KALTEC

BlogApplication notes

Why won't adhesives or sealants stick after cleaning?

Type
Application notes
Published
20 November 2026
Prepared by
Technical team
Product group
Cleaning & Surface Prep

When a bond fails after cleaning, cleaning once more is the usual response, but a surface that looks clean is not necessarily one the adhesive can wet. What actually stops the bond can be an oil film left behind, cleaner residue after rinsing, recontamination before application, a weak oxide on metal or a low-energy plastic, and sometimes more than one at once.

Sometimes a sealant applied to freshly cleaned parts peels off whole when pulled a few days later, or bonded products come back after a humid season with failed joints. On the failed joint, the adhesive sits whole on one side and the other side looks like bare metal. The parts were certainly cleaned, so it is hard to see why they did not bond.

Bond failures usually show up after assembly, after shipment or after months of moisture and temperature cycling. A lifted sealant brings leaks and corrosion, and a failed structural bond means rework, returns and safety issues. Cleaning and bonding are often done by different people on different days, which makes it hard to see that the cause lies in cleaning.

The basic principle is wetting. For a liquid to spread on a solid, the solid's surface energy, the pull its surface exerts on other materials, must exceed the liquid's surface tension (Zisman, 1964; Ebnesajjad, 2011). Zisman set out how to find the critical surface tension, the limit at which a solid wets fully, from the contact angles of several liquids (Zisman, 1964), and Owens and Wendt showed how to split surface energy into a dispersion part, the weak pull every material has, and a polar part, the pull between molecules with uneven charge such as water (Owens & Wendt, 1969). Only an adhesive that wets well fills the fine roughness of the surface, which raises bond strength and fracture energy (Packham, 2003). A single oil layer or a weak oxide either blocks this wetting or, even when wetted, becomes a weak boundary layer that tears first (Kinloch, 1987; Baldan, 2004).

In such cases the technical team sets up the five hypotheses below and rules them out one by one. The start is always the same: pull a failed joint apart to see which side the adhesive stayed on, and run water over a part just after cleaning to see whether the film breaks. Those two checks screen several hypotheses at once (Kinloch, 1987; ASTM International, 2021).

Hypothesis 1: oil film left

The first hypothesis is oil that was not fully removed, because it is the easiest to check. It arises on parts that carry machining oil, rust preventive, stamping oil or mould release, and it tends to show when the cleaning bath has gone long without a change and filled with oil, or when the oil from an earlier step has changed. Picture a press line that switches to a new rust preventive: the cleaning conditions stay the same, yet sealant beads on the steel parts no longer flow out smoothly and pull back at the edges. No oil is visible, but a thin film can still be there.

On an oil film, the adhesive never reaches the metal. Machining oil, rust preventive and release agents are mostly hydrocarbons or silicones with low surface energy, and once surface energy falls below the adhesive's surface tension, the adhesive beads instead of spreading and the contact angle rises (Zisman, 1964; Ebnesajjad, 2011). With poor wetting, the adhesive does not reach the bottom of the roughness, so the real contact area shrinks and gaps remain at the interface (Packham, 2003). If some oil dissolves into the adhesive, the adhesive near the interface may cure poorly and weaken (Kinloch, 1987; Baldan, 2004). The failed joint then shows interfacial failure: adhesive whole on one side, the other side clean.

The quickest check is the water-break test. Run clean water over a part just after cleaning and look for breaks or beading in the film. It is a fast go/no-go test that can catch water-repelling organic films a few molecules thick, but it gives no number and is less sensitive on very rough or porous surfaces (ASTM International, 2021). When a number is needed, calculate surface energy from the contact angles of a polar liquid such as water and a non-polar one, and compare values on the same part before and after cleaning. A low surface energy points to contamination (Owens & Wendt, 1969; ASTM International, 2022).

If the water film breaks, look at the cleaning bath first. Check concentration, temperature and time, and how much oil the bath has picked up. An oil-loaded bath puts oil back on the parts, so set an oil-separation or bath-change interval. If the oil has changed, such as a new rust preventive or a silicone release agent, test on coupons whether the current cleaner can dissolve or emulsify it. After any change, repeat the water-break test or contact angle the same way and record the result.

