• 文献标题:   Modelling and optimization of the mechanical properties of engineered cementitious composite containing crumb rubber pretreated with graphene oxide using response surface methodology
  • 文献类型:   Article
  • 作  者:   ABDULKADIR I, MOHAMMED BS, LIEW MS, WAHAB MMA
  • 作者关键词:   crumb rubber cr, graphene oxide go, engineered cementitious composite ecc, response surface methodology rsm, pretreated crumb rubber
  • 出版物名称:   CONSTRUCTION BUILDING MATERIALS
  • ISSN:   0950-0618 EI 1879-0526
  • 通讯作者地址:  
  • 被引频次:   10
  • DOI:   10.1016/j.conbuildmat.2021.125259 EA OCT 2021
  • 出版年:   2021

▎ 摘  要

One of the biggest obstacles to using crumb rubber (CR) in cementitious composites for structural applications is its adverse effects on mechanical strength. This is caused by the poor bonding between the CR and the hardened cement paste. The problem persists despite several CR pretreatment methods proposed in different studies. This research aims to investigate a novel approach to CR pretreatment using graphene oxide (GO). The effect of GO concentration (GOC) and pretreated CR as a partial substitute to fine aggregate on the mechanical properties of ECC has been investigated. Using response surface methodology (RSM), 16 experimental runs having varying combinations and levels of GOC (0, 0.25, 0.5, 0.75, and 1.0 mg/ml) and CR replacement (1, 3, and 5% by volume) of fine aggregate as the independent variables have been tested for compressive, flexural, tensile strengths and strain capacity (CS, FS, TS, and TC) as the responses. Results show an increase in the mechanical strengths with higher GOC attributed to higher reactivity and better bonding at the CR and hardened cement matrix interface. At a 5% GO-CR replacement, there is a 50.3, 70.4, and 68.3% improvement in CS, FS, and TS between the control mix with 0 mg/ml GOC and the mix with 1.0 mg/ml GOC. Response surface models to predict the mechanical strengths have been developed and validated using ANOVA with R2 values ranging from 94.13 to 99.07% for all the models. Multi-objective optimization has been performed and experimentally validated, with the predicted and experimental results having<5% error.