Casein Glue - A Natural Wonder Glue

Collection: Field Notes - Regenerative Materials

Series: Natural Marine Adhesives & Sealants Hub

Casein Glue: What It Is, How It Works, and Why I Use It for Every Structural Joint

I came to casein glue sideways. I was looking for a structural adhesive that was compatible with the rest of what I was building with — steam-bent hardwood, natural canvas, linseed oil, pine tar — and the obvious candidates kept failing the test. Epoxy works, but it is a petrochemical product with real toxicity during application, no biodegradable end of life, and a philosophical incoherence with the whole project. PVA works for some things, but it is thermoplastic, it creeps under sustained load, and it softens in heat. A hull sitting on a trailer in summer should not have PVA in its structural joints. What I needed was a thermoset adhesive made from natural materials with a centuries-long track record in actual watercraft. Casein glue is exactly that, and I use it for every structural joint in VAKA construction.

The mixing procedure and current experiments with tannic acid are in the how-to post. This post is the background: what casein actually is, how the chemistry works, what the Forest Products Laboratory found when they tested it systematically, and what the honest limits are.

Four Thousand Years of Use, and Why That Matters

The use of casein as a bonding agent is ancient enough that tracing it precisely is impossible. Egyptian tomb paintings contain pigments bound with casein. Roman writers describe woodworking applications. Medieval manuscript illuminators used it to fix gold leaf. What this long record actually tells you is not that the material is traditional in some sentimental sense, but that it works reliably enough to persist across centuries of use by people who needed their joints to hold.

The modern industrial history is more traceable. The Forest Products Laboratory in Madison, Wisconsin, has been investigating casein glue formulations since the early twentieth century. Their Report No. 280, originally published in 1930 and reaffirmed through subsequent editions to 1961, is the most thorough systematic treatment of the subject I've found. It is also available free, which is how primary sources should be. The key formulation from that work — Formula 11, developed by S. Butterman and C.K. Cooperrider and dedicated to the public — is the basis for what I use, modified with copper sulfate for biological resistance.

The aircraft connection is worth mentioning because it dispels a common assumption about natural adhesives. Before synthetic resins became available in the mid-twentieth century, casein glue was used extensively in wooden aircraft manufacture — including in the de Havilland Mosquito, where a birch and balsa sandwich structure bonded with casein-based adhesive formed much of the fuselage. The Mosquito was one of the fastest aircraft of the Second World War. The structural demands of a high-performance airframe are considerably more extreme than anything a sailing canoe or small catamaran will experience. The glue held.

The Chemistry: What Casein Actually Is

Casein is a phosphoprotein — the principal protein constituent of milk, present at roughly 2.5 to 3.5 percent by weight in bovine dairy, held in colloidal suspension. In its natural state it is not soluble in water. To make it function as an adhesive, it must be converted to an alkali-soluble form by reaction with an alkaline agent. This is where the chemistry becomes interesting.

When you add potassium hydroxide or lime to soaked casein, the alkaline environment unfolds the protein chains — denaturation, in the technical sense — and converts the casein to a caseinate salt. The caseinate is viscous, sticky, and bonds strongly to wood fibres as it cures and dries. The degree of alkalinity, the specific alkali used, and the ratio of casein to water all affect the working properties and final joint quality significantly. Too little alkali and the casein doesn't fully activate. Too much lime and the working life collapses to almost nothing, because lime accelerates the setting reaction. The FPL document is precise about this tension, and their formula is the result of systematic work to find the right balance.

The copper sulfate addition, which is what separates Formula 11 from its predecessor Formula 4B, acts as a preservative and significantly increases biological resistance. The FPL notes that copper probably acts against moulds, fungi, and micro-organisms that would otherwise attack the glue line in warm, damp conditions. In the VAKA formula, the glue line is always sealed under natural varnish — Le Tonkinois over a shellac primer — so the copper is never directly exposed to water flow. The bio risk from copper at these concentrations, under that varnish layer, is negligible. But the resistance it confers to the cured joint is real and measurable.

The tannic acid surface pre-wash, which I added to the formula separately from the FPL baseline, creates additional bonding sites in the timber surface. Tannins cross-link with protein, and pre-treating the wood grain with a 3% tannic acid solution before applying the casein gives the caseinate more to grip. Whether incorporating tannic acid directly into the glue mix — rather than only as a surface treatment — improves elongation to break is the subject of the experiments described in the how-to post. That work is ongoing and I don't have results yet.

What the FPL Testing Actually Found

The dry strength figures from the FPL work are the most useful thing to know about this material's structural capability. Testing on Douglas fir using ASTM shear methods produced values of 2,200 to over 3,000 psi for well-made casein glue joints, with wood failure rather than glue line failure as the typical result. Specifications for casein glues at the time required joint test values of 2,800 psi in hard maple. These are not marginal numbers. For the species and joint configurations I'm working with — steam-bent ash and willow, hand-cut lap and lashing joints — the strength is more than sufficient.