Hypothesis 2: cleaner residue

Look at this hypothesis when parts pass the water-break test but still bond poorly. It tends to arise in washers using water-based cleaners when the rinse is short, when the rinse tank has gone long without a change and picked up cleaner, and on parts with many holes and seams that hold rinse water. Say the rinse tank has been in use for a while without a change, and faint white spots start to show on the dried parts, with the sealant letting go only on those spots. Failures may also cluster around places where water collects, such as between overlapping sheets.

Water-based cleaners carry surfactants, which wrap oil and lift it into the water, along with additives such as corrosion inhibitors and defoamers. With too little rinsing, these components dry into a thin film on the surface. A surfactant has one end that likes water and one that likes oil, and if the oil-loving end faces outward after drying, it can lower surface energy instead (Ebnesajjad, 2011; Baldan, 2004). The leftover film sits between adhesive and surface as a weak boundary layer (Kinloch, 1987). In a study of 6082-T6 aluminium bonded with a two-part silicone adhesive, bond strength kept falling as surfactant on the surface increased. The adhesive near the interface absorbed surfactant while curing, and once that limit was passed, failure moved from cohesive failure inside the adhesive to interfacial failure (Brandão et al., 2023). Cleaner components that dissolve into the adhesive can also interfere with its cure (Baldan, 2004).

Start with spots and rinse water. Look for white stains, water marks or tackiness once parts are dry, and log the conductivity and foaming of the final rinse tank regularly. Rising conductivity or foam means cleaner is being carried over into the rinse. Also look for rinse water left in holes, seams and overlaps. Trapped water keeps the adhesive from touching the surface, and with moisture-reactive polyurethanes or moisture-cure sealants it can cause bubbles or poor cure. If that does not settle it, make bonded coupons from parts rinsed as usual and from parts given one extra clean-water rinse, and compare strength and failure mode.

More rinsing is the first remedy. Rinse in several stages and make the final rinse clean water, deionised if possible. Set the rinse-change point by conductivity, and let parts drain well to cut carry-over from the wash tank. A cleaner above its recommended concentration only makes rinsing harder, so keep it within the maker's range. After rinsing, dry with air blow-off or warm air until water in gaps is gone before applying, and where a cleaner with corrosion inhibitor feeds a bonding step, confirm with the adhesive maker that it is suitable before bonding.

Hypothesis 3: recontamination

This is the hypothesis when cleaning conditions stay the same but failures vary from day to day. It tends to arise where cleaned parts wait a long time before bonding, where parts are moved by bare hand, and in plants that use silicone lubricants, release agents or sealants near the bonding area. A typical line leaves cleaned parts on a rack for a day or two before bonding, while at the next bench someone wipes fixtures with silicone spray. On such lines, failures tend to cluster in one shift, with one operator or in lots that waited long.

Why does a clean surface get dirty again? A clean metal surface has high surface energy, so it readily attracts organic matter from its surroundings. It is at its best just after cleaning, and over time oil mist, skin oils and material from gloves and packaging build up again (Ebnesajjad, 2011; Kinloch, 1987). Silicone can cover the surface and block bonding even in tiny amounts. In NASA testing, an epoxy bond to bare D6AC steel lost 75% of its tensile strength at 10 mg/ft² of silicone contamination, while the same epoxy on painted steel showed no loss at that level. The same contamination can matter a great deal or not at all depending on substrate and bond system, and the study also tested cleaning methods for removing it separately for each bond system (Stanley, 2008).

The key to checking is timing. Run a water-break test on the same kind of part once just after cleaning and once just before application. Passing after cleaning but failing before application means contamination in between (ASTM International, 2021). Match the records to see whether failures cluster in one shift, with one operator or in lots that waited long. Also check whether silicone sprays, silicone release agents, hand creams, or gloves and packaging containing silicone are used near the bonding area.

