A Coupled Chemo-Mechanical and Life-Cycle Framework for Oyster-Shell-Derived Cementitious Materials in Aggressive Infrastructure Environments

Oyster-shell powder is promoted as a low-carbon cement substitute, but joint carbon and durability benefits are usually asserted rather than tested. This paper asks when shell-derived concrete (shell replacing cement) delivers a carbon cut with non-inferior service life relative to ordinary Portland cement (OPC) concrete. A coupled hydration–transport–damage–carbon model is run through paired Monte Carlo analysis of a pavement, a dam and a marine culvert. Published data reject the original inert-filler strength model. An effective strength-efficiency coefficient k_f calibrated on one scallop-shell dataset does not generalize to independent oyster-shell data (mean absolute error 0.12 against 0.15 for the inert model; root-mean-square error 0.17 against 0.15), and k_f ranges from -0.40 to 1.36 across four studies, so the primary results use a pooled random-effects distribution. At 25% replacement and constant water-to-binder ratio the median carbon change is −21% with strength 0.88 of OPC; with strength matched it is −15% (P10–P90 −22 to +10%). In the primary configuration (pooled strength-only k_f, sampled diffusivity mapping, OPC reference) the probability of a carbon cut of at least 5% with non-inferior life is at most 0.21 for the pavement, 0.14 for the dam and 0.03 for the marine culvert, and zero against a 50% slag blended cement; other, equally unmeasured transport assumptions give 0.18 to 0.68 for the pavement. All probabilities are conditional scenario results, not empirical probabilities for real infrastructure, and the study is a falsifiable computational screening framework, not a validated service-life model.