Coating Adhesion Strength Evaluation by Scratch Testing: Principles, Standards, and Applications

The scratch test is one of the most widely used quantitative methods for evaluating the adhesion strength of hard coatings and thin films on metallic and ceramic substrates. By drawing a diamond stylus across a coated surface under progressively increasing normal load, the technique identifies the critical loads at which cohesive and adhesive failures occur. This article explains the technical principles, applicable standards, sample requirements, testing procedures, and typical industrial applications of scratch-based coating adhesion evaluation.

Technical Overview

Scratch adhesion testing is a standardized mechanical method used to quantify the bonding quality between a coating and its substrate. A diamond stylus, typically a Rockwell C cone with a 120-degree included angle and a tip radius of about 200 micrometers, is drawn across the coated surface while the applied normal force increases linearly along the scratch track. As the load rises, stresses accumulate at the coating-substrate interface until the coating fails through cracking, delamination, or complete removal. The normal load at which a defined failure event occurs is recorded as the critical load, which serves as the practical measure of adhesion strength for the coating system.

The method is particularly valuable for hard, brittle, thin coatings deposited by PVD, CVD, electroplating, thermal spray, or anodizing processes, where the coating itself is too rigid or too thin for pull-off or tape-based adhesion tests. International standards such as ISO 20502, ASTM C1624, and EN 1071-3 define the equipment, procedure, and reporting requirements for quantitative scratch testing, and accredited third-party laboratories routinely perform this evaluation for coating development, supplier qualification, and incoming inspection.

During a scratch test, several characteristic failure events are distinguished:

• Lc1 (cohesive failure): the first onset of cracking within the coating itself, detected by acoustic emission spikes or microscopic inspection.
• Lc2 (adhesive failure): the first interfacial delamination or localized spalling between coating and substrate, indicating the practical adhesion limit.
• Lc3 (substrate exposure): the load at which the coating is completely removed along the track and the substrate is fully exposed.

Testing Samples

Sample Requirements

Reliable scratch results depend strongly on specimen geometry and surface condition. Typical requirements include:

• Flat, rigid specimens with a test area of at least 10 mm x 10 mm; common coupon sizes range from 20 mm x 20 mm to 30 mm x 30 mm with a thickness of 2 mm to 10 mm.
• Low surface roughness, generally with Ra well below the coating thickness, so that roughness peaks do not mask interface failure signals.
• Clean, degreased surfaces free of oils, oxides, loose particles, and handling contamination.
• Rigid mounting on the tester stage to prevent vibration or elastic deflection that would distort force and acoustic emission signals.
• Known coating thickness, measured independently by calotest, cross-section microscopy, or X-ray fluorescence, since thickness directly influences critical load values.

Suitable Coating and Substrate Types

The scratch method is best suited to coatings in the approximate thickness range of 0.1 micrometers to 50 micrometers, with the highest sensitivity below about 20 micrometers. Typical systems include:

• Hard PVD and CVD coatings: TiN, TiAlN, AlCrN, CrN, TiC, and DLC on tool steels, cemented carbides, and ceramics.
• Electroplated and electroless plated layers such as hard chromium and nickel-phosphorus on steel and aluminum alloys.
• Thermal spray coatings, anodized layers, and PEO coatings on light alloys.
• Optical and functional thin films, including oxides and nitrides on glass, silicon, and polished metal substrates.

Testing Items

A complete scratch adhesion evaluation typically reports the following parameters:

• Critical loads Lc1, Lc2, and Lc3, expressed in newtons, together with their statistical mean and standard deviation.
• Classification of failure modes along the scratch track, for example conformal cracking, hertz cracking, chipping, buckling, spalling, and coating breakthrough.
• Tangential (friction) force and scratch coefficient of friction as a function of normal load.
• Acoustic emission signal versus applied load, used to pinpoint the onset of brittle fracture.
• Penetration depth and residual scratch width at the critical loads.
• Optical or electron microscopy images and profilometry profiles documenting each failure event.

Testing Methods

Progressive-Load Scratch Test

The most common configuration ramps the normal force from a low initial value (often near 1 N or less) up to a maximum of 30 N to 200 N over a scratch length of 5 mm to 10 mm, at a scratch speed of approximately 5 mm/min to 10 mm/min. This single-stroke approach maps the entire failure spectrum of the coating in one track and is the basis of ASTM C1624 and ISO 20502 procedures.