If recontamination is the cause, fix how parts are handled rather than the cleaner. Set a maximum time from cleaning to application, and re-clean when it is exceeded. Do not touch cleaned parts with bare hands; use clean gloves and store parts covered. Keep silicone products out of the bonding area, and once silicone contamination is confirmed, stop the source first rather than repeating ordinary cleaning (Stanley, 2008). The allowed time depends on substrate and adhesive, so set it by coupon testing.

Hypothesis 4: weak oxide

When joints hold at first and fail only after months in humid service, it is time to suspect the oxide. It tends to show when aluminium or steel is only degreased before structural bonding, when parts face moisture and temperature cycling, and when the material or supplier has changed. A common example is an aluminium bracket that was only degreased. Bonded with a structural adhesive, it passes the pull test at final inspection, then comes back from the field after a rainy season with the joint failed. Grey powder may come away on the failed surface.

What an adhesive actually bonds to is not the metal but the oxide on it. Aluminium forms a thin oxide in air almost at once, and rolled or heat-treated surfaces may carry a thick, brittle oxide enriched with alloying elements such as magnesium. Degreasing removes oil but does not change this oxide, so an adhesive on a weak oxide fails when the oxide cracks first under load (Venables, 1984; Critchlow & Brewis, 1996). Long-term durability depends on the oxide's shape and stability. A porous, rough oxide made by etching or anodising interlocks mechanically with the adhesive and raises strength, but when moisture seeps in from the bond edge, the aluminium oxide takes up water and turns into a hydroxide called boehmite. The layer changes volume and shape, and the bond separates at the interface (Venables, 1984). On steel, flash rust formed while parts dry after cleaning can become a similar weak layer.

The colour of the failed surface and the timing of the failure give the clues. If joints pass initial tests but fail after humidity ageing or in service, suspect oxide and moisture first (Venables, 1984). Grey or black powder or rust on the adhesive side means failure ran through the oxide or rust layer. Also check the history: whether cleaning is followed by a pretreatment such as abrading, blasting, etching or anodising, and whether the material or supplier has changed (ASTM International, 2024).

For structural bonds or parts in humid service, do not stop at degreasing; add a pretreatment that suits the metal. Standards set out mechanical and chemical procedures for metals such as aluminium, stainless steel, carbon steel and titanium (ISO, 2012; ASTM International, 2024). Give aluminium a stable oxide by etching or anodising (Critchlow & Brewis, 1996), and use the primer the adhesive maker recommends where needed. Dry steel quickly after cleaning, and if flash rust forms, use a rinse with a corrosion inhibitor but check on coupons that it does not hurt the bond. After changing the pretreatment, judge it by strength after humidity exposure, not by initial strength alone (ISO, 2012).

Hypothesis 5: low-energy plastic

If the parts are a plastic such as polyethylene, polypropylene or a fluoropolymer, suspect the material first, whatever the cleaning. It tends to show when metal parts are switched to plastic, when the material changes to one with recycled content, and when parts treated by corona or plasma are stored a long time before bonding. On one assembly line a metal housing becomes a polypropylene cover, and a sealant that used to hold now peels off by hand however well the part is wiped. A new cleaner and longer cleaning make no difference.

Here the material itself is the cause. These plastics have almost no polar groups on their chains, so their surface energy is low by nature, and removing every trace of oil still exposes a surface that wets poorly (Zisman, 1964; Owens & Wendt, 1969). The international standard notes that most thermoplastics need careful preparation because of their low surface energy, while thermoset composites are less demanding but still need residual mould release removed (ISO, 2012). Corona, flame and plasma treatment raise surface energy by attaching oxygen-containing polar groups. This state is thermodynamically unstable, though: over time the polar groups turn back into the bulk and the surface repels water again, a process called hydrophobic recovery, and temperature, humidity and crystallinity affect its rate (Mozetič, 2023). Long storage after treatment, or cleaning after treatment, can therefore reduce the effect.

First read the material from the marking on the part or the drawing, and check whether the material has changed or contains recycled content. If the surface was treated, apply test liquids of graded surface tension and read the wetting tension from whether they spread or bead. This is a quality-control method for surface treatment, but the value alone does not fully predict adhesion, so read it alongside bond tests (ISO, 2003). Record the time and storage conditions between treatment and bonding, and track how contact angle changes over time (Mozetič, 2023).

