A 2(8-4) reduced factorial design was used to study the influence of differ
ent experimental variables in the synthesis of mesoporous materials. Eight
variables were investigated at high and low levels. This resulted in 18 exp
eriments, including two center points. X-ray diffraction patterns were used
to characterize the samples, which ranged from well-defined MCM-41 structu
res to amorphous materials. Analysis of the data resulted in two models (Mo
del 1 and Model 2). In Model 1, the materials were simply divided into poro
us or non-porous materials. Several variables and interaction terms were fo
und to be significant for this model, of which pH turned out to be the most
important variable. The quality of the materials was taken into considerat
ion in Model 2. The pH was again found to be the most important variable fo
r the model. Model 2 also indicated that at least one of the variables is n
on-linear. (C) 2000 Elsevier Science B.V. All rights reserved.