Collection: Field Notes — Preserving Natural Materials at Sea 

Series Hub: Preserving Wood 

Linseed Oil vs Polyurethane for Wood — Penetrating Oil, Film Finish, and How Each One Fails on a Boat

I came to linseed oil by instinct before I came to it by evidence. The instinct was that a boat built from natural materials should not be sealed under a skin of synthetic polymer. That might be right. It might be sentiment. The only honest way to settle it is to follow the old sources, test the materials, and report what happens rather than what I hoped would happen.

How Each Material Cures

Linseed is a drying oil. Only a few plant oils behave this way. Tung is the other one worth knowing. They cure not by evaporation but by reaction with atmospheric oxygen. The fatty acids in linseed, alpha-linolenic acid in particular, carry double bonds that oxygen attacks, forming cross-links between molecules until the oil hardens through its depth. The finish does not sit on the wood. It sets within it.

The old sources are blunt about one thing. Raw linseed is the wrong material for this. Ashmun Kelly’s The Expert Wood Finisher, a book I go back to often, states it plainly. Raw oil dries soft, stays tacky, and produces nothing durable. It needs metallic driers to cure within a practical timeframe. What results is boiled linseed oil, and when I write linseed oil in these notes, that is what I mean. There is a third path. Heat-polymerised linseed, made by cooking raw oil to around 300°C until it bodies, cures faster without metallic driers and avoids the cobalt and manganese compounds that conventional boiled oil carries. I have not made a batch yet. When I do it gets its own notes.

Polyurethane works the opposite way at every level. It is a synthetic polymer in a solvent carrier. The solvent evaporates and the resin cross-links into a continuous film above the wood. Bob Flexner’s Understanding Wood Finishing, the other book I keep within reach, is precise about what that hardness means in use. Polyurethane is the most abrasion-resistant of the common varnishes, which is exactly why it suits interior floors and dining tables, and exactly why it fails on exterior wood. Hardness and rigidity are the same property. Rigidity is the enemy of any coating on wood that moves.

Where Polyurethane Earns Its Place

Before the case against it, the case for it. Polyurethane is not a bad material. It is the wrong material for a specific application, which is a different charge.

On hardwood floors under daily punishment, chairs dragged, boots wet, things dropped, an oil finish wears through quickly. The same floor in oil-based polyurethane holds for years. The hardness that disqualifies it outdoors is a real virtue on a horizontal interior surface taking mechanical abuse. On figured wood where visual depth is the point, curly maple, quartersawn oak, polyurethane builds a film that flatters the figure in a way a penetrating oil, which more or less disappears into the surface, cannot. Le Tonkinois sits between the two, a tung oil and resin compound that builds more than pure oil while staying more flexible than urethane. I have used it and rate it.

Where the Trouble Starts

Wood moves. It swells across the grain as moisture rises and shrinks as it dries, continuously and significantly in any outdoor or marine situation. A rigid film on a moving substrate has one available response. It cracks. Flexner confirms this directly. He notes that spar varnish, which carries far more oil than standard varnish and cures to a much more flexible film, is the right film finish for exterior wood precisely because flexibility rather than hardness is what exterior wood needs. Polyurethane is the wrong specification for the job. That is a technical judgement, not a preference.

Once the film cracks the damage turns structural without showing itself. Water enters the crack, gets behind the film, and is then sealed in. The intact surrounding film stops it evaporating. You have built a persistently damp pocket in exactly the place rot needs sustained moisture to take hold. The surface looks sound from a few feet away. The rot underneath does not announce itself until the damage is real. This is the standard failure mode of polyurethane on exterior and marine wood. It is reliable, and it is not slow.

How Linseed Oil Fails Differently

There is no film to crack. What happens instead is depletion. The polymerised oil oxidises further under ultraviolet light, goes brittle at the outermost surface, and the wood begins to grey and check. That is the material telling you something before anything structural is at risk. Oiled wood stays readable in a way polyurethane-coated wood does not.

Maintenance is a fresh coat on a clean surface. No stripping, no sanding to key it, nothing beyond removing dirt and letting the wood dry. The new oil integrates with what remains of the old. A Danish condition survey of ten linseed-oil-treated buildings, published in 2002 by E. Brandt and T. Lading, produced a finding I keep returning to. Across every surveyed example, there was no wood in poor condition beneath an essentially intact linseed coating. The authors contrast this with latex paint, where rot behind sound-looking film is documented and common. The material that lets water in also lets it out. The wood beneath it stays open to air.

