1、JEDEC PUBLICATION Solid-State Reliability Assessment and Qualification Methodologies JEP143C (Revision of JEP143B.01, June 2008) JULY 2012 JEDEC SOLID STATE TECHNOLOGY ASSOCIATION NOTICE JEDEC standards and publications contain material that has been prepared, reviewed, and approved through the JEDE
2、C Board of Directors level and subsequently reviewed and approved by the JEDEC legal counsel. JEDEC standards and publications are designed to serve the public interest through eliminating misunderstandings between manufacturers and purchasers, facilitating interchangeability and improvement of prod
3、ucts, and assisting the purchaser in selecting and obtaining with minimum delay the proper product for use by those other than JEDEC members, whether the standard is to be used either domestically or internationally. JEDEC standards and publications are adopted without regard to whether or not their
4、 adoption may involve patents or articles, materials, or processes. By such action JEDEC does not assume any liability to any patent owner, nor does it assume any obligation whatever to parties adopting the JEDEC standards or publications. The information included in JEDEC standards and publications
5、 represents a sound approach to product specification and application, principally from the solid state device manufacturer viewpoint. Within the JEDEC organization there are procedures whereby a JEDEC standard or publication may be further processed and ultimately become an ANSI standard. No claims
6、 to be in conformance with this standard may be made unless all requirements stated in the standard are met. Inquiries, comments, and suggestions relative to the content of this JEDEC standard or publication should be addressed to JEDEC at the address below, or refer to www.jedec.org under Standards
7、 and Documents for alternative contact information. Published by JEDEC Solid State Technology Association 2012 3103 North 10th Street Suite 240 South Arlington, VA 22201-2107 This document may be downloaded free of charge; however JEDEC retains the copyright on this material. By downloading this fil
8、e the individual agrees not to charge for or resell the resulting material. PRICE: Contact JEDEC Printed in the U.S.A. All rights reserved PLEASE! DONT VIOLATE THE LAW! This document is copyrighted by JEDEC and may not be reproduced without permission. For information, contact: JEDEC Solid State Tec
9、hnology Association 3103 North 10th Street Suite 240 South Arlington, VA 22201-2107 or refer to www.jedec.org under Standards-Documents/Copyright Information. JEDEC Publication No. 143C -i- SOLID-STATE RELIABILITY ASSESSMENT AND QUALIFICATION METHODOLOGIES Introduction Reliability is a measure of th
10、e ability of a product to continue to meet data sheet specifications over time. In any population of solid-state products there is a range in the time to failure, which is the time-to-failure distribution. Since the typical time-to-failure distribution will have a central characteristic in the order
11、 of hundreds or thousands of years, means must be derived to assess the time-to-failure distribution in real time. Failure mechanism models provide a physical and mathematical representation of how the time-to-failure distribution will change by changing the influencing factors from application cond
12、itions. Reliability assessment and qualification systems consist of the following: a) a list of applicable failure modes, failure mechanisms, failure kinetics, and causes; b) a failure model, including as appropriate the validity and correctness of the model; c) a means to accelerate the failure pro
13、cess; d) defined application conditions (mission profile) to generate a list of criteria for the acceptability of the failure rates; e) the definitions and/or criteria for a failure, e.g., when is a deviation a failure. Reliability assessment systems provide economic benefit for both manufacturers a
14、nd users. Manufacturers can quantify the reliability of their products in order to understand warranty and other costs associated with potential reliability failures. Users can estimate the reliability of systems using solid-state products in order to quantify costs associated with system operation
15、or failures. Reliability assessment systems consist of a suite of accelerated stresses and the associated sample plans used for monitoring or acceptance. The statistics used to determine the sample plans are not covered herein. Note: the sample plans should be based on the purpose of the sampling, e
16、.g., monitoring or qualification, the expected failure rate, and the expected application requirements. References such as Sampling Inspection Tables Single and Double Sampling by Harold F. Dodge and Harry G. Romig; Some Theory of Sampling by Willian Edwards Deming; Statistical Method from the Viewp
17、oint of Quality Control by Walter A. Shewhart; Guide to Quality Control by Kaoru Ishikawa, and A Sampler on Sampling by Bill Williams provide both the theoretical background and practical applied approaches for choosing appropriate sample plans. Accelerated stress methods are usually categorized by
