Supporting InformationFigure S1. NH 3 temperature-programmed desorption profiles for LTA zeolite samples: Na +ionic form of bulk zeolite (NaA-0), Na + -ionic form of mesoporous zeolite (NaA-2), Ca 2+ionic form of bulk zeolite (CaA-0), and Ca 2+ -ionic form of mesoporous zeolite (CaA-2). The CaA samples exhibit ammonia desorption at high temperature as compared to NaA, which indicates that the CaA samples have stronger acid sites. The mesoporous CaA-2 zeolite exhibits only a slight decrease of desorption peak intensity at high temperature as compared to CaA-0, which is not sufficient to support the large selectivity difference in the methanol conversion reaction. For the desorption measurement, 50 mg powder sample was introduced into a quartz reactor and degassed under vacuum at 823 K. After cooling to room temperature, NH 3 gas was adsorbed for 1 h, which was followed by evacuation at 383 K for removing free and weakly adsorbed NH 3 . Desorption profile was then measured with evacuation at the temperature gradient of 10 K min -1 , using a thermal conductivity detector.

A comprehensive study has been made to predict the adsorption structures and (31)P NMR chemical shifts of various trialkylphosphine oxides (R3PO) probe molecules, viz., trimethylphosphine oxide (TMPO), triethylphosphine oxide (TEPO), tributylphosphine oxide (TBPO), and trioctylphosphine oxide (TOPO), by density functional theory (DFT) calculations based on 8T zeolite cluster models with varied Si-H bond lengths. A linear correlation between the (31)P chemical shifts and proton affinity (PA) was observed for each of the homologous R3PO probe molecules examined. It is found that the differences in (31)P chemical shifts of the R3POH(+) adsorption complexes, when referring to the corresponding chemical shifts in their crystalline phase, may be used not only in identifying Brønsted acid sites with varied acid strengths but also in correlating the (31)P NMR data obtained from various R3PO probes. Such a chemical shift difference therefore can serve as a quantitative measure during acidity characterization of solid acid catalysts when utilizing (31)P NMR of various adsorbed R3PO, as proposed in our earlier report (Zhao; et al. J. Phys. Chem. B 2002, 106, 4462) and also illustrated herein by using a mesoporous H-MCM-41 aluminosilicate (Si/Al = 25) test adsorbent. It is indicative that, with the exception of (TMPO), variations in the alkyl chain length of the R3PO (R = C(n)H(2n+1); n > or = 2) probe molecules have only negligible effect on the (31)P chemical shifts (within experimental error of ca. 1-2 ppm) either in their crystalline bulk or in their corresponding R3POH(+) adsorption complexes. Consequently, an average offset of 8 +/- 2 ppm was observed for (31)P chemical shifts of adsorbed R3PO with n > or = 2 relative to TMPO (n = 1). Moreover, by taking the value of 86 ppm predicted for TMPO adsorbed on 8T cluster models as a threshold for superacidity (Zheng; et al. J. Phys. Chem. B 2008, 112, 4496), a similar threshold (31)P chemical shift of ca. 92-94 ppm was deduced for TEPO, TBPO, and TOPO.