Abstract
Developing a greater understanding of kaolinite dehydroxylation upon calcination is crucial for several industrial applications, including cements. Aluminium coordination in meta-kaolinite indicates the extent of its dehydroxylation and its potential chemical reactivity, and it is typically determined using 27Al magic angle spinning (MAS) nuclear magnetic resonance (NMR) spectroscopy. This technique however presents limitations for Fe-rich materials, given the magnetic properties of Fe ions and minerals containing Fe. In this study, the effect of calcination on Al coordination was assessed in a low-Fe clay used as a reference system, and a Fe-rich clay. Al coordination in the low-Fe clay was quantified via 27Al MAS NMR spectra deconvolution, using data collected at 9.4 T and 11.7 T. Energy dispersive X-ray spectroscopy (EDX) maps and electron energy loss spectroscopy (EELS) measurements were carried out in a scanning transmission electron microscope (STEM) on both clays. Al K-edge EEL spectra showed distinguishable 4/5-fold Al and 6-fold Al sites in both clay types. Differences in line-profile indicated a higher proportion of 4/5-fold Al in kaolinite in the Fe-rich clay compared to the low-Fe clay. Conversely, the Fe-rich clay contained a lower proportion of 4/5-fold Al in meta-kaolinite after calcination, relative to the low-Fe clay. These differences are consistent with the greater structural disorder of the meta-kaolinite identified in the Fe-rich clay by X-ray diffraction and the geological origins of both clays. Overall, this study demonstrates the potential of EELS to provide information about Al coordination for individual kaolinite and meta-kaolinite particles.
Supplementary materials
Title
Supplementary Information
Description
Table S1: Specific surface area of the clay samples used in this study.
Figure S1: Deconvoluted 27Al MAS NMR spectra for a) low-Fe clay 9.4 T; b) low-Fe clay 11.7 T; c) calcined low-Fe clay 9.4 T; d) calcined low-Fe clay 11.7 T.
Table S2: Fitting parameters for 27Al MAS NMR deconvolution.
Actions