Water resistance is the more complicated question, and it's worth being honest about it. The FPL classifies well-formulated casein glue joints as highly resistant to water, broadly comparable to D3/D4 PVA in intermittently wet conditions. After 48 hours' soaking, casein-glue joints in plywood typically showed wet test values of 40 to 60 percent of their dry values in the FPL work. That is substantially better than vegetable or animal glues, but clearly weaker than epoxy or resorcinol under continuous submersion. Casein glue is water-resistant, not waterproof. The distinction matters.

For skin on frame construction, this distinction matters less than it sounds. The structural joints in a VAKA hull are not submerged. They are sealed under varnish, regularly dried out, and maintained at wood moisture contents well below the 18 to 20 percent threshold above which the FPL warns that durability is compromised. The glue's water resistance is adequate for the conditions it actually encounters. If I were building planked boats that would spend extended periods waterlogged, I would need a different adhesive. I'm not, so I don't.

Gap-filling is another area where casein outperforms PVA, and it matters in practice. PVA requires tight-fitting joints for full strength — it is essentially a film adhesive. Casein's working viscosity allows it to bridge small gaps without significant strength loss. When you're working with steam-bent timber and hand-cut joints, tolerances are not CNC-level. A glue that forgives this is genuinely useful.

The Comparison That Matters Most: Casein vs Epoxy

Most people building wooden boats today reach for epoxy by default. It's strong, it's waterproof, and it's familiar. The case for casein in plastic-free natural construction is not just that it's more principled, though it is — it's that epoxy's apparent advantages are mostly irrelevant for the specific application, and its real costs are consistently underweighted.

Epoxy's waterproofing advantage only matters at the glue line in continuously submerged conditions, which is not where the structural joints in a skin on frame hull live. Its strength advantage over well-made casein is marginal for the loads involved. What epoxy brings in exchange for these modest advantages is real toxicity during application, sensitisation risk with repeated exposure, no biodegradable end of life, a carbon footprint in manufacture, and permanent bonding that makes repair significantly harder.

The repair point is the one that matters most in the long run. When a casein joint fails — and eventually, in a well-used hull, something will — the joint releases cleanly under moisture and heat without damaging the timber. You warm the area, release the bond, replace the component, reglue. I've done this in a tent on a river bank with materials that weighed a few grams. An epoxy repair requires angle-grinding, which creates dust containing cured epoxy particles, and produces contaminated waste. It almost certainly requires a workshop. In some configurations it is simply not reversible without destroying the component.

Full reversibility and end-of-life composting are not peripheral advantages for a boat built with the intention of lasting as long as it's useful and then returning cleanly to the ground. They are core to the design logic. The adhesives hub covers how casein fits with shellac and the rubberised bedding compound into a complete natural system.

Uses Beyond Structural Joinery

The protein that makes this a useful structural adhesive is the same protein that makes milk paint work. Casein paint — pigments bound with activated casein — is one of the oldest decorative coatings known. It produces a flat, breathable finish that is chemically compatible with linseed oil and natural varnish, doesn't yellow, doesn't crack the way oil-based paints age, and is fully reversible. I use it on interior timber surfaces where the visual character of the material matters. The same batch of casein that goes into structural joints can go into interior paint, which is an unusual kind of coherence in a material system.

The compound has historically been used for fixing brushes to handles, for binding pigments, and for various craft and light manufacturing applications. The breadth of use reflects genuine versatility — it bonds to wood, fabric, paper, and natural fibres with good adhesion, and its working properties can be adjusted within reasonable limits by changing the alkali type and proportions.

The Mosquito connection I mentioned above sits in this broader history. Casein glue was the structural adhesive of choice for wooden aircraft frames before synthetic resins arrived precisely because it offered the right combination of strength, working time, and workability with hand tools. When synthetic alternatives became available, the industry moved on not because casein was inadequate but because the synthetics were more convenient at industrial scale and offered longer working times. Those trade-offs look different in a one-person workshop building natural-material boats.

At VAKA I design and build boats that don't destroy the environment. Find the plans as they are finalised at VAKA Plans and the full field notes here. If you are looking for a launching spot, the Hithe Finder is a community register of slipways, hards, and beaches for small boats.

VAKA. Traditional craft and natural materials. Nottingham. 2026.

References

Forest Products Laboratory, U.S. Department of Agriculture (1961). Casein Glues: Their Manufacture, Preparation, and Application. Report No. 280. Available via Oregon State University library: 

Sutermeister, E. and Browne, F.L. (1939). Casein and Its Industrial Applications. 433 pp. Reinhold Publishing Co., New York. Cited in FPL Report No. 280.

Truax, T.R. (1929). The Gluing of Wood. U.S. Department of Agriculture Bulletin 1500, 78 pp. Cited in FPL Report No. 280.

Kelly, A. Ashmun (1921). The Expert Wood Finisher. Available via Woodworkers UK: 

Henley, W.T. (ed.) (1914). Henley's Twentieth Century Formulas, Recipes and Processes. Available via Project Gutenberg