Case Study: Mapping Surface Prep Consistency on Composite and Aluminum Surfaces

Key Takeaways

The Challenge

Adhesion is impacted by the top few molecular layers of a material. A “surface” is roughly 1,000 times thinner than a fingerprint, and about a million times thinner than a human hair. 

Figure 1. The surface of a substrate is about 1,000 times thinner than a fingerprint and about 1 million times thinner than a human hair.

At that scale, you can't judge prep quality by looking at a part. But, taking water contact angle measurements on a surface before bonding gives you a real number to judge your prep quality. A lower contact angle means higher surface energy and thus a surface more ready to bond. Brighton Science research has shown repeatedly that adhesion strength decreases as contact angle increases.

But that’s not where the story ends. The challenge is that surfaces are dynamic and ever-changing. After prep, it's difficult to know whether some areas were missed, re-contaminated, or have aged back to a lower-energy state. We wanted to see how consistent common prep methods used in manufacturing really are, across the whole surface and over time.

How we measured

We mapped differently prepared surfaces with an in-house robotic water contact angle tool, currently in the final stages of laboratory demonstration, which builds on our commercial robotic system to integrate our BConnect software. Each surface was mapped immediately after prep and again after aging up to 24 hours. Each square on the contact angle map represents one measurement.

What we found: composites

The abrasion, grit blasting, and plasma treatment results below all use the same CFR epoxy. The peel ply result uses a CFR PAEK composite.

Abrasion damage spreads past what you could see

We abraded the center of an IPA-wiped CFR epoxy coupon. A technician would likely flag only the heavily abraded middle, which overshot “sand-to-black”. But the less abraded edges of the abraded zone read nearly the same as the center, while untouched IPA-wiped areas read 20-30° lower.

Figure 2. Contact angle map overlaid on a CFR epoxy coupon with heavy abrasion in the center. Mean 53°, standard deviation 9°.

FTIR confirmed that the abrasion stripped all the epoxy resin off the center and exposed the carbon fiber. For more on why "sand to black" undermines composite bonds, see our post on measuring surface prep before composite repairs.

Grit blasting made the composite worse 

An IPA wipe brought the new CFR epoxy coupon from a mean contact angle of 44° down to 40°. Grit blasting raised it to 52°, and an additional IPA wipe only brought it down to 51°. The increase wasn't leftover grit or contamination. It was damage to the composite surface from over abrasion.  

Figure 3. CFR epoxy as received, after IPA wipe, after grit blasting and dry wipe, and after a second IPA wipe.

Many common prep methods were developed for metals, but they don't always work as well for polymer-based materials.

Plasma produced a stable, high surface energy surface

On an IPA-wiped coupon (mean 32°), a single pass of atmospheric plasma lowered the mean contact angle to 11°. After 24 hours, it had only risen to 17°. By removing contaminations before plasma treatment, we allow the substrate to be activated by the plasma, rather than the contaminants on the surface.

Figure 4. CFR epoxy untreated (39°), after IPA wipe (32°), after atmospheric plasma (11°), and aged overnight (17°).

Peel ply doesn't always mean you can skip surface prep

The surface of the CFR PAEK coupon looked uniform right after the wet peel ply was removed, but it had an average contact angle of 51°. The composite's outer layers take on the character of the peel ply resin, and many peel plies have low surface energy. Plasma treatment right after removal with no IPA wipe brought the mean to 19°, with no reading above 26°. This means the peel ply did a good job of protecting the surface from contamination, but it still needed a helping hand to optimize surface energy.

Figure 5. CFR PAEK composite directly after peel ply removal and after atmospheric plasma treatment. No IPA wipe was used before or after plasma treatment.

What we found: aluminum

Clean aluminum is very reactive, starting with high surface energy but forming an oxide layer quickly. A metal-oxide layer is lower in surface energy and mechanically weaker than the metal beneath it. The order of prep steps made a clear difference in how the surface aged.

Figure 6. Aluminum untreated (79°), after IPA wipe (40°), after atmospheric plasma (14°), and aged overnight (37°).

Figure 7. Aluminum untreated (68°), after IPA wipe (46°), after aluminum oxide grit blast (20°), after atmospheric plasma (9°), and after 24 hours (28°).

Grit blasting removed the oxide layer before plasma treatment, giving a lower starting angle, slower drift, and a longer usable out-time. Meanwhile, abrading the composite decreased the surface energy because of carbon fiber damage. These trends show that prep must be tuned to each substrate.

Lessons for your process

• Match the method to the material. A step that helps a metal can hurt a composite. Validate each step on each substrate, rather than carrying procedures over.
• Measure after each step, not just at the end. In the grit blast example, measuring between steps showed the problem was substrate damage, not contamination. That changes the fix.
• Know your out-time. A surface that reads 9° right after prep can be much higher by the next shift. Measure the drift on your own parts and build that window into your process.
• Set your own threshold. The numbers in this study describe these coupons, not a universal spec. The right limit depends on your substrate, adhesive or coating, and process, so correlate contact angle with your own mechanical testing before setting a max.
• You don't have to map every part. Mapping during development shows where prep tends to fall short. In production, a few targeted measurements at those spots give you a heads up that something’s gone wrong.

Brighton Science's Surface Analyst™ instruments give you a water contact angle reading in seconds, on the production floor or in the lab, so your team can confirm a surface is ready before bonding or coating.

Common questions

What contact angle is good for bonding? There's no single number. The reliable approach is to correlate contact angle with mechanical test results on your own system and set a maximum allowable angle from that data.

How long does a plasma-treated surface stay ready to bond? It depends on the material. In this study, plasma-treated CFR epoxy went from 11° to 17° over 24 hours, while plasma-treated aluminum went from 9° to 28°. Measure the drift on your own parts to set an out-time.

Why did grit blasting help aluminum but hurt the composite? On aluminum, blasting removes the weak, low-energy oxide layer. On a polymer-based composite, it can damage the surface itself, as well as reveal too much carbon fiber. Here, grit blasting raised the composite's contact angle, and solvent wiping didn't reverse it, demonstrating irreversible mechanical damage.

Does a wet peel ply surface still need prep? It can. The composite's outer layers take on the character of the peel ply resin, which is often low in surface energy. In this study, the CFR PAEK surface read 51° after peel ply removal until plasma treatment brought it to 19°.

About this study

This case study is based on Identifying Inconsistencies in Surface Preparation Through Robotic Water Contact Angle Mapping, by Sonia Stanciu, Alan Jechort, and Giles Dillingham, presented at CAMX 2026. 

Talk with a Brighton Science applications engineer about setting a surface prep spec for your materials.