NUS gives up on recovering the carbon fibre and sells the whole composite scrap as aerogel

The National University of Singapore has published a composite recycling route that stops trying to rescue the carbon fibre ^[1][2]. Instead of separating fibre from cured epoxy, the team mills the entire scrap stream, fibre and resin together, holds the powder with carboxymethyl cellulose and freeze-dries it into an aerogel ^[3][4].

That inverts the usual business case. Conventional processes use high energy, harsh chemicals or high temperatures to recover only the fibres, then discard or destroy the epoxy ^[6], so the resin fraction is a cost line from end to end, and because thermoset epoxy cannot be melted down it has no cheap exit ^[5]. Grinding everything and selling the porosity converts that liability into product mass. The feedstock argument is the familiar one: turbine blades, fuselages such as the Boeing 787 and Airbus A350, and automotive frames generate thousands of tonnes of thermoset waste a year that goes to landfill or high-emissions incineration ^[7].

The lab figures describe a soft, mostly empty material. Density is 0.08 to 0.12 g/cm3 ^[8] and porosity 91.25 to 94.51 percent ^[9], which leaves solids at no more than about 8.75 percent of volume ^[11]. Elastic modulus tops out at 418.95 kPa ^[10], under 0.42 MPa ^[12], which is compliant-foam territory: something to be faced and protected, not loaded. Reported thermal conductivity is 0.042 to 0.049 W/m.K from lab tests ^[13][14], noise reduction coefficient reaches 0.51 ^[15], and oil uptake is near 15 g/g once a silane coating makes the surface reject water and take hydrophobic fluids ^[16][17]. Biocompatibility tests with human cells came back clean ^[18].

The volume arithmetic is where the commercial claim gets tested. At those densities, a tonne of finished aerogel occupies roughly 8 to 12.5 cubic metres ^[19]. Insulation, acoustic absorption and marine oil-spill sorbents ^[1] are three markets with three qualification regimes, and all of them buy by the cubic metre while the waste side is priced by the tonne. The report gives no price point, no per-site scrap volumes and no yield figure for how much binder the mix needs.

It also leaves the decisive process number blank. The team names the main hurdle as moving from laboratory batch freeze-drying to continuous large-scale manufacturing ^[20], and freeze-drying is not a finishing step here, it is the process. Nothing in the account quantifies its energy or cost, which is what determines whether avoiding the high-temperature and chemical fibre-recovery routes ^[6] actually reduces anything. Assoc Prof Duong Hai Minh, who led the study, frames the contribution as showing that composite waste need not be treated only as a disposal problem, and that using both fibre and epoxy fractions produces functional materials with higher value ^[21].

Three things to watch. Whether a named partner appears from the aerospace, manufacturing or waste-management industries the team is courting ^[22], since a sorbent or insulation product needs an offtaker before it needs a plant. Whether anyone demonstrates continuous drying, or substitutes a cheaper dewatering step, at a stated throughput ^[20]. And whether the property ranges hold on real mixed end-of-life scrap rather than controlled lab batches ^[14], because a modulus of 418.95 kPa ^[10] does not leave much headroom for feedstock variation.