Research digest

The figures quoted across this site come from the published literature rather than from product data sheets. This page lists the principal sources, says what each one established, and marks which are open access.

Review articles: start here

Four review papers between them cover the field, and reading them in this order is the fastest route into the subject.

Amran, Farzadnia and Abang Ali (2015) — properties and applications

The standard general review. Covers composition, production, fresh and hardened properties, durability and the application range, with the property-versus-density relationships that most subsequent work builds on. The best single starting point for someone new to the material.

Ramamurthy, Kunhanandan Nambiar and Indu Siva Ranjani (2009) — a classification of studies

Rather than surveying properties directly, this paper organises the research field itself: what has been studied, by what methods, and where the results conflict. Useful for understanding why published figures for the same property vary as widely as they do.

Fu, Wang, Wang and Li (2020) — state of the art and state of the practice [open access]

The most recent broad review, and freely available. Bridges laboratory findings and industrial practice, which is where most of the useful detail on production control lives.

Amran and co-authors (2020) — fibre-reinforced foamed concretes [open access]

A focused review of what fibre reinforcement achieves in foam concrete. The short version: fibres improve tensile and flexural behaviour and control shrinkage cracking, but do relatively little for compressive strength, and above roughly 0.5 % by volume they interfere with the flow the material depends on. Published in Materials and freely available.

Void structure and strength

Nambiar and Ramamurthy (2007) — air-void characterisation

The paper that established quantitatively what practitioners suspected: at constant density, compressive strength is governed by the void system rather than by the mix proportions. Larger average void size, a wider size distribution and a higher proportion of merged voids all reduce strength. This is the technical basis for the claim that the foaming agent is the largest single lever on strength at a given density.

Hilal, Thom and Dawson (2015) — void structure with and without additives

Extends the void-structure work to mixes containing superplasticiser and fly ash, and reports how those additions change void size, shape and connectivity — and therefore strength — independently of their effect on the paste.

Jones and McCarthy (2005) — foamed concrete as a structural material

An assessment of how far foam concrete can be pushed toward structural use. Sets out the strength and stiffness that are realistically achievable in the higher density range, and the shrinkage and durability constraints that limit it. Still the clearest statement of where the structural ceiling sits. The shrinkage figures it cites — and the 0.1 to 0.35 % drying shrinkage range quoted throughout this site — trace to Nambiar and Ramamurthy (2009), who measured foam concrete shrinkage behaviour directly and identified falling density as the strongest single driver of the effect.

Why foam survives in cement paste

Foam concrete depends on a paste that is fluid enough to accept foam without shearing it, yet stiff enough to hold the bubbles in place before the material sets. That balance has a physical basis, and it has been measured directly rather than inferred from production experience.

Feneuil, Pitois and Roussel (2017) — surfactants and paste yield stress

Measured how surfactants — the same class of compound used to generate foam — change the yield stress of cement paste. Yield stress, not viscosity, is the property that decides whether a bubble embedded in the paste can rise under buoyancy: below a critical paste yield stress a bubble drains and rises through the paste; above it, the paste resists the small stress the bubble exerts and holds it in place.

Feneuil, Roussel and Pitois (2019) — an optimal paste yield stress

Follows directly from the 2017 result with a search for the optimum. A paste needs to be soft enough to be worked without excessive shearing during mixing, but stiff enough that once the foam is folded in it does not drain or coalesce before the paste sets. Too low a yield stress and the bubbles rise and drain during the dormant period; too high, and the foam cannot be mixed in without being sheared apart in the attempt. The paper reports the optimal paste yield stress sitting between those two failure modes.

Together these two papers give a physical mechanism for what the technology page describes qualitatively as the dormant period problem — the interval after mixing and before set during which the void structure is at its most vulnerable. They also explain why two different correctives both work. A viscosity modifier raises the paste's yield stress directly, toward the stable range these papers identify. An accelerator does not touch yield stress at all; it shortens the dormant period itself, so the paste spends less time in the window where instability matters. Each addresses the same mechanism from a different side. See foaming agents and additives for how these correctives are applied.