What Linseed Oil Does Not Do

Here I want to be careful, because the case for linseed is routinely overstated and the overclaims are easy to knock down. Flexner says it plainly. Of the common finishes, only wax is less protective than boiled linseed oil for surface hardness, scratch resistance, and stain resistance. Liquid water works through a brush-applied coat and can raise the grain within minutes. Water vapour passes through almost as if the finish were not there. If protection means a hard barrier against knocks and liquid, polyurethane wins outright, and pretending otherwise is dishonest.

The honest case for linseed rests elsewhere. Work by Miha Humar and Bostjan Lesar at the University of Ljubljana, published in International Biodeterioration and Biodegradation in 2013, tested linseed- and tung-oil-treated timber against wood-decay fungi and measured water uptake in the lab and outdoors. Oil treatment measurably cut water uptake and gave real fungal resistance. Fungal mass losses in linseed-treated timber ran roughly half those of untreated controls. The part that matters most is their reason why. One of the main advantages of oil-based treatments is that they do not seal the surface, so the wood can dry after rain. Lowering the average moisture content over time is enough to slow fungal attack, even where the oil does not keep water out entirely.

The caveats are real. The Ljubljana study used vacuum-pressure impregnation at high oil retentions, far beyond what a brush achieves. The benefits scale with penetration depth. For brushed coats on a frame the fungal resistance and moisture reduction are genuine but modest next to deeply impregnated specimens. The oil slows the moisture cycling. It does not stop it. Tung oil, for what it is worth, beat linseed on both water exclusion and fungal resistance throughout that study. I have not yet run my own field comparison between the two. It is on the list.

Application, Where the Result Is Actually Decided

Brandt and Lading found wildly variable results across their ten buildings, from failure inside two years to excellent condition after eight or more, with workmanship and conditions accounting for most of the spread. This is the part most published sources understate.

The wood has to be genuinely dry, below 12 to 15 percent moisture content. Brandt and Lading name this the single most important factor. Damp wood cannot be saturated and the treatment fails from the start regardless of later care. Temperature matters too. Below roughly 5 to 10 degrees, boiled linseed gets difficult to apply thin and drying times stretch out. A cold February workshop gives me unpredictable results. A mild October day in a shed with a small heater running does not. And the inter-coat intervals are longer than almost every guide admits. The most durable treatment in the Danish survey used five coats at one-week intervals, more than six weeks before the surface was finished. Rush it and you get thick, half-cured layers that wrinkle and crack instead of building into sound treatment. Thin coats, all excess wiped back before it goes tacky, each one cured fully. The patience is the hard part to pass on to anyone in a hurry to launch.

The Environmental Question

The boats I build are made from materials that compost when they are done. The finish they carry should match that. Polyurethane weathers into synthetic polymer fragments. Sanding it produces microplastic particles directly, and weathered coatings break up progressively over their service life into the same water the boat floats on. Boiled linseed degrades into organic fatty acid compounds that break down rather than accumulate. This is the same question as the rest of the build, not technical reasoning dressed up as ethics. The fibreglass disposal problem is the loud version. Synthetic coatings on wooden boats are the quiet version of the same story.

Where This Investigation Is Going

I have not yet run a controlled comparison of brushed linseed against polyurethane on matched timber outdoors. I have used both, on different projects, over several years, and the qualitative difference in maintenance behaviour is as the sources describe. What I do not have is the quantitative difference in moisture uptake under field conditions, brushed on, real temperatures, actual marine exposure. I want to test heat-polymerised oil against conventional boiled linseed under the same conditions, and tung against both. The published research gives a starting point. What it does not give is results from a small workshop on the Essex coast, in the conditions the people reading this are actually working in. That is what the Field Notes are for. For now, this is where I am..

Sources: Ashmun Kelly, The Expert Wood Finisher (1921). Bob Flexner, Understanding Wood Finishing (2005). Miha Humar and Bostjan Lesar, Efficacy of linseed- and tung-oil-treated wood against wood-decay fungi and water uptake, International Biodeterioration & Biodegradation (2013). E. Brandt and T. Lading, Linseed Oil Paint As An Alternative To Wood Preservatives, 9th DBMC Conference (2002).

Plans for skin-on-frame boats built in natural materials and finished to be maintained rather than replaced at VAKA Boatplans. The full knowledge base at Field Notes.