A low-energy plastic will not be fixed by changing the cleaner. Remove release agent and dust by cleaning, then apply a surface treatment such as corona, flame or plasma, or use a primer or adhesive made for low-energy surfaces (ISO, 2012). With surface treatment, set a time window from treatment to bonding, and handle the treated faces so they are not recontaminated. If cleaning after treatment is unavoidable, confirm by wetting tension that it does not undo the treatment (ISO, 2003; Mozetič, 2023).

Testing the hypotheses

Five hypotheses do not call for five full investigations. Rule them out one at a time, starting with what is easy to see and quick to answer.

The first step is to pull a failed joint apart and look at how it failed. If adhesive remains on both faces, the problem lies in the adhesive itself or its cure rather than the surface, so set the cleaning hypotheses aside (Kinloch, 1987). If one face is bare, an interfacial failure, identify the substrate next. If it is a low-energy plastic such as polyethylene or polypropylene, look at hypothesis 5 before cleaning (ISO, 2012).

For metal, move on to the water-break test. If the film breaks just after cleaning, that is hypothesis 1, the oil film, so look at the cleaning conditions and oil in the bath. If it passes after cleaning but breaks just before application, it is recontamination, hypothesis 3 (ASTM International, 2021). If it passes at both points and still does not bond, check for spots after drying, rinse conductivity and water left in gaps to decide on hypothesis 2. These two tests alone screen out much of hypotheses 1, 2 and 3.

If that still does not settle it, look at timing and control coupons. Joints with normal initial strength that fail later in humid service point to hypothesis 4, the oxide (Venables, 1984). Make coupons with only the suspected cause changed and set them beside coupons made under normal conditions. Change one thing at a time, such as an extra rinse, bonding straight after cleaning or an added pretreatment; only then can you tell which one worked. Where moisture looks like the issue, check strength after humidity exposure as well as initial strength (ISO, 2012).

Once the cause is found and the process fixed, the last step is to measure again and record it. It is done only when the water-break test or contact angle, measured the same way, passes and bonded coupons show cohesive failure within the adhesive (ASTM International, 2022; Kinloch, 1987).

One more thing to remember: the hypotheses can overlap. Parts that hold rinse water in their gaps take longer to dry and so wait longer, and recontamination and flash rust can follow in the meantime. In such cases, put cleaning, rinsing and drying right first, then recheck the rest, because no other result can be trusted until the surface is clean.

Limits and cautions

The water-break test and wetting tension show surface condition quickly, but their values alone do not guarantee bond strength or durability (ASTM International, 2021; ISO, 2003). Acceptance criteria for a given part follow the industry's specifications and the adhesive maker's process guidance. Where a bond bears directly on safety, as in aerospace, automotive safety parts, medical devices or pressure vessels, do not decide on floor judgement alone; validate the process with coupon tests together with a bonding specialist.

Safety matters too. Before changing or handling cleaners and surface-treatment chemicals, read the safety data sheet, the SDS. Organic solvents can harm the nervous system, liver, kidneys, lungs and skin, and many are highly flammable. Provide ventilation, control ignition sources, wear suitable gloves and eye protection, use respiratory protection where needed and keep within exposure limits (OSHA, n.d.). The acids and alkalis used for etching and anodising can severely injure skin and eyes, so handle them only where the equipment and training are in place.

