HOT SEARCHES :
Semiconductor Failure Analysis: Root Cause Analysis, Inspection Techniques, and Reliability Engineering
Release Time:
2026-07-24
Source:
www.hsmicroscope.com
Author:
HS Microscope
Learn how semiconductor failure analysis identifies electrical, mechanical, thermal, and material defects using optical microscopy, X-ray, C-SAM, SEM, FIB, EDS, decapsulation, and cross-section analysis.
Quick Answer
Semiconductor failure analysis is a systematic process used to identify the physical location, failure mechanism, and root cause of a defective integrated circuit or semiconductor package.
A complete investigation typically combines:
- Electrical Testing
- Optical Microscopy
- Digital Microscopy
- X-ray Inspection
- Industrial CT
- Scanning Acoustic Microscopy (C-SAM)
- Decapsulation
- Cross-Section Analysis
- Scanning Electron Microscopy (SEM)
- Energy-Dispersive X-ray Spectroscopy (EDS)
- Focused Ion Beam (FIB)
- Transmission Electron Microscopy (TEM)
No single method can identify every semiconductor defect. Effective failure analysis normally progresses from non-destructive inspection to increasingly localized and destructive techniques.
SEM is widely used in microelectronics manufacturing for dimensional analysis, process development, defect characterization, and failure analysis.
What Is Semiconductor Failure Analysis?
Semiconductor failure analysis is the investigation of devices that:
- Fail electrical testing
- Perform outside specification
- Fail reliability qualification
- Stop functioning during operation
- Exhibit intermittent behavior
- Show abnormal physical damage
- Generate unexpected customer returns
The purpose is not only to find where the device failed, but also to determine:
- What failed?
- Where did the failure occur?
- How did the failure develop?
- Why did it occur?
- Which manufacturing or design condition caused it?
- How can the same failure be prevented?
A successful investigation converts an individual failure into actionable information for manufacturing, design, supplier control, and reliability improvement.
Why Semiconductor Failure Analysis Matters
A failed semiconductor device may represent more than an isolated defective component.
It may indicate:
- Wafer fabrication instability
- Contamination
- Lithography defects
- Weak interconnects
- Packaging process problems
- Material incompatibility
- Assembly damage
- Electrical overstress
- Reliability weaknesses
- Design-related vulnerabilities
Failure analysis supports:
- Yield improvement
- Process optimization
- Product qualification
- Customer complaint resolution
- Corrective and preventive action
- Supplier quality management
- Reliability engineering
- New product development
Systematic failure mechanisms must be separated from random defects because recurring mechanisms may affect entire production lots or product families.
Main Categories of Semiconductor Failures
1. Electrical Failures
Electrical failures directly affect device functionality.
Common examples include:
- Open circuits
- Short circuits
- Excessive leakage current
- Abnormal resistance
- Parametric drift
- Timing errors
- Functional failure
- Intermittent electrical connections
Typical causes include broken interconnects, contamination, damaged junctions, poor wire bonds, solder defects, or electrical overstress.
2. Mechanical Failures
Mechanical failures affect the physical integrity of the die or package.
Common examples include:
- Die cracks
- Package cracks
- Edge chipping
- Wire breakage
- Bond lifting
- Substrate fractures
- Package warpage
- Solder fatigue cracks
Mechanical failures frequently develop because of handling stress, vibration, impact, thermal cycling, or coefficient-of-thermal-expansion mismatch.
3. Thermal Failures
Thermal failures are caused or accelerated by excessive temperature or inefficient heat dissipation.
Typical examples include:
- Localized overheating
- Thermal runaway
- Solder fatigue
- Delamination
- Die attach degradation
- Metallization damage
- Burned semiconductor structures
Thermal imaging and electrical testing are commonly used to locate abnormal heat generation before destructive analysis begins.
4. Material Failures
Material-related failures include:
- Corrosion
- Oxidation
- Contamination
- Metal migration
- Intermetallic compound growth
- Plating defects
- Dielectric breakdown
- Adhesive degradation
EDS, SEM, FIB, and cross-section analysis are frequently used to identify the material composition and structure of these defects.
