Humidity and Cure: Moisture-Cure, Evaporative and Reactive

Why this matters

"Humidity slows cure" is true for some products, false for others, and backwards for a third group. A tech who carries one rule for all of them will tell a customer the silicone will be fine because it is a dry winter day, when dry cold air is exactly what slows it down, or will put an epoxy down on a damp autumn morning and come back to a greasy film the next coat will not stick to.

The way out is to know which of three cure mechanisms the product uses. The mechanism tells you which weather helps, which hurts, and what the failure looks like when it goes wrong. The data sheet still sets the limits (see related: Weather as a Work Condition: The Rule for Every Trade); this card is how to read humidity against them.

The three ways a material cures

Evaporative. The product is a binder dispersed or dissolved in a carrier, water or solvent, and it cures by losing that carrier to the air. Latex paints, acrylic latex caulks, most water-based stains and sealers, and solvent-borne lacquers work this way. Waterborne products also need the binder particles to fuse into a continuous film as the water leaves, a step called coalescence, and coalescence has a minimum temperature. Humidity matters because the air can only take water at the rate its dryness allows: the closer the air is to saturation, the slower the water leaves.

Moisture-cure. The product reacts with water vapor from the air to cure. Most silicone sealants, polyurethane sealants and adhesives, and moisture-cure urethane coatings work this way. They cure from the outside in: a skin forms first, and the cure front works inward as moisture diffuses through the skin. Too little moisture in the air slows them; too much, on some urethanes, makes them react too fast and release gas.

Reactive (two-part). Two components are mixed in a fixed ratio and cure by chemical reaction between them, with no need for anything from the air. Epoxies, two-part polyurethanes and polyureas, and many two-part adhesives work this way. Humidity is not part of the cure, but it can attack the surface of an uncured film, and temperature controls the reaction speed far more than for the other two families.

The field table

The sheet for a specific product states its limits. The table tells you which direction humidity pushes before you read it.

Relative humidity is not the moisture a sealant sees

Relative humidity is the air's water content as a share of what it could hold at its temperature. Cold air can hold much less. That is why "60 percent humidity" means two very different things in January and July, and why moisture-cure products slow down in winter even on days the forecast calls humid.

Work it out with two days at the same relative humidity:

• A July afternoon, 85 F and 60 percent RH. The dew point is about 69 F, and the water vapor pressure of that air is about 24.6 hPa (hectopascals, the unit meteorologists use for vapor pressure).
• A January morning, 25 F and 60 percent RH. The dew point is about 13 F, and the vapor pressure is about 2.75 hPa.

24.6 divided by 2.75 is about 8.9, so the July air carries roughly nine times as much water as the January air at the same 60 percent. Check it from the other side: water vapor content roughly doubles for every 20 F or so of temperature at saturation in this range, and 85 F minus 25 F is 60 F, three doublings, which is eight times. Close enough to confirm the order of magnitude.

A moisture-cure sealant in that January air has about a ninth of the water to react with, and the reaction itself runs slower in the cold. The two effects stack. That is why a silicone bead that skins in minutes in summer can take much longer in a hard freeze, and why a deep joint can stay uncured in its core for days. Many sheets state cure rates at a standard condition near room temperature and 50 percent RH; outside it, the sheet's numbers are optimistic.

Worked example: one cold, damp week, three products

A handyman shop has three jobs on the board for the same week in late October. The forecast is cool and damp: mornings in the low 40s F rising to the mid 50s, humidity 80 to 90 percent, overnight fog two of the five nights.

Job 1: caulk exterior trim joints before painting. The plan says acrylic latex caulk. Say its sheet gives a 40 F minimum and a paintable time at standard conditions. At 45 F and 85 percent, the water leaves the bead slowly, and a fog night puts liquid water on a bead that is still mostly water inside. The office's call is to caulk on the day with the driest afternoon, do it before noon so the bead gets the warmest hours, and paint no sooner than the sheet allows, pushing the paint to the following day rather than the same afternoon. A caulk painted over while still wet underneath can crack the paint film as the bead shrinks.