18、the location of the failure mechanisms they are expected to stimulate. The stress methods are normally classified as die-related, package-material-related, or packaged-product-related, including the interconnection of die and package. Eventually there may be an additional category and section within
19、 this publication for stress methods associated with the failure mechanisms of the interface between either the package and the socket or the package and the printed circuit board (PCB). This publication pertains primarily to integrated circuits; however, that does not prohibit adding additional cat
20、egories and sections for stress methods that include system-level failure mechanisms. Stress methods are also categorized by the cause of failure, e.g., process variance, defects, or wearout. The stress factor, e.g., temperature, electric field, humidity, mechanical stress, voltage, or current densi
21、ty, may also be used for categorization. JEDEC Publication No. 143C -ii- Introduction (contd) While reliability modeling has been around for more than a century, for integrated circuits the seminal paper “Reliability in the Bell System” published in Proceedings of the IEEE, February 1974 by Peck R0i
22、s a constant; Eais the activation energy (eV); k is Boltzmanns constant (8.62 105eV/K); T is the temperature (K) Since we are looking for the changes in the failure time (i.e., the probability-density function of time-to-failure), we use the common equation for the temperature acceleration factor: A
23、T= T1/T2= exp(Eaa/k)(1/T1 1/T2) where AT is the acceleration factor due to changes in temperature; T1is the observed failure rate at test temperature T1 (h-1); T2is the observed failure rate at test temperature T2 (h-1); Eaais the apparent activation energy (eV); k is Boltzmanns constant (8.62 105eV
24、/K); T1is the temperature of test 1 (K); T2is the temperature of test 2 (K) JEDEC Publication No. 143C -iii- Introduction (contd) The temperature acceleration factor is derived from the Arrhenius equation for failure rate, and may be calculated from the equation: T= 0exp (Eaa/kT) where Tis the failu
25、re rate at temperature T; 0is the base failure rate (h-1); Eaais the apparent activation energy (eV); k is Boltzmanns constant (8.62 105eV/K); T is some temperature (K) NOTE The Arrhenius relationship is for temperature accelerated failure mechanisms, whether the failure rate increases or decreases
26、with temperature. Some failure mechanisms have stronger dependencies on other factors (e.g., voltage, humidity) so the Arrhenius relationship may not be applicable. Various population failure distributions, called probability density functions (PDF) and cumulative distribution functions (CDF), are m
27、athematical models of failure rates. These distributions are specifically the cumulative distribution function of the time-to-failure (cumulative mortality function), cumulative hazard function, cumulative reliability function, instantaneous failure rate, or probability density function of time-to-f
28、ailure (mortality function) depending on what failure rate is intended to be modeled and described. The simplest way of visualizing reliability as a function of time is a graph. The most common graph is the bathtub curve showing the areas of infant mortality, useful life, and wearout. The reliabilit
29、y is quantified in units of FITs, MTTF (Mean-Time-To-Failure), or MTBF (Mean-Time-Between-Failures). Failure rates are experimentally determined by stressing devices. The observed failure rate from stress is converted to the quoted failure rate predicted for field applications using an acceleration
30、factor calculated using the Arrhenius Equation for reliability and an apparent activation energy estimated from reliability stress results at different conditions. While the Arrhenius Equation for reliability is most often used to model temperature acceleration of the failure rate distribution, the
31、equation can be used for other stress factors, e.g., voltage, humidity. JEDEC Publication No. 143C -iv- JEDEC Publication No. 143C Page 1 SOLID-STATE RELIABILITY ASSESSMENT AND QUALIFICATION METHODOLOGIES (From JEDEC Board Ballot JCB-12-02, formulated under the cognizance of the JC-14.3 Subcommittee
32、 on Silicon Devices Reliability Qualification and Monitoring.) 1 Scope This publication applies to all integrated circuits and their associated packages. The document summarizes the suite of reliability documents and publications available. These documents address reliability qualification, reliabil
33、ity stress testing, and reliability modeling. The purpose of this publication is to provide an overview of some of the most commonly used systems and test methods historically performed by manufacturers to assess and qualify the reliability of solid-state products. The appropriate references to exis