Foaming agents

Panesar (2013) — synthetic versus protein foaming agents

The principal published comparison of the two established agent families, examining fresh and hardened properties of cellular concrete produced with each. Confirms the practical experience that protein agents give a finer void structure and higher strength at equal density, while synthetic agents offer better consistency and yield. Note that this covers protein hydrolysate and synthetic surfactant agents; comparable peer-reviewed data on biotechnological protein agents is not available in the open literature.

Indu Siva Ranjani and Ramamurthy (2010) — foam generated using sodium lauryl sulfate [open access]

A detailed study of the foam itself rather than the concrete: how foam density, stability and drainage vary with surfactant concentration and generation method. Freely available, and the best introduction to foam characterisation as a measurable discipline.

Curing and binder composition

Falliano, De Domenico, Ricciardi and Gugliandolo (2018) — curing conditions and compressive strength

A large experimental programme varying curing conditions, cement type, foaming agent and dry density independently. Establishes how strongly the curing regime governs the outcome, which is the evidence base for treating curing as a specification item rather than a site detail. Directly relevant to the steam curing regimes discussed on the technology page.

Kearsley and Wainwright (2001) — high fly ash content

Quantifies the effect of substituting a large proportion of cement with fly ash. Early strength falls, long-term strength rises, and the 90-day and later figures can exceed the equivalent neat cement mix. The standard reference for anyone considering high-volume fly ash foam concrete.

Thermal behaviour

Othuman Mydin (2022) — density, porosity and void size

Separates the influence of density, total porosity and void size on thermal conductivity, which are usually conflated. The finding that at equal density a finer void structure conducts less has a direct practical consequence: the same foaming agent choice that raises strength also improves thermal performance.

Durability

Indu Siva Ranjani and Ramamurthy (2012) — sulfate environments

Examines foam concrete exposed to sulfate solutions across binder compositions. Behaviour depends strongly on the binder: fly ash-containing mixes perform considerably better than neat Portland cement mixes, consistent with sulfate resistance in conventional concrete.

Kearsley and Wainwright (2001) — porosity and permeability

A separate 2001 paper by the same authors, not to be confused with their fly ash study above, established the distinction that governs transport in foam concrete: the artificial voids the foam creates are mostly closed and are not, by themselves, a significant transport path, while the capillary porosity of the paste between them is what actually governs water and gas movement. It is this capillary porosity, not the void content, that determines absorption, carbonation rate and sulfate ingress.

Recent work

Two papers published in 2024 are worth flagging separately from the rest of this bibliography, because they are recent enough that their findings have not yet been independently replicated — worth reading, but not yet worth treating as settled in the way the older, more established results above can be.

Xu and Garrecht (2024) [open access] examined how mixing technique and material composition affect compressive strength and thermal conductivity in ultra-lightweight foam concrete, at densities well below the structural-grade classes discussed elsewhere on this site. This is exactly the range where published data is thinnest and where production practice has historically relied on in-house trial mixes rather than the literature. Anyone specifying or producing below roughly 200 kg/m³ should read it directly rather than relying on this summary, since the interaction between mixing method and composition it reports is more detailed than a digest can usefully compress.

Dang and co-authors (2024) is the clearest current example of foam concrete work on a binder other than Portland cement: an alkaline activator and a solid precursor used in place of cement, applied to the same foam-characterisation questions this bibliography covers for conventional mixes. Alkali-activated systems bring their own curing chemistry and durability profile, neither of which can be assumed to match the Portland-cement figures quoted throughout this site — a caution that applies to this paper's findings as much as to the broader alkali-activation literature it belongs to.

Open questions

Four gaps in the literature are worth naming, because they are places where a specifier should require project-specific testing rather than rely on published data.

• Biotechnological foaming agents. The performance claims — very fine foam, high stability, reliable densities down to 75 kg/m³ — come from manufacturers' technical literature. There is no independent comparative study against protein hydrolysate and synthetic agents in the peer-reviewed record.
• Long-term behaviour. Most published data stops at 28 or 90 days. Field evidence at 10 and 20 years, particularly for carbonation depth and for material in permanently saturated conditions, is thin.
• Reinforced foam concrete. The corrosion protection actually required for embedded steel in a material with low alkaline reserve and fast carbonation is not settled.
• Additive manufacturing. Published work on foam concrete for 3D printing remains at laboratory scale, and the conflict between rapid stiffening and foam stability has no general solution yet.

Full bibliography

Last reviewed: 7 September 2026