Job 2: reseal a shower surround and a few exterior window perimeters with silicone. Silicone is moisture-cure, so the damp week helps the exterior windows more than it hurts them; 85 percent RH at 45 F carries more moisture than the dry winter air in the earlier comparison. The interior shower is the better place to watch: the bathroom is warm, the cure is normal, and the customer is told the sheet's no-water time before the shower is used. The temperature still sets the floor on the exterior beads; the sheet's minimum and the surface reading govern (see related: Air, Surface and Material Temperature: Three Different Limits).

Job 3: an epoxy coating on an unheated garage floor. This is the one that fails if nobody thinks. At 7 a.m. the garage air is 55 F at 80 percent RH, which puts the dew point at about 48.9 F. The slab reads 52 F. The spread is 52 minus 48.9, which is 3.1 F, under the 5 F margin coatings commonly use (see related: The Surface Five Degrees Above Dew Point Rule for Coatings). On top of that, a cool, humid garage is the textbook condition for amine blush: the curing agent in many epoxies reacts with moisture and carbon dioxide in the air to leave a waxy carbamate film on the surface. It may not show until the next coat, which then fails to bond.

The shop does three things. It waits for the afternoon, when the slab has warmed and the spread has opened past the margin. It checks the sheet's minimum slab temperature and its recoat window at the expected temperature, which at 55 F may be well over a day. And it plans for blush rather than hoping against it: before the second coat, the first is checked for a greasy or waxy feel, and if it is there, the sheet's removal method (commonly washing with warm water and a scrub pad, then drying) comes before any sanding or recoat. Sanding a blushed film without washing it first just spreads the film around.

Temperature and the reactive products

For two-part products, temperature is the bigger lever. A common rule of thumb is that the reaction rate roughly halves for every 18 F (10 C) drop in temperature, and the cure tables on most epoxy sheets bear that out approximately: if the sheet says walk-on in about a day at 77 F, expect closer to two days at 59 F. Pot life moves the same way in reverse: a hot day shortens working time sharply, and a large mass mixed in a bucket heats itself and can kick off early.

Cold also changes the mix. Components that are cold are thick, they mix poorly, and a poor mix leaves soft spots that never cure, however long the crew waits. Store and condition the components to the sheet's product temperature before mixing.

Hazards these products carry

• Epoxy and urethane components are skin sensitizers, and a reaction developed on one job can recur at lower exposure on every later one: nitrile or the glove the SDS names, sleeves, and no skin contact with uncured material.
• Solvent-borne products and some urethanes give off vapors that are both inhalation hazards and flammable: ventilate, keep ignition sources away, and use a respirator only under a written 29 CFR 1910.134 program (applied to construction work by 29 CFR 1926.103) where the SDS calls for one.
• Acetoxy-cure silicones release acetic acid as they cure, the vinegar smell; in a small closed room, ventilate.

What changes the answer

• Indoor work. Inside a conditioned building, humidity and temperature are set by the HVAC, not the weather, unless the space is unheated or under construction with no conditioning.
• Thick sections. A deep sealant joint or a thick coating behaves differently from a thin film. Moisture-cure products take far longer to cure through depth, which is part of why joint design calls for a backer rod to control depth; evaporative products shrink more in thick beads.
• Hybrid chemistries. Some products cure by more than one mechanism. The sheet says which, and the table above applies to each part.

References

• Manufacturer product data sheets and SDS for the specific product (cure mechanism, limits, blush removal)
• 29 CFR 1910.134, Respiratory Protection (where an SDS requires a respirator), applied to construction work by 29 CFR 1926.103
• See related: The Surface Five Degrees Above Dew Point Rule for Coatings; What Dew Point Predicts
• See related: Sealant Joints and the Temperature You Caulk At
• See related: The Sealant That Never Cured