5. Process-Induced Failures
Process-related defects may originate during:
- Wafer fabrication
- Photolithography
- Etching
- Thin-film deposition
- CMP
- Wafer dicing
- Die attach
- Wire bonding
- Flip-chip assembly
- Molding
- Reflow soldering
- Cleaning
Defects, process variations, and design-related factors can all contribute to semiconductor yield loss.
6. Field-Induced Failures
Some semiconductor devices pass production testing but later fail during use.
Possible causes include:
- Electrostatic discharge
- Electrical overstress
- Excessive voltage
- Excessive current
- Moisture penetration
- Thermal cycling
- Mechanical vibration
- Radiation exposure
- Long-term material fatigue
Field failures are especially important in automotive, aerospace, medical, industrial, and power-electronics applications.
Common Semiconductor Failure Mechanisms
Open Circuit
An open circuit interrupts the electrical path.
Possible causes include:
- Broken wire bonds
- Cracked metal traces
- Incomplete solder joints
- Via failure
- Bond lift
- Electromigration
Electrical testing helps locate the affected circuit, while optical microscopy, X-ray, SEM, or FIB can identify the physical defect.
Short Circuit
Short circuits create unintended electrical connections.
Typical causes include:
- Metal bridging
- Solder bridging
- Conductive contamination
- Dielectric breakdown
- Metal migration
- Damaged interconnects
Shorts may be localized using electrical techniques, thermal imaging, emission microscopy, or physical cross-sectioning.
Die Crack
Die cracks may originate during:
- Wafer dicing
- Die picking
- Die attach
- Wire bonding
- Package molding
- Reliability testing
Surface cracks can often be identified using industrial microscopes, while hidden cracks may require C-SAM, X-ray, CT, or cross-section analysis.
Delamination
Delamination is the separation of bonded material interfaces.
It commonly occurs between:
- Die and mold compound
- Die and underfill
- Substrate layers
- Lead frame and mold compound
- Die attach adhesive and die
Scanning acoustic microscopy is widely used to detect delamination and internal package cracking non-destructively.
Wire Bond Failure
Typical wire bond failures include:
- Lifted bonds
- Broken wires
- Heel cracks
- Non-stick on pad
- Non-stick on lead
- Wire sweep
- Pad cratering
Optical microscopy provides rapid visual examination, while pull testing, shear testing, X-ray, SEM, and cross-section analysis provide further verification.
Solder Joint Failure
Common solder-related failures include:
- Voids
- Bridging
- Non-wet opens
- Head-in-Pillow
- Fatigue cracks
- Insufficient solder
- Intermetallic degradation
X-ray is normally used for non-destructive screening, followed by cross-section and SEM analysis when detailed structural evaluation is required.
Electromigration
Electromigration is the movement of metal atoms caused by high current density.
It may create:
- Voids
- Hillocks
- Open circuits
- Short circuits
- Increased resistance
Advanced localization and electron microscopy are often required to evaluate electromigration damage.
Electrostatic Discharge
Electrostatic discharge can damage:
- Gate oxides
- Junctions
- Metal interconnects
- Protection circuits
The damage may be extremely small and require electrical localization, emission microscopy, SEM, FIB, or TEM for confirmation.
Electrical Overstress
Electrical overstress occurs when voltage, current, or power exceeds the device design limit.
Typical physical evidence includes:
- Melted metal
- Burned structures
- Junction damage
- Localized carbonization
- Bond wire damage
Electrical overstress damage is often larger and more thermally destructive than typical ESD damage.
Corrosion
Corrosion may result from:
- Moisture
- Ionic contamination
- Improper cleaning
- Packaging defects
- Harsh operating environments
Optical microscopy, SEM, and EDS are commonly combined to evaluate corrosion morphology and identify contaminating elements.
Semiconductor Failure Analysis Techniques
Electrical Testing
Electrical testing is normally the first analytical stage.
Typical methods include:
- Continuity testing
- Resistance measurement
- Current-voltage testing
- Leakage testing
- Functional testing
- Curve tracing
- Parametric testing
- Time-domain reflectometry
Electrical characterization helps define the failure signature and narrows the suspected failure location.
