Solder Mask Adhesion: Cure Verification and Tape Test Methods

A solder mask that is under-cured looks perfect until the board reaches a soldering temperature or a cleaning bath, and then it lifts from the copper or blisters off a trace. Cure and adhesion are the same question asked twice, and both are verified with cheap tests that most lines already have the equipment to run.

Why Cure and Adhesion Are One Question

A photoimageable mask is an epoxy or acrylic system that is cured in two stages, first by ultraviolet light through the artwork and then by heat. The ultraviolet stage fixes the pattern, and the thermal stage completes the cross-linking that gives the film its chemical resistance and its grip on the copper beneath.

A film that has been fully exposed but under-baked still holds its shape, so the board passes visual inspection and pattern checks. The deficiency appears as loss of adhesion during thermal stress, and it is measured with the same board quality tests that are used for a cosmetic defect.

Cure Schedules and What They Deliver

A typical thermal cure for a liquid photoimageable mask is 60 minutes at 150 degrees Celsius, though the exact schedule belongs to the material and is quoted by the supplier with a tolerance. The board has to reach the temperature rather than the oven air, and a heavy copper plane or a thick panel lengthens the time to reach it.

Dual cure systems add a final ultraviolet flood after the thermal step to cure the areas that the artwork shadowed, such as the sidewalls of a via or a region under a wide trace. Where that step is omitted, the shadowed areas remain tacky and are the first places to fail a solvent rub.

Measuring Cure in Production

The usual production check is a solvent rub with methyl ethyl ketone, commonly twenty five double rubs across the surface with a swab under defined pressure, and the mask passes if the film is not broken through. The test is comparative: a known good sample is rubbed alongside the production panel so the operator has a reference.

Qualification uses instrumentation instead, with differential scanning calorimetry to measure the glass transition and residual exotherm, and infrared spectroscopy to follow the disappearance of the reactive bands. Those methods belong with a material change rather than with a shift check, because they need a laboratory and a baseline.

The Tape Test and How It Is Run

The tape test for mask adhesion is a cross-hatch performed to IPC-TM-650 method 2.4.28.1, cutting a grid of lines one millimetre apart with a blade held at a defined angle and pressing a length of tape over the grid. The tape is then pulled back at about 180 degrees in one movement, and the grid is examined for removal.

The pass criterion is no removal of the film between the lines and no lifting at the cut edges. A result that is marginal usually means the cut went through the copper rather than through the mask, so the cut depth is part of the method and not an operator preference.

Surface Energy and Wettability

Adhesion depends on the mask wetting the copper before it is cured, and that is controlled by the surface energy of the copper. A dyne pen drawing a continuous line at 32 to 38 millinewtons per metre indicates a surface that will wet; a line that beads or breaks indicates contamination.

The contamination is usually organic, left by a rinse that was not effective or by handling before the mask was applied, and it can be removed by a micro-etch or a plasma step. Our assembly cleanliness notes describe the same measurement from the assembly side of the process.

Solder Float and Thermal Stress

The thermal stress test floats a masked sample on molten solder at 288 degrees Celsius for ten seconds, and the mask is examined for blistering, lifting and delamination. It is defined in IPC-TM-650 method 2.4.13, and it is the check that most closely reproduces what happens to the board during assembly.

Cracks at the edges of a trace and blisters over a plane usually indicate that the escape of moisture or solvent was blocked, which points to a cure that ran too fast or to a rinse that left chemistry under the film. Both are process faults rather than material faults, and both are fixed upstream of the test.

Chemical Resistance Checks

The mask sees flux, cleaning chemistry and, in some products, a conformal coating solvent, so the resistance is checked against the specific chemicals the product will meet rather than against a generic list. A common check is immersion in the flux or the cleaner at its working temperature for a defined time followed by a tape test.

The results are recorded per chemistry and per exposure time, which makes it possible to tell whether a failure follows the mask or the cleaning process. Our conformal coating notes describe how the coating compatibility is checked on top of the same evidence.

Process Variables That Break Adhesion

The variables that matter are copper surface condition, rinse effectiveness, exposure energy, cure temperature and cure time, and thickness of the mask itself. A mask applied too thickly cures from the outside inwards and stays soft at the interface, while one applied too thinly can be abraded through in handling.

Tented vias add a further risk, because the film bridges a hole and is unsupported over it. The tent is the first feature to lift under thermal stress, and where the product cannot tolerate that, the via is left open or filled and capped instead of tented.

Sampling, Records and Acceptance

The tape test is run per lot on a coupon that carries the same mask and copper as the product, and the solvent rub is run per shift as a trend. A failure on the coupon stops the lot, because a mask that fails on a flat coupon will not survive assembly on a board that is more highly loaded.

The record should carry the cure schedule actually run, the oven loading, the exposure energy and the test results with the pass criterion used. Without the schedule, a later adhesion problem cannot be traced to the process, and our fabrication notes treat the schedule as part of the drawing package rather than as a shop setting.

FAQ

Is a fully cured mask always better adhered? Up to a point. Over-curing makes the film brittle and can cause it to crack at the edges of traces during thermal stress, so the schedule has a window rather than a direction, and the upper end is as important as the lower one.

Why does the tape test sometimes pass on a board that fails in the field? Because the coupon is flat and lightly loaded, while the product has steps, tented vias and a heavier copper area. The test detects a gross adhesion loss, and the thermal stress test is what covers the geometry effects.

Can surface energy be restored after contamination? Usually it can, with a micro-etch, a plasma treatment or a change in the rinse, and the dyne pen is what confirms the surface has actually changed. Treating the surface without re-measuring is the commonest way to repeat the same failure twice.