Improving the workability and rheological properties of Engineered Cementitious Composites using factorial experimental design


Creative Commons License

Sahmaran M., Bilici Z., Ozbay E., Erdem T. K. , Yucel H. E. , Lachemi M.

COMPOSITES PART B-ENGINEERING, cilt.45, sa.1, ss.356-368, 2013 (SCI İndekslerine Giren Dergi) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 45 Konu: 1
  • Basım Tarihi: 2013
  • Doi Numarası: 10.1016/j.compositesb.2012.08.015
  • Dergi Adı: COMPOSITES PART B-ENGINEERING
  • Sayfa Sayıları: ss.356-368

Özet

In the development of Engineered Cementitious Composites (ECC), micromechanics-based design theory is adopted to properly select the matrix constituents, fiber, and fiber-matrix interface properties to exhibit strain hardening and multiple cracking behaviors. Despite the micromechanics design constraints, practical applications show that the workability and rheological properties of matrix can affect the fiber dispersion uniformity, which have also direct concerns on composite mechanical properties. For this reason, in this research, parameters of micromechanics-based optimized ECC mixture design, which most possibly affecting the workability and rheological properties, are investigated. An experimental program that contains 36 different ECC mixtures was undertaken to quantitatively evaluate the combined effects of the following factors on workability and rheological properties: water-binder (w/b), sand-binder (s/b), superplasticizer-binder (SP/b) ratios and maximum aggregate size (D-max). A mini-slump cone, a Marsh cone and a rotational viscometer were used to evaluate the workability and rheological properties of ECC mixtures. Compressive strength and four point bending tests were used for mechanical characteristics of ECC mixtures at 28 days. The effects of studied parameters (w/b, s/b, SP/b and D-max) were characterized and analyzed using regression models, which can identify the primary factors and their interactions on the measured properties. Statistically significant regression models were developed for all tested parameters as function of w/b, s/b, SP/b and D-max. To find out the best possible ECC mixture under the range of parameters investigated for the desired workability and mechanical characteristics, a multi-objective optimization problem was defined and solved based on the developed regression models. Test results indicate that w/b, s/b and SP/b parameters affect the rheological and workability properties. On the other hand, for the range of studied aggregate sizes, D-max is found to be statistically insignificant on the rheological and workability properties of ECC. (c) 2012 Elsevier Ltd. All rights reserved.