Optical Microscopy
Industrial and metallurgical microscopes are widely used for initial physical examination.
Typical applications include:
- Surface contamination
- Package cracks
- Die cracks
- Burn marks
- Corrosion
- Wire bond defects
- Bond pad damage
- Metallization defects
- Solder joint surfaces
Optical inspection is fast, non-destructive, and suitable for documenting the device condition before further analysis.
Digital Microscopy
Digital microscopes provide:
- High-resolution image capture
- Dimensional measurement
- Extended depth of field
- Image stitching
- Annotation
- Defect comparison
- Report generation
They are especially useful for customer reports, process documentation, and collaboration between engineering teams.
X-ray Inspection
X-ray inspection reveals hidden structures without opening the semiconductor package.
It can identify:
- Broken wires
- Wire sweep
- Solder voids
- Missing bumps
- Bridging
- Die displacement
- Internal package damage
- Abnormal die attach coverage
Micro-focus X-ray systems are commonly used for BGA, flip-chip, and advanced semiconductor packages.
Industrial CT
Computed tomography generates a three-dimensional representation of internal package structures.
It supports:
- Crack visualization
- Void volume measurement
- Internal alignment evaluation
- Solder joint inspection
- Substrate analysis
- Non-destructive failure localization
Three-dimensional defect detection is increasingly valuable as semiconductor structures become smaller and more complex.
Scanning Acoustic Microscopy
C-SAM uses ultrasonic energy to detect internal interface abnormalities.
Typical applications include:
- Delamination
- Underfill voids
- Die attach defects
- Package cracks
- Moisture damage
- Interface separation
C-SAM is normally performed before decapsulation or destructive sectioning so that the original internal condition is preserved.
Thermal Imaging
Infrared thermal imaging identifies abnormal heat patterns while the device operates.
It can help locate:
- Short circuits
- Leakage paths
- High-resistance connections
- Localized overheating
- Power-distribution abnormalities
Thermal localization often reduces the area that must later be examined by SEM or FIB.
Emission Microscopy
Emission microscopy detects faint light generated by electrically active failure sites.
It can help identify:
- Junction leakage
- Oxide breakdown
- ESD damage
- Latch-up
- Hot-carrier effects
- Abnormal transistor activity
This technique is valuable when the failure is electrical but has no immediately visible physical defect.
Decapsulation
Decapsulation removes the package material to expose the semiconductor die and internal connections.
Methods may include:
- Chemical decapsulation
- Plasma decapsulation
- Laser-assisted opening
- Mechanical removal
The method must be carefully selected to avoid destroying evidence or introducing new damage.
After decapsulation, engineers typically inspect:
- Die surfaces
- Wire bonds
- Bond pads
- Contamination
- Corrosion
- Burned structures
Cross-Section Analysis
Cross-sectioning exposes the internal structure of a device by cutting, grinding, and polishing the sample.
It is used to evaluate:
- Layer structures
- Solder joints
- Bond interfaces
- Die attach
- Cracks
- Voids
- Delamination
- Plating thickness
- Intermetallic compounds
Because the method is destructive, it should normally be performed only after non-destructive inspection and accurate failure localization.
Scanning Electron Microscopy
SEM produces high-resolution images of surfaces and cross-sections.
Applications include:
- Fracture analysis
- Interconnect inspection
- Bond interface evaluation
- Corrosion examination
- Crack morphology
- Solder microstructure
- Process-defect analysis
SEM is considered a core analytical tool for semiconductor defect and failure analysis.
Energy-Dispersive X-ray Spectroscopy
EDS is commonly integrated with SEM to identify the elemental composition of a selected area.
Typical applications include:
- Contamination identification
- Corrosion analysis
- Foreign-material analysis
- Plating verification
- Solder composition evaluation
- Process-residue investigation
EDS results should be interpreted together with imaging, process history, and reference samples.
Focused Ion Beam
FIB uses a focused ion beam to remove material from a highly localized area.
