Adhesion of glass to steel using a geopolymer

Abstract

Many potential and actual uses for geopolymers (GPs) have been listed, such as waste encapsulation, refractories, fire resistant paneling, sewer pipes, building products, and acid-resistant coatings [1]. Other proposed technical uses include fire protection coatings as applied to concrete [2]. Understanding the nature of bonding of GPs to other materials is desirable if GPs are to be applied as protective coatings or adhesives. Two common methods of making GPs are to add fly ash or metakaolinite (MK: kaolinite heated to ∼750 ◦C to render it amorphous) to concentrated alkali solutions for reaction and subsequent polymerization to take place. The GPs produced consist of amorphous to semi-crystalline three-dimensional aluminosilicate networks [3]. For industrial applications, the use of fly ash as a precursor gives a cost advantage over MK. However to understand the science of the polymerization process, the use of the latter is preferred. It has been reported that ball milling the aluminosilicate precursors up to 4 hr improves the compressive strength of GPs [4]. In the present work, we attrition-milled the MK, which is a more effective method of milling than ball milling, to determine the effect on the compressive strength. Furthermore, we examined the adhesion of stainless steel to glass using GPs derived from the milled MK and determined the interfacial fracture energy by a fracture mechanics approach. The MK was produced by heating kaolinite (Kingwhite 80, Unimin, Australia) at 750 ◦C for 15 hr in air. An MKwater slurry was attrition-milled at 300 rpm with 5 mm zirconia balls for 30, 60 and 120 min. The milled slurry was dried in a stainless steel pan at 40 ◦C for 5 days. The particle size distributions of each batch, including the unmilled sample were determined using laser diffraction techniques (Mastersizer 2000, Malvern Instruments Ltd., UK) and are shown in Fig. 1. The MK exhibits a bimodal distribution and the amount of fines (∼0.18 μm) increased with milling time. The 30 min milling time reduced the coarse particle size from 8.7 to 4.4 μm and doubled the volumes of fines. A further 30 min again reduced the coarse particles to 2.9 μm and increased the fines. Additional milling for 60 min did not