With increasing global emphasis on sustainable development, the CO₂ emissions from concrete as the most widely used building material have attracted much attention. Statistics show that cement production accounts for approximately 8% of global CO₂ emissions. Low-carbon concrete minimizes lifecycle CO₂ emissions through optimized cementitious materials, use of solid waste, and low-carbon technique (e.g., carbon capture), while still meeting engineering performance standards. However, incorporating recycled aggregates or mineral admixtures, although reducing CO₂ emissions, often leads to a deterioration in mechanical properties and durability of concrete. Therefore, how to effectively balance reduced CO₂ emissions while maintaining the engineering performance of concrete remains a challenge in low-carbon concrete mix designs.

The complex interactions among multiple components of concrete materials make mix proportion design highly complex. Conventional trial-and-error methods show obvious deficiencies in efficiency, cost and precision. Previous research established some performance prediction models based on machine learning methods, which often focused on single performance indicators. There is little systematic research on multi-objective collaborative optimization of mechanical properties. This failure to adequately balance environmental benefits with performance hinders the effective use of AI in designing low-carbon concrete. The result demonstrates that optimizing the proportions of recycled aggregates and supplementary cementitious materials can effectively limit carbon emissions to the range of 240–260 kg CO₂/m³ when the mechanical property requirements (i.e., compressive strength >50 MPa) and chloride ion permeability resistance (i.e., electric flux <500 C) are satisfied. This study was to offer an approach for achieving a balanced design between material performance and CO₂ emissions in low-carbon concrete, and to promote the application of Artificial Intelligence (AI) technology in its mix proportion design.

This study introduced an AI-based intelligent design optimization method for low-carbon concrete. The process started with data preprocessing, including normalization and outlier treatment, to enhance dataset quality. Subsequently, various models, optimized via different hyperparameter tuning methods, were benchmarked for high-precision prediction of key performance indicators. The SHapley Additive Explanations (SHAP) analysis was employed to elucidate the impact of mix proportion parameters on concrete performance. The The Non-dominated Sorting Genetic Algorithm Ⅲ (NSGA-Ⅲ) algorithm was then utilized for multi-objective optimization of CO₂ emissions, cost, strength, and chloride ion penetration resistance. Two indices, C_i (Carbon intensity index) and Q_i (Chloride ion penetration carbon efficiency index), were introduced to quantify the balance between environmental impact and concrete performance. These indices were subsequently employed to evaluate the solutions from the multi-objective optimization, guiding the decision-making process towards a comprehensive and synergistic optimization of material properties and low-carbon characteristics.

During data preprocessing, normalization addresses dimensional inconsistencies among feature variables, enhancing their comparability. Furthermore, the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm effectively identifies specific outliers (i.e., 43 related to compressive strength and 4 concerning electric charge passed), leading to a marked improvement in overall data quality. Comparison of machine learning models shows that eXtreme Gradient Boosting (XGB) model exhibits a superior performance in all prediction tasks, with Particle Swarm Optimization (PSO) achieving R² of 0.91 for compressive strength and 0.89 for electric charge passed on test sets. Interpretability through SHAP analysis reveals that cement content and water content are the main factors affecting compressive strength, while superplasticizer dosage is a key factor affecting electric charge passed. For carbon efficiency indices, fly ash replacement of cement is an important approach to reduce C_i and Q_i, while the negative impact of recycled aggregates on performance often exceeds their carbon reduction benefits. Multi-objective optimization results indicate significant trade-offs among compressive strength, durability, CO₂ emissions, and cost. Decision schemes based on carbon efficiency indices further confirm that CO₂ emissions can be effectively reduced, while ensuring concrete performance via scientifically proportioning supplementary cementitious materials, and optimizing water-binder ratio and superplasticizer content.

This study highlighted a significant potential of an AI-driven approach for designing low-carbon concrete, effectively balancing engineering performance with environmental impact. This study could provide valuable insights into achieving tailored concrete properties via utilizing predictive modeling, parameter influence analysis with SHAP, and NSGA-Ⅲ for multi-objective optimization. The proposed intelligent design method incorporating carbon efficiency indices could offer a practical strategy for developing low-carbon concrete, promoting sustainable practices in the construction industry.

