Response surface modeling of boron adsorption from aqueous solution by vermiculite using different adsorption agents: Box-Behnken experimental design

Demircivi P., Saygili G.

WATER SCIENCE AND TECHNOLOGY, vol.76, no.3, pp.515-530, 2017 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 76 Issue: 3
  • Publication Date: 2017
  • Doi Number: 10.2166/wst.2017.200
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.515-530
  • Istanbul Technical University Affiliated: Yes


In this study, a different method was applied for boron removal by using vermiculite as the adsorbent. Vermiculite, which was used in the experiments, was not modified with adsorption agents before boron adsorption using a separate process. Hexadecyltrimethylammonium bromide (HDTMA) and Gallic acid (GA) were used as adsorption agents for vermiculite by maintaining the solid/liquid ratio at 12.5 g/L. HDTMA/GA concentration, contact time, pH, initial boron concentration, inert electrolyte and temperature effects on boron adsorption were analyzed. A three-factor, three-level Box-Behnken design model combined with response surface method (RSM) was employed to examine and optimize process variables for boron adsorption from aqueous solution by vermiculite using HDTMA and GA. Solution pH (2-12), temperature (25-60 degrees C) and initial boron concentration (508,000 mg/L) were chosen as independent variables and coded x(1), x(2) and x(3) at three levels (-1, 0 and 1). Analysis of variance was used to test the significance of variables and their interactions with 95% confidence limit (alpha = 0.05). According to the regression coefficients, a second-order empirical equation was evaluated between the adsorption capacity (q(i)) and the coded variables tested (x(i)). Optimum values of the variables were also evaluated for maximum boron adsorption by vermiculite-HDTMA (HDTMA-Verm) and vermiculite-GA (GA-Verm).