水和メカニズムに基づく強度モデリングと低炭素スラグ・フライアッシュ三元系コンクリートの結合材レベル逆設計
Hydration-Mechanism-Based Strength Modeling and Binder-Level Inverse Design of Low-Carbon Slag–Fly Ash Ternary Concrete (原題)
Li-Na Zhang, Rui-Xuan Zhu, Run-Sheng Lin, Xiao-Yong Wang
🤖 gxceed AI 要約
日本語
高炉スラグとフライアッシュを大量使用した三元系コンクリートは炭素排出を大幅に削減できるが、複雑な水和反応により従来の経験的強度モデルでは信頼性が低い。本研究は毛細管水と水酸化カルシウムの連成進展に基づく統一水和モデルを構築し、1030点の実験データで有効反応寄与に基づく圧縮強度モデルを校正した。これを遺伝的アルゴリズムに組み込み、28日強度と組成制約下で結合材レベルの低炭素逆設計を実施し、混和材置換率の上限付近と低水量が排出削減に寄与することを示した。
English
High-volume slag–fly ash ternary concrete cuts carbon but complicates strength prediction. This study builds a unified hydration model (capillary water and Ca(OH)2 coupling) and calibrates an effective-reaction strength model on 1030 data points (3–365 days, 2.33–82.60 MPa). Embedded in a genetic algorithm, it performs binder-level low-carbon inverse design under 28-day strength constraints, showing boundary-dominated solutions with high replacement ratios and reduced water content. Results are theoretical binder candidates, not full mix designs.
Unofficial AI-generated summary based on the public title and abstract. Not an official translation.
📝 gxceed 編集解説 — Why this matters
日本のGX文脈において
日本の建設業はScope 3排出の削減とグリーン調達の要請が強まっており、低炭素コンクリートの設計手法はサプライチェーン排出削減に直結する。SSBJや有報でのScope 3開示を進める建設・建材企業にとって、材料レベルでの炭素最適化は説明責任と調達リスク低減に資する。
In the global GX context
Global construction decarbonization is central to net-zero pathways, and this work offers a physically interpretable, optimization-ready method for low-carbon binder design. It supports Scope 3 accounting and green procurement in the built environment, though it stops short of full LCA or field validation.
👥 読者別の含意
🔬研究者:水和モデルと強度予測を統合し、遺伝的アルゴリズムで低炭素結合材を逆設計する枠組みを提供する。
🏢実務担当者:混和材置換率の上限と低水量が強度と炭素削減の両立に有効であることを示唆する。
🏛政策担当者:低炭素コンクリートの普及には、材料設計段階からの炭素最適化を促す基準や調達指針が有効である。
📄 Abstract（原文）
Reducing carbon emissions in concrete production while improving structural performance has become a priority in the transition to carbon neutrality. High-volume fly ash and slag systems significantly reduce carbon emissions. However, their complex hydration interactions alter strength development, rendering conventional empirical strength models inadequate for reliable low-carbon binder design. To overcome this limitation, this study proposes an integrated hydration-mechanism-based framework that links strength prediction with carbon-oriented binder optimization. A unified hydration model is developed based on the coupled evolution of capillary water and calcium hydroxide, enabling a consistent description of cement hydration, slag latent hydraulic reaction, and fly ash pozzolanic reaction in ternary binder systems. A single set of material-specific kinetic parameters is applied across the investigated mixture proportions and curing ages, without recalibration for each mixture. Building on the predicted degree of reaction, a compressive strength model centered on effective reaction contributions is calibrated using 1030 experimental data points encompassing 3–365 days and 2.33–82.60 MPa. Within the calibration database, the model reproduces early-age strength reduction, later-age compensation, and strength crossover behavior induced by mineral admixtures. The strength model is subsequently embedded into a genetic algorithm framework to perform theoretical binder-level low-carbon inverse design under 28-day strength and composition-domain constraints. The optimization results reveal boundary-dominated solutions, with mineral admixture replacement ratios approaching upper limits and a reduced water content contributing to further calculated emission reduction. Because aggregates, paste volume, superplasticizer dosage, and workability are not included, the resulting binder compositions are theoretical candidates rather than complete concrete mixture designs. This work presents a physically interpretable and optimization-ready framework for binder-level low-carbon design within the adopted model domain.
🔗 Provenance — このレコードを発見したソース
- openalex https://doi.org/10.3390/buildings16183719first seen 2026-09-23 05:00:43
🔔 こうした論文の新着を逃したくない方は キーワードアラート に登録（無料・3キーワードまで）。
gxceed は公開メタデータに基づく研究支援データセットです。要約・翻訳・解説は AI 支援で生成されています。 最終的な解釈・検証は利用者が原典資料に基づいて行うことを前提とします。