With the ongoing expansion of global housing and infrastructure, cement as an essential component of concrete is projected to reach a global demand of 468 million tonnes by 2050, contributing approximately 6%–7% of total CO₂ emissions. Reducing carbon emissions in cement and concrete production is thus critical for achieving carbon neutrality goals. In the central and western regions of China, a growing demand for large infrastructure projects such as dams, tunnels, and mountain bridges increasingly relies on the use of mass concrete. However, a great heat release from cement hydration with a low thermal conductivity leads to internal heat accumulation and significant temperature differentials. These thermal gradients increase cracking risk and compromise long-term durability. Conventional strategies like low-heat cement, mineral admixtures, and chemical retarders are adopted to mitigate early hydration heat. Nevertheless, these approaches are limited due to high production costs, reduced strength, or insufficient control over heat evolution rate. Recently, carbon capture, utilization, and storage (CCUS) technologies, particularly semi-dry carbonation, show a promising potential for reducing carbon emissions and regulating hydration heat, especially for high-alkaline materials. In this study, semi-dry carbonation pretreatment was applied to Portland cement, forming a surface layer of CaCO₃ that serves as a physical barrier to suppress early hydration. The effect of semi-dry carbonation on the hydration behavior and microstructure was investigated. The results could provide some insights into applying CCUS technology for hydration heat control in cement-based materials.

A P·I 42.5 Portland cement (Fushun Cement Co., Ltd., China) was used. Semi-dry carbonation was performed in a NELD-CA070 carbonation chamber using simulated kiln gas containing 20% CO₂ at a RH of (70% ± 5%) and a temperature of (20 ± 2) ℃. The cement was loosely spread in trays and carbonated for 2, 4 h, or 8 h, respectively, under manually stirring for every 30 min. The resulting samples were designated as C2, C4, and C8, respectively, while the uncarbonated sample was labeled as C0. All the samples were dried at 40 ℃ for 24 h after carbonation.

Cement pastes were prepared at a water-to-cement ratio of 0.4 and cured under standard conditions (i.e., 20 ± 1 ℃, 95 ± 5% RH). For the determination of compressive strength, cube specimens (40 × 40 × 40 mm) were tested at 1, 3, 7 d, and 28 d, respectively. For the microstructural analysis, fresh pastes were cast in sealed vials, demolded after 24 h, immersed in water, and cured until the specified age. Hydration was stopped by solvent exchange with isopropanol, followed by vacuum drying and grinding to a fine powder with the particle size below 80 μm.

The early hydration heat evolution was measured by isothermal calorimetry (TAM Air). The phase composition was determined by isothermal calorimetry, thermogravimetric analysis (TGA), X-ray diffraction (XRD), and Fourier transform infrared spectroscopy (FTIR). The pore solution alkalinity was assessed through pH measurement and elemental analysis by inductively coupled plasma optical emission spectrometry (ICP-OES). The pore structure was characterized by mercury intrusion porosimetry (MIP), and the microstructure was determined by scanning electron microscopy (SEM).

The semi-dry carbonation alters the phase composition and microstructure of cement. The TG analysis shows a weakened gypsum decomposition peak and an intensified CaCO₃ decomposition, indicating that gypsum participates in carbonation. The mild carbonation (i.e., 2–4 h) increases CH content due to enhanced pre-hydration. CO₂ sequestration reaches 1.60%, 2.27%, and 3.62% after 2, 4 h, and 8 h, corresponding to 3.62%, 5.16%, and 8.23% CaCO₃ formation, respectively. As carbonation progresses, C₃S, C₂S, C₃A, C₄AF, and gypsum decrease, while calcite and amorphous silica increase, driven by reactions with CO₂ and moisture. The FTIR spectra show strengthened C—O stretching (1410 cm^–1), weakened H₂O peaks (3527 cm^–1), and intensified SO₄^2– bands (1124 cm^–1 and 1197 cm^–1), indicating interactions among gypsum, CO₂, and alkali oxides. The SEM images reveal roughened clinker surfaces caused by the deposition of carbonation products such as calcite and silica gel.

The semi-dry carbonation significantly reduces early compressive strength, with 1-d reductions of 31.2%, 77.5%, and 89.6% after 2, 4 h, and 8 h, respectively. However, this effect diminishes over time, with less than 5.4% strength loss at 28 d. Carbonation extends the induction period and reduces hydration heat, lowering cumulative heat by 67.2% at 10 h. Early hydration degrees and chemically bound water also decrease, corresponding to reduced CH content and early strength. Despite this, hydration products at < 400 ℃ are largely unaffected, having the minimal difference in 28-d strength. Early-age pore solution alkalinity declines significantly. In the C8 group, Na⁺ and K⁺ concentrations at 1 d are reduced by 60% and 40.7%, and the pH value is decreased from 13.52 to 13.18. At 28 d, the pH value becomes 13.5, indicating no risk to reinforcement passivation. At this age, total porosity increases slightly from 21.1% to 22.8%, with a minor rise in capillary and gel pores. The SEM images and MIP analyses confirm similarly dense microstructures across all the samples, indicating that the calcite layer is eventually penetrated and has no adverse long-term effect.

