| Summary: | Raman spectroscopy was used for the quantitative determination of Mo and W in Mo- and W-supported mesoporous silica (Mo/SBA-15 and W/SBA-15, respectively) and Mo-supported beta zeolite (Mo-BEA). Three Raman quantitative models were developed and optimized for the metal contents of Mo/SBA-15, W/SBA-15, and Mo/BEA. Subsequently, the models were characterized using the root mean square error of calibration (RMSEC), root mean square error of cross-validation (RMSECV), root mean square error of prediction (RMSEP), correlation coefficient, and predicted residual error sum of squares (PRESS) diagnostic function. The calibration range of the models were in the range of approximately 2–40 wt% for the SBA-15 support and 1–21 wt% for the BEA support because the BEA support presented lower Mo absorption than the SBA-15 support. The RMSEC, RMSECV, and RMSEP values were below 1.80% for all developed models. The highest and lowest correlation coefficients corresponded to the W/SBA-15 (0.9984) and Mo/BEA (0.9777) models, respectively. The change in catalyst support affected the mentioned chemometric parameters (Mo/SBA-15 vs. Mo/BEA). Subsequently, Raman spectroscopy combined with the temperature control stage was used to study the calcination of Mo/BEA, Mo/SBA-15, and W/SBA-15 using three-dimensional diagrams, in which the changes in catalyst structure were analyzed as functions of the temperature and time. Raman spectroscopy was determined to be a suitable analytical tool for the quantitative analysis of the metal contents of the catalyst and optimization of the calcination process. Therefore, Raman spectroscopy can be used during catalyst manufacturing.
|
|---|