Applications include:
- Precision cross-sectioning
- Buried-defect exposure
- Circuit modification
- Via inspection
- Interconnect analysis
- TEM sample preparation
FIB is particularly valuable when the suspected failure site is too small for conventional mechanical cross-sectioning.
Transmission Electron Microscopy
TEM provides extremely high-resolution analysis of thin samples.
Typical applications include:
- Crystal defects
- Gate oxide failure
- Atomic-scale interfaces
- Nanoscale contamination
- Interconnect structures
- Advanced process-development failures
TEM is generally used only after electrical and physical localization has reduced the target area to a very small region.
Non-Destructive and Destructive Methods
| Method | Type | Typical Application |
|---|---|---|
| Electrical Testing | Non-destructive | Define the failure signature |
| Optical Microscopy | Non-destructive | External defect inspection |
| Digital Microscopy | Non-destructive | Measurement and documentation |
| X-ray | Non-destructive | Hidden structure inspection |
| Industrial CT | Non-destructive | 3D internal analysis |
| C-SAM | Non-destructive | Delamination and void detection |
| Thermal Imaging | Non-destructive | Hotspot localization |
| Emission Microscopy | Usually non-destructive | Electrical defect localization |
| Decapsulation | Destructive | Expose the die and wire bonds |
| Cross-Sectioning | Destructive | Examine internal structures |
| SEM | Usually destructive or semi-destructive | High-resolution analysis |
| FIB | Destructive | Site-specific material removal |
| TEM | Destructive | Nanoscale structural analysis |
Typical Semiconductor Failure Analysis Workflow
1. Collect Background Information
Record:
- Device model
- Lot number
- Manufacturing date
- Test history
- Operating conditions
- Failure symptoms
- Customer application
- Environmental exposure
- Previous reliability results
Incomplete background information can lead to incorrect conclusions.
2. Confirm the Failure
Repeat the reported test under controlled conditions.
Determine whether the failure is:
- Permanent
- Intermittent
- Temperature-dependent
- Voltage-dependent
- Load-dependent
- Environment-dependent
The original failure signature must be preserved throughout the investigation.
3. Perform External Inspection
Use an industrial or digital microscope to document:
- Package condition
- Markings
- Contamination
- Cracks
- Lead damage
- Solder condition
- Signs of overheating
Photograph the sample before cleaning, opening, or sectioning.
4. Perform Non-Destructive Internal Inspection
Depending on the suspected failure, use:
- X-ray
- Industrial CT
- C-SAM
- Thermal imaging
- Emission microscopy
These techniques help identify the probable failure location without destroying evidence.
5. Conduct Electrical Localization
Use electrical measurements to isolate the failed:
- Pin
- Circuit
- Die region
- Power rail
- Interconnect
- Memory cell
- Transistor structure
Electrical localization should guide subsequent physical analysis.
6. Open the Package
Perform controlled decapsulation only after completing the required non-destructive tests.
Avoid damaging:
- Wire bonds
- Die surfaces
- Bond pads
- Corrosion evidence
- Contamination residues
7. Inspect the Die and Interconnects
Use optical microscopy and SEM to inspect:
- Wire bonds
- Metallization
- Passivation
- Burn marks
- Cracks
- Contamination
- Corrosion
- Bond pads
8. Perform Site-Specific Analysis
Use one or more of the following:
- EDS
- FIB
- Mechanical cross-section
- TEM
- Material analysis
The selected method should directly address the suspected failure mechanism.
9. Determine the Root Cause
Differentiate between:
- Physical failure site
- Failure mechanism
- Manufacturing root cause
- Design root cause
- Application-induced cause
For example, a broken wire is the physical failure site, fatigue is the failure mechanism, excessive package vibration may be the operating cause, and insufficient wire-loop support may be the design or process root cause.
10. Implement Corrective Action
Possible corrective actions include:
- Adjusting process parameters
- Improving material selection
- Modifying package design
- Strengthening supplier controls
- Improving cleaning
- Updating handling procedures
- Changing inspection criteria
- Adding reliability screening
- Improving ESD protection
11. Verify Effectiveness
Corrective action must be validated through:
- Repeat testing
- Reliability testing
- Pilot production
- Lot comparison
- Statistical analysis
- Long-term monitoring
A failure analysis is not complete until corrective action effectiveness has been confirmed.
