Journal of Building Engineering 31 (2020) 101351 Available online 19 March 2020 2352-7102/© 2020 Elsevier Ltd. All rights reserved. Experimental assessment and optimization of mix parameters of fy ash-based lightweight geopolymer mortar with respect to shrinkage and strength K. Mermerdas ¸ * , Z. Algın , S ¸ . Ekmen Harran University, Faculty of Engineering, Civil Engineering Department, S ¸ anlıurfa, Turkey A R T I C L E INFO Keywords: Lightweight geopolymer mortar Compressive strength Shrinkage Optimisation ABSTRACT The aim of the study is to determine the impact of various NaOH molarities, sodium silicate-to-sodium hydroxide ratios, binder amounts and the age on the compressive strength and free shrinkage of fy ash based lightweight geopolymer mortars. After experimental evaluation, the data were used to model and optimize mix proportions of lightweight geopolymer mortars through response surface method. NaOH molarities were determined as 6 M, 8 M, and 10 M. sodium silicate-to-sodium hydroxide ratios were chosen as 0.5, 1.5, and 2.5 and binder amounts were used as 600, 700, and 800 kg/m 3 . For the same variables, except the binder amount, the compressive strengths of geopolymer pastes were also determined. The usage of lightweight aggregate as a replacement of river sand (25%) affected the results due to pore structure and its strength. Thus, it was concluded that strength and shrinkage of the samples reduced in comparison to the geopolymer mortars only including sand. The in- crease of molarity improved compressive strength of pastes and mortars. However, higher sodium silicate-to- sodium hydroxide ratio resulted in lower strength values. Geopolymer mortars with higher binder amounts and NaOH molarities showed a lower drying shrinkage strains, while resulted in increase of the autogenous shrinkage. The effect of sodium silicate-to-sodium hydroxide ratios were also observed in both shrinkage types. Besides, the optimum variables based on maximum strength and minimum shrinkage strain values were reached in optimisation study conducted with response surface method (RSM). The optimum values for the parameters of NaOH molarity, binder amount and sodium silicate-to-sodium hydroxide ratio were specifed as 9.892 M, 600 kg and 0.5, respectively. The corresponding desirability value for this optimisation was obtained as 0.833 within the acceptable range. 1. Introduction It is clear that a high level of CO 2 causes an adverse effect directly on the atmosphere and indirectly on human health. In the construction industry, the cement, which causes a considerable amount of CO 2 emission during manufacture, is used as the main binder. Therefore, as the consumption of cement increases CO 2 footprint also increases. The attempt to decrease the negative effect has prompted researchers to seek for new materials. For a long time, supplementary cementitious mate- rials such as by-products and waste materials (fy ash, slag, etc) have been utilised in concrete production. Geopolymers, also called inorganic polymers, can be both crystalline and amorphous and formed using alumino-silicate sources together with a suitable alkaline solution. Geopolymer concrete produced by the use of waste materials and solutions replacing Portland cement may have a positive impact on the environment by lesser CO 2 emissions between 40% and 90%. Hence, it will lead to a signifcant reduction in the energy quantity consumed during the cement production process [1]. There is a remarkable attention on geopolymer materials during the last decades [2–14]. For example, Shoaei et al. [12] investigated the effect of various curing temperatures and alkali solution to binder ratio (S/B) on density, compressive strength, fexural strength properties of geopolymer mortar. The optimum value for curing temperature and S/B were determined as 90 � C and 0.6, respectively. In the study of Zama- nabadi et al. [13] the infuence of using sodium hydroxide solution molarities of 8 M,12 M, 16 M and sodium silicate to sodium hydroxide (SS/SH) ratios of 1, 2.5, and 4 on ambient cured alkali-activated slag paste were investigated. They reached the conclusion that sodium * Corresponding author. E-mail address: kasim.mermerdas@harran.edu.tr (K. Mermerdas ¸). Contents lists available at ScienceDirect Journal of Building Engineering journal homepage: http://www.elsevier.com/locate/jobe https://doi.org/10.1016/j.jobe.2020.101351 Received 25 November 2019; Received in revised form 10 February 2020; Accepted 12 March 2020