POWER Electronics System Failure (4)

POWER Electronics Devices Failure:

One of the Biggest Problems that lead to Power Electronics System Failure is the Failure of Power Electronics Devices or Power Semiconductors Devices.

In the Last Article, we talked about the 6 most Common Failure Mechanisms for Power Electronics Devices [Fatigue - Electromigration - Stress Induced Voiding - Corrosion - Conduction Filament Formation - Time Dependent Dielectric Breakdown]

In this Article, we will Provide more information about these Physical Failure Mechanisms that depend on the Lifetime & Conditions of Semiconductor Devices, and not Failure Mechanisms resulting from Design Errors, or resulting from incorrect use of the Power Electronics System.

1-Fatigue:

Fatigue Failures of the materials of Power Electronics Devices are accelerated because Power Electronics Devices always suffer Continuous Variations of Temperature & Stress in their long-term Operations. Meanwhile, with the development of Renewable Energy Technology, the demand of Power Density of Power Electronics Converters is becoming Higher and Higher. It presents a New Challenge for Designing High-Reliability Power Electronics Devices.

Therefore, intensive Studies of Fatigue Failure Mechanism of Power Electronics Devices are important to the optimizing Design of Packages, Material Selection, and Manufacturing Technique, and also the premise of Designing High-Reliable Power Electronoics Devices. Physics-of-Failure (PoF) is a vital method to Study the Fatigue Failure, which can predicate the Reliability via Modeling and Simulating based on the Failure Analysis Process & Mechanism.

2-Electromigration:

In an Electric Field atoms are subjected to a Force due to the Field, and to a Force which results from the motion of Electrical Carriers, Electrons, or Holes. In bulk Samples, and at High-Temperatures, these Forces cause the displacement of atoms by a Lattice Mechanism which is also responsible for the diffusion of atoms in a concentration gradient. In thin films, Electromigration has been found to occur at Lower Temperatures (and Higher Current Densities) by a Grain Boundary Diffusion Mechanism. Electromigration may cause Failures in Material Discontinuities, such as found at Terminals, at Temperature Gradients, or at Structural Inhomogeneities. The Crack Formation Process, as observed in Aluminum Thin Films, is described. The Effects of Film Purity, Orientation, Grain Size, Glass Overcoat, and Solute Additions on Lifetime are reviewed. Practical guidelines for the Design of Thin-Film Interconnections, and for the interpretation of accelerated test data are given.

3-Corrosion:

Corrosion on Electronic & Power Electronic Devices is one of the major Effects, which negatively affect the Reliability of the Respective Device. The ZESTRON Experience from over many years of Failure Analysis and Risk Assessment in this area clearly reveal that in Power Electrics, especially the Anodic Migration Phenomenon (AMP) is the Corrosion Mechanism most often found under High-Voltage Conditions in Combination with High-Humidity Load. This Study shows that the AMP is a deviation from the “Basic Mechanism” of Electrochemical Migration, which is commonly known from Low-Voltage Applications.

4-Conductive Filament Formation:

In this Section, a Failure Mechanism driven by the Combined Presence of an Applied Electric Field and Liquid Medium known as CFF is reviewed. This Failure Mechanism was first observed in 1955, by researchers at Bell Laboratories on Silver. In their investigation, Silver in contact with Insulating Materials was found to be Removed from the Anode and deposited in a different Location, which resulted in Breakdown of the Insulating Material. The Failure Mode for this Failure Mechanism includes Formation of Shorts that Grows through the substrate and between Conductors of same Size as a Substrate. Generally, CFF is known to occur in two steps: Organic Material Delamination and Moisture Absorption at the interface and Metal Migration leading to Loss of Insulation Resistance.

5-Stress Induced Voiding Formation:

Stress Induced Voiding (SIV) is a key Reliability issue common to Line Interconnect. The Induced Stress occurs due to Processing Activities associated with both Single & Multi-Layered Interconnect System. Common Sources of these Stresses on Metal Line include the Deposition of the inter-Level Dielectric and the Subsequent Cooling, CTE mismatch associated with the Dielectric Material and the Metal following Thermal Treatment, as well as other Thermal Treatments that may have an influence on the Stress State of the Metal. The main Driving Force for SIV is Stress Gradient. Void Formation and Growth Effect in an Interconnect depends on the Location. Owing to the sudden open Failure related to voids below the via, these Failure Modes are referred to as SIV, while the Partial Resistance rise at the via is often called Stress Migration. This Characteristic is Experimentally Correlated with Metal Geometry. Over the years, many Physics-based Models have been developed to facilitate Reliability Prediction for SIV Failure Mechanism.

6-Time Dependent Dielectric Breakdown:

Time Dependent Dielectric Breakdown (TDDB) is a Failure Mechanism whereby Breakdown takes place, often below the Breakdown Strength of the Material (dielectric) with Time when Stored at a Non-varying Electric-Field (e-field). In the present of Electric-Field, several Conduction Mechanisms have been reported to facilitate TDDB. [Ionic Conduction, Direct Tunneling, Schottky Emission, Space-Charge-Limited Conduction, Poole–Frenkel (PF) Emission, Ohmic Conduction, and Fowler–Nordheim (FN) Tunneling] were reported as the common Conduction Mechanisms. Generally, TDDB Reliability Models are developed based on either current or field induced Degradation collected data.

A Critical Observation from this Article is that most Modeling efforts only Focused on a Single Failure Mechanism. In practice, Components may Experience Combined Failure Mechanism. For instance, EM and Fatigue. Models Based on such Combine Failure Mechanism may be more Accurate from a Reliability Prediction perspective.

Integrated Reliability Models based on data-driven approach, though still developing show great potential for Real-time Power Electronics Reliability Prediction.

Miled Kafrouni

MKafronics Instruments

#Power_Electronics_Failure