Recommended Equipment
| Investigation Stage | Recommended Equipment |
| External inspection | Industrial Microscope |
| Image documentation | Digital Microscope |
| Surface and metallurgical inspection | Metallurgical Microscope |
| Hidden package structures | X-ray |
| Three-dimensional internal defects | Industrial CT |
| Delamination and interface defects | C-SAM |
| Electrical hotspot localization | Thermal or Emission Microscope |
| Die exposure | Decapsulation System |
| Internal structure evaluation | Cross-Section Equipment |
| High-resolution surface analysis | SEM |
| Elemental analysis | SEM-EDS |
| Site-specific sectioning | FIB |
| Nanoscale analysis | TEM |
Failure Analysis by Defect Type
| Failure Type | Initial Inspection | Advanced Analysis |
| Package Crack | Optical Microscope | SEM + Cross-Section |
| Die Crack | Optical Microscope + C-SAM | SEM + Cross-Section |
| Delamination | C-SAM | Cross-Section |
| Solder Void | X-ray | CT + Cross-Section |
| Wire Bond Failure | Optical Microscope | Pull Test + SEM |
| Electrical Short | Electrical Test + Thermal Imaging | FIB + SEM |
| Electrical Open | Curve Tracing + X-ray | Cross-Section + SEM |
| Corrosion | Optical Microscope | SEM-EDS |
| Contamination | Optical Microscope | SEM-EDS |
| Gate Oxide Failure | Electrical Localization | FIB + TEM |
| Electromigration | Electrical Testing | SEM + FIB |
| ESD Damage | Emission Microscopy | SEM + FIB |
Root Cause Analysis Tools
Five Whys
The Five Whys method repeatedly asks why a failure occurred until the underlying process or design cause is identified.
It is useful for relatively straightforward manufacturing failures.
Fishbone Diagram
A fishbone diagram organizes possible causes into categories such as:
- Material
- Machine
- Method
- Measurement
- Environment
- Human factors
- Design
It is especially useful when several process factors may have contributed to the failure.
Fault Tree Analysis
Fault Tree Analysis starts with the failure event and maps the combinations of conditions that could produce it.
It is commonly used for complex, safety-critical, and reliability-related investigations.
Failure Mode and Effects Analysis
FMEA evaluates potential failure modes based on:
- Severity
- Occurrence
- Detection capability
Failure-analysis findings can be fed back into design and process FMEAs to reduce future risk.
Reliability Testing After Failure Analysis
Reliability tests help reproduce or accelerate suspected failure mechanisms.
Common tests include:
- Temperature Cycling
- Thermal Shock
- High-Temperature Operating Life
- High-Temperature Storage
- Temperature-Humidity Bias
- Highly Accelerated Stress Testing
- Pressure Cooker Testing
- Mechanical Shock
- Vibration Testing
- Drop Testing
- Power Cycling
JESD47 establishes baseline stress-test-driven qualification requirements for integrated circuits, while the JESD22 family includes individual environmental and reliability test methods.
Best Practices
- Preserve the original failure condition before destructive analysis.
- Begin with non-destructive inspection.
- Document every analytical step.
- Use control samples from known-good devices.
- Avoid cleaning the sample before contamination analysis.
- Correlate physical evidence with electrical results.
- Localize the defect before using FIB or TEM.
- Distinguish the failure site from the true root cause.
- Verify corrective actions through reliability testing.
- Build a searchable failure-mode image library.
- Maintain calibration and maintenance records for inspection equipment.
- Use cross-functional teams involving quality, design, process, manufacturing, and reliability engineers.
Common Failure Analysis Mistakes
Destroying Evidence Too Early
Immediate decapsulation or cross-sectioning may remove contamination, alter cracks, or damage internal structures.
