Abstract: A quadratic regression model was constructed based on the response surface methodology using water/binder ratio, sodium silicate modulus, alkali equivalent (Na₂O content) as independent variables and 28 d compressive strength, 28 d flexural strength, setting time and fluidity as response target values. The model was analyzed for variance and significance, and the optimal mix ratio was determined. The results show that the most significant effect of water/binder ratio on compressive strength, flexural strength, setting time and fluidity, followed by sodium silicate modulus and alkali equivalent, there is a more significant interaction between the factors, in which the interaction between water/binder ratio and sodium silicate modulus is the most significant. The optimal mix ratio is 0.39 water/binder ratio, 1.38 sodium silicate modulus and 4.7% (mass fraction) alkali equivalent, and the error between the model prediction and the experimental results is less than 10%. The high accuracy of this regression fitting model further confirms the effectiveness of the response surface methodology in determining the optimal mix ratio parameters of alkali-activated materials.

Key words: alkali-activated material, response surface methodology, compressive strength, flexural strength, setting time, fluidity