This study showed that emi-dry carbonation could alter the phase composition and microstructure of Portland cement. After 2-h, 4-h, and 8 h-carbonation, CO₂ sequestration increased to 1.60%, 2.27%, and 3.62%, respectively, accompanied by a reduction in clinker minerals and gypsum and formation of amorphous silica and calcite layers on cement particle surfaces. These calcite layers could effectively suppress early hydration by reducing the contact area between water and reactive clinker phases. As a result, cumulative heat release at 3 d was decreased by 14.7%, 35.1%, and 58.1%, respectively, leading to a lower early compressive strength, a pore solution alkalinity, and a looser microstructure. However, the long-term effect was limited. At 28 d, the continuous formation of hydration products penetrated the calcite layer, and hydration degree and product quantity were reduced by less than 7%, and the compressive strength, pH value, and porosity were altered by no more than 5.4%, 0.1, and 8.1%, respectively.

Concrete infrastructure in China has entered a large-scale phase of maintenance and repair, with organic adhesives being essential materials for enhancing structural durability. Enhancing the long-term effectiveness of adhesion between these adhesives and cementitious materials is crucial for extending the service life of concrete structures. This study reviews the primary types of organic adhesives used for infrastructure repair work, including epoxy resins, polyurethanes, and silicone adhesives, and discusses the performance advantages of their various applications. The adhesion mechanisms are determined in terms of mechanical interlocking, intermolecular interactions, and thermodynamic interactions, identifying the main sources and driving forces behind adhesion. The degradation mechanisms of interfacial adhesion under moisture, salt solutions, adhesive aging, and cyclic load coupling are analyzed. Finally, the study summarizes various techniques for enhancing interface adhesion, providing insights and innovative approaches for the design of organic adhesives and adhesion durability. The main conclusions are as follows:

The adhesion mechanism of organic adhesives to cement-based materials is primarily considered from three perspectives: mechanical interlocking, intermolecular interactions, and thermodynamic interactions. Compared to the roughness of the interface, the penetration of organic adhesives into the cement-based material promotes mechanical interlocking more effectively. In the absence of interface modification, hydrogen bonds formed between the adhesive and the substrate dominate the intermolecular interactions. Moreover, hydrophobic interactions between the adhesive and the substrate play a significant role in driving the interface adhesion process.

Moisture is the predominant factor responsible for debonding at the interface between organic adhesives and cement-based materials. Moisture molecules replace the hydrogen bonds formed during the adhesion process, while aggressive ions accelerate the degradation of adhesion. Furthermore, moisture and aging effects gradually degrade the mechanical properties of the organic adhesives, weakening the interfacial adhesion strength. Environmental temperature and vibration loads also contribute to the deterioration of adhesion performance.

Techniques such as interface roughening, jet treatment, and coating with silane coupling agents can significantly enhance the adhesion strength between organic adhesives and cement-based materials. These methods primarily alter the surface roughness, surface tension, and surface-active groups of the substrate. Additionally, reducing the viscosity of the organic adhesive over time, increasing the content of polar groups in the adhesive molecular structure, and improving the wetting of both the adhesive and substrate phases can further strengthen the interface adhesion.

Both domestic and international scholars have conducted extensive research on the adhesive performance between organic adhesives and cement-based materials from a multi-scale perspective. Significant progress has been made in understanding adhesion mechanisms and enhancement techniques. However, research on the driving forces of adhesion at the micro-nano scale remains relatively underdeveloped, especially when compared to the adhesion mechanisms observed with metal materials. Furthermore, there are still several pressing issues regarding the environmental adaptability of adhesives and cement-based materials, which demand higher standards for their long-term reliability in practical applications. Although researchers have gained a deeper understanding of the mechanism of mechanical interlocking and clarified the corresponding intermolecular interactions, there is still a need to further analyze the individual contributions of mechanical interlocking and intermolecular interactions to overall adhesion. From a thermodynamically driven molecular adhesion mechanism perspective, optimizing the molecular structure of adhesives remains an area worth exploring. Existing studies typically link macroscopic adhesion strength directly with intermolecular interactions at the nano-scale, such as hydrogen bonds, while overlooking the important role of the mesoscopic mechanical properties of the adhesive interface. Therefore, it is essential to develop a method for characterizing the mesoscopic mechanical properties of the adhesive interface, taking into account the phase composition characteristics of cement-based materials. Furthermore, a more thorough understanding of the quantitative relationship between these properties, macroscopic adhesion strength, and nano-scale chemical composition is needed. The degradation mechanisms of interfacial adhesion of organic adhesives under coupled working conditions remain unclear. How factors such as vibration loads, temperature, and moisture coupling accelerate debonding of organic adhesives from cement-based materials requires further investigation. In addition, the time-dependent damage behavior of interface adhesion under service conditions should be explored in greater detail. This research is crucial for identifying key parameters that can enhance the durability of concrete structures, thereby providing technical support for ensuring the long service life of infrastructure.