Acrylate gel and curtain injection, explained

What acrylate gel is

A two-component system, mixed at the head by a calibrated pump, that cures into a flexible rubbery gel. Cured elongation often runs into the hundreds of per cent, which is dramatically more movement tolerance than either polyurethane foam or epoxy offers.

Its defining property is the low viscosity. Where polyurethane foam bridges over a fine crack because it cannot enter it, and microfine cement grout has a hard particle-size floor below which it physically cannot go, acrylate gel travels into micro-cracks, capillary networks and the pore structure of the soil itself.

The trade-offs are equally defining. It has zero structural strength — it is a gel, not a plug, and it contributes nothing to the capacity of the element. Several formulations shrink irreversibly if they dry out completely, so it inherits a version of the same drying problem that governs the hydrophilic and hydrophobic resin argument. And gel time is strongly temperature dependent, which on a hot Queensland afternoon is a working constraint rather than a footnote.

How it sits against the other two chemistries

Curtain injection: treating the ground, not the wall

Curtain injection is the technique the material makes possible, and it is a genuinely different scope from anything else on this site.

Instead of injecting into a crack, holes are drilled right through the wall on a grid — typically 250 to 500 mm in each direction — and gel is pumped into the soil on the far side. As it permeates and cures it forms a continuous impermeable barrier between the ground and the structure. In effect it builds an external membrane without excavating to reach the outside face.

That is why it appears in the list of alternatives whenever negative side waterproofing hits its limits. Ordinary crack injection stops water at a defined path through the element; curtain injection stops water reaching the element at all, which sidesteps the redirection problem that governs every dry-side repair — sealing one path leaves the water looking for another.

It is used where excavation is impossible or absurdly expensive: a basement wall against a boundary, a lift pit below the water table, a structure in saturated sand, or a wall where the outside face is under a neighbouring building.

When acrylate gel is the right call

• Diffuse seepage across a broad face with no injectable line, where there is nothing for a foam to grip and nothing to key into.
• Honeycombed or porous concrete, typically at the base of a pour where the mix segregated, which behaves as a permeable zone rather than as a crack.
• Hairline cracks below the reach of any polyurethane, and below the particle-size floor of a microfine cement grout.
• Structures with genuine elastic movement, where the elongation is the reason to choose it — though real structural movement is an engineer’s question before it is a materials one.
• Curtain work into the ground behind a wall, where no other injection material permeates soil.
• Permanently wet elements. Like hydrophilic polyurethane, the gel depends on staying wet, so a tank wall, a planter box or a structure below the water table suits it and a wall that dries out every winter does not.

When it is the wrong call

The failure modes are as specific as the applications, and most of them come from using it where a foam belongs.

• A running or gushing leak under head. The gel can be washed out or diluted before it gels, which is precisely the condition a fast-accelerated hydrophobic foam is designed for. The correct sequence on a high-flow joint is foam first to kill the flow, gel second for the residual weeping the foam bridged over.
• Anywhere the element dries out completely and permanently. Several formulations shrink irreversibly, and unlike hydrophilic polyurethane there is not even a re-swell argument to fall back on.
• Anywhere strength is part of the requirement. The gel contributes nothing structural, and no amount of it substitutes for epoxy on a dormant structural crack.
• Any job where the pump calibration cannot be verified. A two-component material mixed off-ratio does not gel, and the result is a quantity of unreacted chemistry in the ground behind somebody’s wall.

Why it is priced differently, and what to look for in a quote

Ordinary crack injection is priced per linear metre of crack, because the quantity is visible and countable before anybody starts. Curtain injection is not, and a contractor who quotes it as a lump sum is either carrying a very large contingency or has not thought about it.

The quantity of gel a curtain takes depends on the porosity of ground nobody can see. The same wall length in sand and in clay takes materially different volumes, and neither can be established from the inside face. Curtain injection is therefore normally priced from a trial panel — a small area injected first to establish the actual take — at a rate per injection point or per kilogram of gel, with the trial result then extrapolated across the area.

An indicative band for curtain injection sits around $400 to $1,200 per square metre of wall face, but that band is genuinely wide and it should be treated as an order of magnitude rather than a price. Published Australian rates for curtain work are scarce, and the quantity is the hardest thing to predict in this trade — which is the honest reason the band is not narrower. The methodology behind figures like this is set out in the Australian cost guide.

