A laboratory test can be faster without being a direct clock for field deterioration. That distinction matters when engineers use accelerated alkali–silica reaction (ASR) testing to judge whether fly ash adequately limits concrete expansion. A study published in Materials on 15 October 2026 shows how extending an accelerated test changes the apparent replacement level needed to pass.
This Civil Insight article reviews a peer-reviewed paper by Xiaojun Liu, Yan Zhang, Guangye Lv, Wei Li, Xiaojun Huang and Min Deng. It is an editorial assessment of their published research, not an independently peer-reviewed study or an Australian mix-design recommendation.
What the researchers tested
The team investigated four reactive tuff aggregates from Cangnan, Zhejiang, China, with a non-reactive limestone reference. They used one cement and one Class F fly ash, testing six cement-replacement levels by mass: 0%, 15%, 20%, 25%, 30% and 35%. That gives 24 tuff–replacement combinations, rather than a survey of every available aggregate or supplementary cementitious material.
They compared three laboratory approaches. The accelerated mortar-bar test used crushed aggregate in 25 × 25 × 285 mm bars immersed in 1 mol/L sodium hydroxide at 80°C. The conventional concrete-prism test used 75 × 75 × 285 mm prisms with coarse aggregate, stored at 38°C and approximately 99% relative humidity for 728 days. An accelerated concrete-prism test used prisms immersed in 1 mol/L sodium hydroxide at 80°C.
The study assessed the 28-day mortar-bar results against a 0.10% expansion index, and the concrete-prism results against 0.040%. Those are the criteria adopted in this experimental comparison. They should not be read as a substitute for the test method and acceptance requirements specified for an Australian project.
The answer moved with the test age
In the 728-day conventional prism test, the four mixtures without fly ash expanded between 0.037% and 0.064%. The 15% replacement mixtures expanded between 0.018% and 0.029%, all below the study’s 0.040% index. Notably, one unmitigated tuff mixture was already below that index; the starting behaviour was not identical across the four aggregates.
The 28-day mortar-bar assessment also identified 15% as the lowest tested replacement level passing for all four tuffs. The accelerated concrete prisms gave a different sequence. At 182 days, 20% replacement kept all four below 0.040%, with a maximum expansion of 0.036%. At 273 days, the corresponding lowest tested level rose to 30%, with a maximum of 0.037%.
At 364 days, no tested replacement level up to 35% kept every tuff mixture below the same index. These percentages describe the discrete levels investigated; they are not exact optimised dosages. The result demonstrates why an expansion threshold without a specified assessment age can produce an ambiguous decision.
What the photographs add
The authors also examined SA3 concrete after two years. Their laboratory photographs show reacted aggregate and deposits in the moist-air specimens, and reaction regions in specimens exposed to hot sodium hydroxide. The photographs relate to the study itself, rather than an unrelated construction site.
The paper reports a smaller visibly reacted depth as fly-ash replacement increased in the hot-alkali specimens. This supports the observed mitigation trend, but the imaging concerns a particular aggregate and severe laboratory exposure. It does not establish how quickly a structure in Melbourne would develop damage.
How reliable and transferable is the evidence?
A useful strength is the side-by-side comparison with a 728-day conventional prism benchmark. Holding the aggregate sources and replacement levels within one investigation makes disagreement between methods easier to interpret than comparing unrelated studies. The accompanying microstructural observations provide a second line of evidence about reaction.
The limits are substantial. Four tuffs, one cement and one fly ash cannot represent the full variability of Australian concrete supply. The paper reports replicate means and standard deviations, but does not clearly state specimen counts per group, limiting the reader’s ability to assess precision. It declares two authors’ employment by China General Nuclear Power Corporation and reports no external funding.
The benchmark is another laboratory test, not validation against monitored structures. Hot external alkali changes exposure severity as well as reaction speed. Longer immersion therefore cannot simply be translated into a fixed number of service years. The authors themselves identify the need to connect accelerated assessment with field conditions and broader materials.
Civil Insight’s practical interpretation
For construction professionals, the defensible lesson is about specifying and interpreting evidence. Ask a test report to state the materials, specimen method, conditioning, assessment age and criterion together. If two reports disagree, check those elements before treating one result as a contradiction or automatically adopting the larger replacement percentage.
The study supports investigating accelerated prism testing as a screening tool for these materials. It does not demonstrate universal dosage selection, certification, Australian Standards compliance or complete durability performance. Project acceptance still needs evidence relevant to the proposed materials and exposure.
Original paper
Liu, X.; Zhang, Y.; Lv, G.; Li, W.; Huang, X.; Deng, M. (2026). An Accelerated Concrete Prism Test for Evaluating Fly Ash Mitigation of Alkali–Silica Reaction in Reactive Tuff Aggregates. Materials, 19(20), 4335. Published online 15 October 2026. DOI: 10.3390/ma19204335. Full published article reviewed; publisher record checked on 8 November 2026, with no linked correction or retraction identified.
Featured photograph
SA3 concrete without fly ash after two years in moist air at 38°C. Laboratory photographs, Figure 10. Photograph: Xiaojun Liu et al., Materials (2026) / CC BY 4.0.
