMS, ..)

Two Approaches
� Measures to deal with failures at system/module level (increase resilience)
� The ability to cope with stress and catastrophe � A resilient product or system remains functional during external stress even in the case of internal degradation or malfunction. Example: High speed fault tolerant networks (Flexray IVN)

� Make devices more robust against failure
� Classical approach we know from pure play semiconductor devices � 30 years of reliability research may not be 100% applicable to the new fields

IIRW 2009, John Schmitz, 24

Examples of Areas that ask for Extra Reliability Attention
MEMS:
� We still have limited knowledge on how MEMS devices fail. � Limited tools and models are available. � How to model the reliability of MEMS is a challenge.

Bio-sensors and other sensors
� Many cases require robust shelf lifetime but almost no operational lifetime � Often solid state/liquid interface: chemical resistance against premature attack

PV electronics, LED/drivers
� LED's live much longer than CFL's: driver electronics to survive under harsh temp conditions � PV cells live for 20 years.......

Automotive
� Hybrid cars: 400Volt under the hood conditions � IVN: drive by wire/brake by wire: extreme reliability requirements

Advanced Packaging
� Hermetic packaging for some MEMS devices (out gassing, back end compatible) � Certain sensors need to have an open window to environment � Automotive high temperature (up to 200 C) require other moulding, wiring and bond pads
IIRW 2009, John Schmitz, 25

Example: RF MEMS capacitive switches
Switching function through capacitance change Movable Aluminum structure in PASSITM process
upper electrode / membrane lower electrode spring

upper electrode / membrane

dielectric lower electrode 100 �m gap 3 �m etching hole anchor
IIRW 2009, John Schmitz, 26

RF MEMS C-V curve
Con

VR

Coff

VPI

Hysteresis as a result of position dependent electrostatic force Symmetric because electrostatic force is proportional to V2
IIRW 2009, John Schmitz, 27

RF MEMS reliability
Device should meet reliability specifications:
� Lifetime >10 years operation under harsh conditions.

Test bench to characterize failure mechanisms:
� High electric field stress test � High mechanical stress and crash tests � Pressure cooker test (high temperature and humidity)

Accelerated reliability tests Understand and improve device lifetime MEMS test bench

IIRW 2009, John Schmitz, 28

Spring force and mechanical beam properties

=64.5 MPa

=2.18 GPa

=62.7 GPa

IIRW 2009, John Schmitz, 29

High electric field stress test

Vstress=+50 V
++++++++++++++++++++++++++ ++++++++++++++++++++++++++ -------------------------

0V

V=50 V

C

- - - - - - - - - - - - - - - - - - - - - - - -

0V

F=2/20
0

Vshift

V

++++++++++++++++++++++++++ -------------------------

0V

IIRW 2009, John Schmitz, 30

Time and voltage dependence of Vshift
Vstress

Vshift = t exp( Vstress )

dVshift/d(t1/2) (V/s1/2)

Vstress (V) These results enable accelerated lifetime tests

IIRW 2009, John Schmitz, 31

Electric Force Microscopy (EFM)

++

SiN Vstress=50V
+ ++

Al

+ ++

Apply high voltage stress, then remove top electrode
IIRW 2009, John Schmitz, 32

Electric Force Microscopy
Charge is trapped in SiN dielectric
++ ++

0.5 V

hole +
+ ++

0V

Measure injected charge in SiN dielectric Non-uniform charge distribution
IIRW 2009, John Schmitz, 33

mechanical stress level in spring

Transient characterization impact degradation v
impact
1.0 0.9 0.8 0.7

0.33 m/s 0.50 m/s 0.65 m/s 0.80 m/s 0.75 m/s 0.80 m/s 1.0 1.3 m/s m/s 1.8 m/s
1 10 100 1000 10000 100000 1000000

Vpi / Vpi,0

0.6 0.5 0.4 0.3 0.2 0.1 0.0

1.0 m/s

Nr. of stress pulses

Spring height profile
Height (�m)
4

Before stress After stress 2 �m

2

m)

0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Position (mm)

IIRW 2009, John Schmitz, 34

Closing of the switch seen by interferometer
Each second in this movie is about 4 �s in reality

IIRW 2009, John Schmitz, 35

MEMS Reliability
Field in development New test methods need to be developed Not just electrical testing, but also mechanical testing:
� Acceleration test for accelerometers � Acoustic/pressure test for microphones and pressure sensors � New optical and mechanical tests to characterize failure mechanisms

Physics needs to be understood to improve reliability In particular the micro-mechanical and electrical effects between contacting structures, like adhesion and charging, are challenging

IIRW 2009, John Schmitz, 36

Example: FlexRay (X-by-wire)
Safety-Related high-speed, fault-tolerant Vehicle Systems
X = Safety-related vehicle application (e.g. brake-by-wire, steer-by-wire, etc) Replacement of mechanical/hydraulic parts by high-speed, fault-tolerant electric/electronic systems without conventional backup Implications
� new bus concept: => real-time system => deterministic access => speed beyond 1 Mbps => fault-tolerant � reliability Example: Brake-By-Wire

MISSION Critical Reliability (life critical)!!!!
IIRW 2009, John Schmitz, 37

Example: Biosensor

Changing society
Aging population Less blockbusters drugs More chronic diseases Pressure on healthcare budgets

Changing IVD market
Personalized medicine Increased prevention On-the-spot diagnostics

Need for new technologies
Semiconductor biochips, label-free, miniaturisation, wireless, portable, ease of use, ....

NXP Biosensors Platform
IIRW 2009, John Schmitz, 38

What are Biosensors ?
Biosensor breakdown
Detection signal Binding event Analyte Non-specific analyte Sample matrix Output or analysed signal Biological receptors Transducer or detection system

A biosensor is a device for the detection of an analyte that combines a biological component with a physicochemical detector component
IIRW 2009, John Schmitz, 39

Biosensor Reliability
Electrodes

Guaranteed shelf life time
� Surface preparation and sealing � Self diagnostics

Modified

Only minutes life time required during actual usage conditions New protocols needed how to do product qualifications
500 nm

Standard

Transistors

MISSION Critical Reliability (life critical)!!!!
IIRW 2009, John Schmitz, 40

Outline
Introduction NXP Society Trends
� Society Trends and Applications � Changing landscape: More than Moore

More than Moore devices Changing Reliability requirements in More-than-Moore devices Conclusions Acknowledgments

IIRW 2009, John Schmitz, 41

Conclusions
Trends in society raise demand for IC solutions in areas like health, security and the environment ICs will more often be integrated with chemical, biological and mechanical devices to provide solutions that can be used at any time and at any place (in vivo/vitro, indoor/outdoor, car etc) Because life critical factors can be at stake in these new fields than in consumer electronic performance, the requirements for a reliable performance of these applications can differ widely from that what we accept today. 30 years of experience in SoCs is not enough to handle all reliability issues in More-than-Moore applications: new approaches at system, module and device level are required!
IIRW 2009, John Schmitz, 42

Acknowledgements
Fred Kuper Peter Steeneken Guoqiao Tao Reinout Woltjer Caroline Beelen-Hendrikx Rene M Jansen

IIRW 2009, John Schmitz, 43

Thank you

IIRW 2009, John Schmitz, 44

Subscribe
x