• Is the price per point, per kilogram of gel, or per square metre — and if per square metre, what gel take per square metre is that based on?
• Is there a trial panel in the scope, and does the rate get revisited after it?
• What happens to the price if the actual take is double the assumed take? That question has to have an answer before work starts, not after.
• Is the pump two-component and calibrated, and when was it last verified?
• Which product, and what does its technical data sheet say about shrinkage on drying?

The fourth material nobody mentions

For completeness, because it comes up in specifications and is routinely misdescribed: microfine cement grout. It fills honeycombing, void networks and large defects where resin volume would be uneconomic, and it permeates coarse soils.

Cement in concrete matches the modulus of the parent material and will not creep or embrittle, which is a real advantage over any polymer for filling volume. Its limitation is a hard floor set by particle size — the crack needs to be roughly three to five times the d95 particle size, so it will not reliably enter cracks below about 0.3 to 0.5 mm.

One point of pedantry that matters in a specification: specify microfine cement by d95 in microns rather than by the word microfine, because the definitions differ between bodies. ISRM uses d95 under 16 microns, ACI Committee 552 uses a maximum diameter under 15 microns, and the Portland Cement Association uses under 10 microns. A specification that says only "microfine" has not specified the thing that determines whether the grout enters the defect.

Common questions

Is acrylate gel the same as polyurethane injection?

No, and they are not interchangeable. Polyurethane is a single-component material that reacts with water in the crack, foams and sets into a plug keyed into the crack faces. Acrylate gel is a two-component material about as thin as water that permeates a medium — soil, honeycombing, a capillary network — and cures within it as a flexible gel with no structural strength. The practical difference is that a foam fills a defined path fast, and a gel goes where there is no defined path at all.

What is curtain injection, in plain terms?

Drilling right through a wall on a grid, typically 250 to 500 mm in each direction, and pumping gel into the ground on the far side so it permeates the soil and sets into a continuous barrier between the ground and the structure. It builds the equivalent of an external membrane without excavating to reach the outside face, which is why it is used where excavation is impossible — a wall on a boundary, a structure in saturated sand, or an outside face that sits under a neighbouring building.

Why can curtain injection not be quoted as a fixed price?

Because the quantity depends on the porosity of ground nobody can see. The same wall length in sand and in clay absorbs materially different volumes of gel, and there is no way to establish which from the inside face. That is why the honest pricing basis is a trial panel — inject a small area first, measure the actual take, then extrapolate — with a rate per injection point or per kilogram of gel rather than a lump sum. A fixed price over unknown ground either carries a large hidden contingency or is heading for a variation.

Does acrylate gel shrink like hydrophilic polyurethane does?

Several formulations do, and irreversibly, which is arguably worse. Hydrophilic polyurethane shrinks when the water leaves and at least has a partial re-swell response when water returns, even if that response is weaker than the marketing suggests. A gel formulation that dries out completely can shrink permanently. The practical consequence is the same in both cases: these materials belong in elements that stay wet, and a wall that dries out every winter is a hydrophobic foam job.

Can gel be used to fix a leaking construction joint under pressure?

Not on its own, and not first. Gel injected into fast-moving water can be washed out or diluted before it gels, so on a joint with real flow the sequence is a fast-accelerated hydrophobic polyurethane first to kill the flow, then a flexible hydrophilic grade or an acrylate gel to seal the residual weeping the foam bridged over rather than penetrated. Two materials, two jobs, one joint — and a quote that specifies both for one element is usually a sign somebody has actually rated the flow before pricing it.

Is a two-component pump really necessary?

Yes, and it is not a detail. A two-component material mixed off-ratio does not cure into what its data sheet describes — at the extreme it does not gel at all, leaving unreacted chemistry in the ground behind a wall. Gel time is also strongly temperature dependent, so the calibration that was right in July is not automatically right in February. It is entirely reasonable to ask a contractor what pump is being used and when its calibration was last verified.

Where to next

• negative side waterproofing and its alternatives
• the resin argument that governs wet and dry elements
• how injection work is priced in Australia
• below-ground and basement leak sealing