Bonded Joint & Adhesive Failures

What Is Bonded Joint Failure?

Bonded joints transfer load through an adhesive layer between two substrates. Unlike bolted joints, which transfer load through mechanical fasteners, bonded joints rely on surface adhesion and the cohesive strength of the adhesive. Bonded joint failures can occur in the adhesive itself (cohesive failure), at the adhesive-substrate interface (adhesive failure), or in the substrate material (substrate failure). The failure path reveals the weak link in the joint and points to the root cause.

THE LOCATION OF A HIGH ANALYTICAL PEEL STRESS DOES NOT BY ITSELF IDENTIFY THE ROOT CAUSE OF A BOND FAILURE. The analysis identifies where stress is high, but the root cause may lie in surface preparation, cure, contamination or environment.

Failure Paths

A bonded joint can fail in three principal locations. Each failure path has different implications for the root cause investigation.

Peel and Shear Loading

Bonded joints are strong in shear but weak in peel. Shear loading distributes stress through the adhesive layer, allowing a large bond area to contribute. Peel loading concentrates stress at the bond edge, where the adhesive is loaded in tension through a thin section — the weakest direction. Joints designed to load the adhesive in shear are far more robust than joints that introduce peel. Peel stresses are highest at the edges of the bondline and at points where the substrate stiffness changes.

• Shear loading — stress distributed through the adhesive; efficient load transfer
• Peel loading — stress concentrated at the edge; adhesive loaded in its weakest direction
• Mixed mode — combination of peel and shear; common at bondline edges
• Cleavage — opening mode at the bond edge, similar to peel but with a prying action

Stress Singularities at Bondline Edges

At the edges of a bonded joint, the stress field is theoretically singular — the stress approaches infinity at the sharp corner where the adhesive meets the free surface. In linear elastic analysis, this produces a stress singularity that cannot be used as a failure prediction directly. The maximum stress at the singularity depends on the mesh refinement — it increases as the element size decreases. This is a mathematical artefact, not a physical prediction.

A numerical stress singularity at a bondline edge is not a physical failure prediction. The stress concentration is real, but the infinite stress from a linear elastic model is a mathematical artefact of the sharp corner. Use cohesive-zone modelling, fracture mechanics or a characteristic-distance approach instead.

Manufacturing Factors

The strength and durability of a bonded joint depend critically on the manufacturing process. A joint that is analytically adequate can fail in service because the manufacturing conditions were not correct. In a failure investigation, the manufacturing records and the fracture surface evidence must be examined to determine whether the joint was produced correctly.

• Surface preparation — the most critical factor; inadequate cleaning, abrasion or treatment produces weak interfaces
• Bondline thickness — too thin gives poor toughening; too thick introduces defects and reduces strength
• Cure — incorrect temperature, time or pressure produces incomplete cure and reduced properties
• Voids and porosity — trapped air or volatiles reduce the effective bond area
• Contamination — oil, grease, moisture or release agent on the surface prevents adhesion
• Resin-rich / resin-starved regions — uneven adhesive distribution

Environmental Effects

Adhesive bonds are sensitive to environmental conditions. Moisture absorption can plasticise the adhesive, reducing strength and stiffness. Elevated temperature can reduce the glass transition temperature and dramatically weaken the adhesive. Thermal cycling can produce differential expansion stresses at the bondline. Chemical exposure can degrade the adhesive or the interface. These effects are progressive — a joint that is strong at manufacture may degrade over years of service exposure.

Cohesive-Zone Modelling

Cohesive-zone modelling (CZM) is an advanced analysis technique for bonded joints. Instead of modelling the adhesive as a continuum and computing stress, CZM uses a traction-separation law that describes how the adhesive transfers load as the bond opens or shears. The law includes a maximum traction (strength) and a separation at which the bond fails (toughness). CZM can predict both crack initiation and propagation, and it naturally handles the stress singularity at the bond edge by working with energy rather than stress. CZM requires material-specific calibration data — traction-separation parameters measured from tests.

ANALYSIS CONSIDERATION: Cohesive-zone modelling requires material-specific traction-separation parameters. Using generic values without calibration to the actual adhesive and substrate system produces unreliable predictions.

Investigation of Bond Failures

When investigating a bonded joint failure, the fracture surface is the primary evidence. The failure path — cohesive, adhesive or substrate — reveals the weak link. The surface morphology, contamination traces and adhesive condition provide further information. The manufacturing records should be reviewed for surface preparation, cure and bondline thickness. The service environment should be assessed for moisture, temperature and chemical exposure. The analysis should use as-built geometry and actual material properties — not the idealised design values.

1. Examine the fracture surface — determine failure path (cohesive, adhesive, substrate)
2. If adhesive failure — investigate surface preparation and contamination
3. If cohesive failure — investigate adhesive properties, cure and bondline thickness
4. If substrate failure — the adhesive is not the root cause; investigate the substrate
5. Review manufacturing records — surface treatment, cure cycle, bondline control
6. Assess environmental exposure — moisture, temperature, chemicals
7. Perform analysis using as-built geometry and actual properties
8. Compare predicted failure location and mode with observed evidence

Key Takeaways

• Bonded joints can fail cohesively (in the adhesive), adhesively (at the interface) or in the substrate
• The failure path identifies the weak link and directs the investigation
• Bonded joints are strong in shear but weak in peel — peel stress at edges is the critical design concern
• Surface preparation is the most critical manufacturing factor — inadequate preparation produces weak interfaces
• Moisture, temperature and chemical exposure can degrade bond strength over service life
• Stress singularities at bondline edges are mathematical artefacts — use CZM or fracture mechanics instead
• Cohesive-zone modelling requires calibrated traction-separation parameters for the specific adhesive system