Sources

  1. ASTM International. (2021). ASTM F22-21, Standard test method for hydrophobic surface films by the water-break test. https://store.astm.org/f0022-21.html
  2. ASTM International. (2022). ASTM D7490-13(2022), Standard test method for measurement of the surface tension of solid coatings, substrates and pigments using contact angle measurements. https://store.astm.org/d7490-13r22.html
  3. ASTM International. (2024). ASTM D2651-01(2024), Standard guide for preparation of metal surfaces for adhesive bonding. https://store.astm.org/d2651-01r24.html
  4. Baldan, A. (2004). Adhesively-bonded joints and repairs in metallic alloys, polymers and composite materials: Adhesives, adhesion theories and surface pretreatment. Journal of Materials Science, 39(1), 1–49. https://doi.org/10.1023/B:JMSC.0000007726.58758.e4
  5. Brandão, R., Borges, C. S. P., Marques, E. A. S., Carbas, R. J. C., Akhavan-Safar, A., Nunes, P. D. P., Ueffing, C., Weißgraeber, P., Schmid, F., & da Silva, L. F. M. (2023). Effect of surfactant contamination on the properties of aluminum/silicone adhesive joints. Mechanics of Advanced Materials and Structures, 30(9), 1875–1888. https://doi.org/10.1080/15376494.2022.2046217
  6. Critchlow, G. W., & Brewis, D. M. (1996). Review of surface pretreatments for aluminium alloys. International Journal of Adhesion and Adhesives, 16(4), 255–275. https://scilit.com/publications/380505d646ad4aa9dfe7cf98a8ee319d
  7. Ebnesajjad, S. (Ed.). (2011). Handbook of adhesives and surface preparation: Technology, applications and manufacturing. Elsevier/William Andrew. https://shop.elsevier.com/books/handbook-of-adhesives-and-surface-preparation/ebnesajjad/978-1-4377-4461-3
  8. International Organization for Standardization [ISO]. (2003). ISO 8296:2003, Plastics — Film and sheeting — Determination of wetting tension. https://www.iso.org/standard/38451.html
  9. International Organization for Standardization [ISO]. (2012). ISO 17212:2012, Structural adhesives — Guidelines for the surface preparation of metals and plastics prior to adhesive bonding. https://www.iso.org/standard/50596.html
  10. Kinloch, A. J. (1987). Adhesion and adhesives: Science and technology. Chapman and Hall. https://link.springer.com/book/10.1007/978-94-015-7764-9
  11. Mozetič, M. (2023). Aging of plasma-activated polyethylene and hydrophobic recovery of polyethylene polymers. Polymers, 15(24), 4668. https://doi.org/10.3390/polym15244668
  12. Occupational Safety and Health Administration [OSHA]. (n.d.). Solvents. U.S. Department of Labor. https://www.osha.gov/solvents
  13. Owens, D. K., & Wendt, R. C. (1969). Estimation of the surface free energy of polymers. Journal of Applied Polymer Science, 13(8), 1741–1747. https://doi.org/10.1002/app.1969.070130815
  14. Packham, D. E. (2003). Surface energy, surface topography and adhesion. International Journal of Adhesion and Adhesives, 23(6), 437–448. https://doi.org/10.1016/S0143-7496(03)00068-X
  15. Stanley, S. D. (2008). Evaluation of the effect of silicone contamination on various bond systems and the feasibility of removing the contamination (NASA Technical Reports Server 20080023444). https://ntrs.nasa.gov/citations/20080023444
  16. Venables, J. D. (1984). Adhesion and durability of metal-polymer bonds. Journal of Materials Science, 19(8), 2431–2453. https://doi.org/10.1007/BF00550796
  17. Zisman, W. A. (1964). Relation of the equilibrium contact angle to liquid and solid constitution. In Contact angle, wettability, and adhesion (Advances in Chemistry, Vol. 43, pp. 1–51). American Chemical Society. https://doi.org/10.1021/ba-1964-0043.ch001

Product and site consultation

Leave your company, site and the product groups in use. Our engineer will get in touch to go through your site conditions and concerns. Materials, equipment, operating conditions and downstream steps show what should change and what can stay. If nothing needs to change, we will say so.

Contact us
After you contact us
  1. 01
    Site setup

    We confirm process conditions and needs first.

  2. 02
    Spec design

    We set the product spec and selection criteria.

  3. 03
    Comparison

    We compare matching candidates and propose.

  4. 04
    Tailored supply

    We supply on time and manage the stock.

  5. 05
    Site check

    We visit, check conditions of use and record.

  6. 06
    Use management

    We track use and change cycles to catch issues.

  7. 07
    Regular analysis

    We analyze the records and send a yearly report.