Nature and structure of aluminum surface sites grafted on silica from a combination of high-field aluminum-27 solid-state NMR spectroscopy and first-principles calculations
AuthorsKerber, Rachel Nathaniel
Florian, Pierre A.
KAUST DepartmentKAUST Catalysis Center (KCC)
KAUST Grant NumberUK-00017
Online Publication Date2012-04-09
Print Publication Date2012-04-18
Permanent link to this recordhttp://hdl.handle.net/10754/562158
MetadataShow full item record
AbstractThe determination of the nature and structure of surface sites after chemical modification of large surface area oxides such as silica is a key point for many applications and challenging from a spectroscopic point of view. This has been, for instance, a long-standing problem for silica reacted with alkylaluminum compounds, a system typically studied as a model for a supported methylaluminoxane and aluminum cocatalyst. While 27Al solid-state NMR spectroscopy would be a method of choice, it has been difficult to apply this technique because of large quadrupolar broadenings. Here, from a combined use of the highest stable field NMR instruments (17.6, 20.0, and 23.5 T) and ultrafast magic angle spinning (>60 kHz), high-quality spectra were obtained, allowing isotropic chemical shifts, quadrupolar couplings, and asymmetric parameters to be extracted. Combined with first-principles calculations, these NMR signatures were then assigned to actual structures of surface aluminum sites. For silica (here SBA-15) reacted with triethylaluminum, the surface sites are in fact mainly dinuclear Al species, grafted on the silica surface via either two terminal or two bridging siloxy ligands. Tetrahedral sites, resulting from the incorporation of Al inside the silica matrix, are also seen as minor species. No evidence for putative tri-coordinated Al atoms has been found. © 2012 American Chemical Society.
SponsorsThis publication is based on work supported by Award No. UK-00017, made by King Abdullah University of Science and Technology (KAUST), and by the TGE RMN THC Fr3050. The authors thank the PSMN at ENS of Lyon for the attribution of CPU time and Raphael Wischert for sharing data (ref 60).
PublisherAmerican Chemical Society (ACS)