Relying on One Inspection Method
A visually obvious defect may not be the electrical failure source. Electrical and physical evidence must be correlated.
Confusing the Failure Site With the Root Cause
The location where damage is found may be the final effect rather than the original cause.
Ignoring Process History
Wafer maps, assembly records, test logs, equipment maintenance, and material lots may reveal systematic patterns.
Failing to Use Reference Samples
Comparing failed devices with known-good samples helps distinguish abnormal structures from normal process variation.
Stopping After Identifying the Defect
The purpose of failure analysis is not only to identify damage. It must also support corrective action and recurrence prevention.
Frequently Asked Questions
What is the first step in semiconductor failure analysis?
The first step is to collect the failure history and confirm the reported electrical or functional symptom. Physical inspection should begin only after the failure signature has been clearly documented.
Why should non-destructive inspection be performed first?
Non-destructive techniques preserve the original sample condition and help localize hidden defects before decapsulation, cross-sectioning, or other destructive procedures.
Can an industrial microscope be used for semiconductor failure analysis?
Yes. Industrial, metallurgical, stereo, and digital microscopes are essential for initial inspection, package evaluation, die-surface examination, wire bond inspection, corrosion analysis, and documentation.
However, internal and nanoscale failures may additionally require X-ray, C-SAM, SEM, FIB, or TEM.
What is the difference between defect detection and failure analysis?
Defect detection identifies an abnormal feature. Failure analysis determines whether that feature caused the electrical or mechanical failure, how the failure developed, and what underlying condition produced it.
What is the difference between failure mechanism and root cause?
A failure mechanism describes the physical process that produced the damage, such as corrosion, fatigue, electromigration, or dielectric breakdown.
The root cause identifies why that mechanism occurred, such as contamination, improper material selection, excessive current density, or incorrect process parameters.
Why is SEM-EDS commonly used together?
SEM provides high-resolution imaging of defect morphology, while EDS identifies the elements present in the selected area. Together, they help investigate contamination, corrosion, plating defects, and foreign materials.
When is FIB required?
FIB is used when a buried or extremely small defect must be exposed at a precise location. It is particularly useful for advanced interconnects, vias, transistor structures, and TEM sample preparation.
Is failure analysis always destructive?
No. Electrical testing, optical microscopy, X-ray, CT, C-SAM, thermal imaging, and emission microscopy are generally non-destructive. Decapsulation, cross-sectioning, FIB, and TEM preparation are destructive.
Related Articles
Semiconductor Inspection
- Wafer Inspection
- Wafer Defect Inspection
- Wafer Edge Inspection
- Die Inspection
- Die Crack Inspection
- Wire Bond Inspection
- Flip Chip Inspection
- Lead Frame Inspection
- IC Package Inspection
Inspection Methods
- Bright Field Microscopy
- Dark Field Microscopy
- Polarized Microscopy
- Differential Interference Contrast
- Optical Microscopy Techniques
Product Guides
- Semiconductor Inspection Microscope
- Metallurgical Microscope
- Industrial Microscope
- Digital Microscope
- Measuring Microscope
- Stereo Microscope
Comparison Guides
- Optical Microscopy vs SEM
- X-ray vs C-SAM
- SEM vs TEM
- FIB vs Mechanical Cross-Sectioning
- Destructive vs Non-Destructive Testing
Defect Library
- Die Crack
- Delamination
- Solder Void
- Wire Bond Failure
- Electromigration
- Corrosion
- ESD Damage
- Electrical Overstress
- Package Crack
- Contamination
Keyword:
Semiconductor reliability testing methods,Electrical failure localization techniques,X-ray and C-SAM semiconductor inspection,Semiconductor decapsulation and cross-section analysis,FIB failure analysis for integrated circuits,SEM analysis of semiconductor defects,IC package root cause analysis,Best microscope for semiconductor failure analysis,Semiconductor failure analysis methods,How to perform semiconductor failure analysis,Semiconductor Reliability Analysis,Electronic Component Failure Analysis,Semiconductor Defect Analysis,IC Failure Analysis,Semiconductor Root Cause Analysis,Semiconductor Failure Analysis