Constant-Load Scratch Test

For confirmation and repeatability studies, multiple parallel tracks are made at fixed normal loads bracketing the critical load. Constant-load scratching isolates a specific stress level and helps verify whether a failure is reproducible or load-dependent, which improves confidence in the reported adhesion value.

Micro-Scratch Testing

For very thin films below about 1 micrometer, a sharper stylus with a tip radius of 2 micrometers to 50 micrometers and millinewton-level loading is used. Micro-scratch testing, often combined with high-resolution depth sensing, allows adhesion evaluation of optical coatings, metallization layers, and nanostructured films that would fail undetectably under macro loads.

Failure Detection and Data Analysis

Critical loads are determined by the combined interpretation of three complementary signals: sudden changes in the slope of the tangential force curve, bursts in the acoustic emission trace, and direct optical or SEM observation of the scratch track after testing. Most standards recommend performing at least three to five valid scratches per sample on different locations and reporting the test conditions in full, including stylus radius, loading rate, scratch speed, and coating thickness, because critical load values are comparative rather than absolute material constants.

Relevant Standards

• ASTM C1624: quantitative single-point scratch testing of adhesion strength and mechanical failure modes of ceramic coatings.
• ISO 20502: determination of adhesion of ceramic coatings by scratch testing.
• EN 1071-3: methods of test for ceramic coatings, determination of adhesion by the scratch test.
• Corresponding national GB/T series standards adopted for specific coating systems and industries.

Applications

Scratch adhesion evaluation supports quality control, process optimization, and failure analysis across many industries:

• Cutting and forming tools: verifying adhesion of TiAlN, AlCrN, and DLC coatings on end mills, drills, inserts, punches, and molds before production release.
• Automotive and aerospace: qualifying wear-protective and low-friction coatings on fuel injection components, valve train parts, bearings, and landing gear hardware.
• Consumer electronics and optics: assessing durable decorative coatings, anti-reflective layers, and transparent conductive films on glass and polymer substrates.
• Medical devices: evaluating bio-compatible surface coatings on surgical instruments and implantable components.
• Coating development: comparing pretreatment strategies, interlayer designs, deposition parameters, and substrate bias settings during R&D programs.
• Failure analysis: distinguishing adhesive delamination from cohesive wear when investigating premature coating loss in service.

FAQ

Does the scratch test report adhesion strength in megapascals?

Not directly. The primary reported quantity is the critical load in newtons, which integrates the effects of coating hardness, thickness, residual stress, and interfacial bonding. While analytical models can estimate interfacial shear strength from scratch data, the critical load is the internationally recognized comparative adhesion index.

How does scratch testing compare with pull-off and cross-cut methods?

Pull-off tests (ASTM D4541, ISO 4624) and cross-cut or tape tests (ASTM D3359, ISO 2409) are suited to organic paints and thicker, more ductile coatings. The scratch test is preferred for hard, brittle, thin coatings where adhesives cannot bond reliably to the coating or where tape methods produce no measurable failure.

What coating thickness range is appropriate?

Conventional macro-scratch testing covers roughly 0.1 micrometers to 50 micrometers. Very thin films below about 1 micrometer are better evaluated with micro-scratch instrumentation, while very thick soft coatings are usually assessed with tensile or shear methods.

How many scratches are needed for a valid result?

Standards and good laboratory practice typically require a minimum of three to five scratches per specimen at different positions, with the mean critical load and scatter reported. Additional tracks may be necessary when failure events are ambiguous.

What factors influence repeatability?

Stylus tip radius and wear condition, loading rate, scratch speed, substrate hardness, coating roughness, coating thickness uniformity, and friction at the stylus-coating interface all affect results. Consistent fixturing, calibrated force sensors, and identical test parameters are essential for comparing batches or suppliers.

How should samples be prepared and shipped?

Samples should be flat, clean, and rigidly mountable, with coating thickness data supplied. Specimens should be packaged to prevent surface contact damage, and flexible or curved parts should be noted in advance so that suitable test areas or fixturing can be agreed with the laboratory.