34、ting and proposed JEDEC or joint standards and publications are cited. This document is also intended to provide an educational background and overview of some of the technical and economic factors associated with assessing and qualifying microcircuit reliability. 2 Terms and definitions The followi
35、ng terms and definitions are fundamental in the field of reliability. Virtually all of the terms have been defined before but some were merely used without definition in one or more of the publications listed in clause 9. Some terms and/or definitions have been revised technically and/or grammatical
36、ly and are not the same as currently exist in other publications. The updating of publications, including JESD88, JEDEC Dictionary of Terms for Solid State Technology, is ongoing. The intention is that the other publications will be aligned with the terms and definitions presented herein as those pu
37、blications are revised. Terms and definitions associated with failure mechanisms and reliability models can be found in JEP122. If not otherwise noted, in general for the below definitions, when one sees “time” or “time-to-failure”, the definitions also apply for “cycles” or “cycles-to-failure”, or
38、other measure of the duration measurement required for failure. Repeating the alternate terms is not necessary to convey understanding. JEDEC Publication No. 143C Page 2 2 Terms and definitions (contd) acceleration factor (A, AF): For a given failure mechanism, the ratio of the time it takes for a c
39、ertain fraction of the population to fail, following application of one stress or use condition, to the corresponding time at a more severe stress or use condition. NOTE 1 Times are generally derived from modeled time-to-failure distributions (lognormal, Weibull, exponential, etc.). NOTE 2 Accelerat
40、ion factors can be calculated for temperature, electrical, mechanical, environmental, or other stresses that can affect the reliability of a device. NOTE 3 Acceleration factors are a function of one or more of the basic stresses that can cause one or more failure mechanisms. For example, a plot of t
41、he natural log of the time-to-failure for a cumulative constant percentage failed (e.g., 50%) at multiple stress temperatures as a function of 1/kT, the reciprocal of the product of Boltzmanns constant in electronvolts per kelvin and the absolute temperature in kelvins, is linear if one and only one
42、 failure mechanism is involved. The best-fit linear slope is equal to the apparent activation energy in electronvolts. NOTE 4 The abbreviation AF is often used in place of the symbol A. acceleration factor, stress (Af): The acceleration factor due to the presence of some stress (e.g., current densit
43、y, electric field, humidity, temperature cycling). acceleration factor, temperature (AT): The acceleration factor due to changes in temperature. NOTE 1 This is the acceleration factor most often referenced. The Arrhenius equation for reliability is commonly used to calculate the acceleration factor
44、that applies to the acceleration of time-to-failure distributions for microcircuits and other semiconductor devices: AT = T1/T2 = exp(Eaa/k)(1/T1 1/T2) where Eaa is the apparent activation energy (eV); k is Boltzmanns constant (8.62 105 eV/K); T1 is the absolute temperature of test 1 (K); T2 is the
45、absolute temperature of test 2 (K); T1 is the observed failure rate at test temperature T1 (h-1); T2 is the observed failure rate at test temperature T2 (h-1). NOTE 2 The best-fit linear slope of a plot of the natural log of the time-to-failure as a function of 1/kT, the reciprocal of the product of
46、 Boltzmanns constant in electronvolts per kelvin and the absolute temperature in kelvins, is equal to the apparent activation energy in electronvolts. NOTE 3 q = o AT, where q is the quoted (predicted) system failure rate at some system temperature Ts, o is the observed failure rate at some test tem
47、perature Tt, and AT is the temperature acceleration factor from Tt to Ts. acceleration factor, voltage (AV): The acceleration factor due to changes in voltage. acceleration model: A mathematical formulation of the relationship between (1) the rate (speed) of a degradation mechanism or the time-to-fa
48、ilure and (2) the conditions or stresses that caused the degradation. JEDEC Publication No. 143C Page 3 2 Terms and definitions (contd) activation energy (Ea): The excess free energy over the ground state that must be acquired by an atomic or molecular system in order that a particular process can o
49、ccur. NOTE The activation energy is used in the Arrhenius equation for the thermal acceleration of physical reactions. The term “activation energy” is not applicable when describing thermal acceleration of time-to-failure distributions, e.g., in the Arrhenius equation for reliability, hence the need for the term “apparent activation energy”. apparent activation energy (Eaa): An energy value, analogous to activation energy, that can be inserted in the Arrhenius equation for reliability to calculate an acceleration factor applicable to changes
