Molecules 2001, 6
1007
Introduction
Usually, the removal of methoxymethylene protective groups is carried out in liquid phase using
mainly sulphuric or hydrochloric acids as catalysts [1-3]. However, the problems associated with the
handling and disposal of the inorganic acids, and the environmental and potential hazards of the same
have raised interest in the development of an alternative process using solid acid catalysts. The use of
heteropolyoxoanions (HPAs) as catalysts is an interesting area that has widely been studied recently. The
6
+
6+
HPAs are composed of a close-packed framework of metal-oxygen octahedra, MO (M=Mo , W )
x
4
+
5+
surrounding a central atom, X (Si , P ). In particular, the Wells-Dawson acid structure (H P W O .aq.)
6
2
18 62
consists of two identical “half units” (P W ) linked through the oxygen atoms.
2
9
In this work, we present a deprotection study carried out using such a heteropolyacid (HPA) with
Wells-Dawson structure in organic solvents. The Wells-Dawson acid (WD) was tested using both bulk
solid samples and solid-supported ones using on large pore silica as the support.
Results and Discussion
The deprotection reaction was studied using compounds 1-9 as the substrates. Different reaction
conditions were checked, e.g. temperature, time, reaction solvent, concentration of the solution and
aggregation state of the catalyst. Concentration of the catalyst on the support and the molar ratio of WD to
substrate were also varied.
Reactions using bulk catalyst
When 1,2-dichloroethane was the solvent, the bulk catalyst did not dissolve, even at 80ºC. The
reaction is slow under these conditions and a maximum yield of 80% is reached in 5 hours. Secondary
products begin to appear at longer reaction times. On the other hand, the use of methanol as the solvent
allows the catalyst to dissolve, the reaction is fast and the yields are excellent, e.g. 98% within 45 min or
100% after 1 hr. reaction at 65ºC (Table 1).
Methanol is therefore a good solvent for the reaction, but it has the disadvantage that it dissolves the
WD off the silica when the supported catalyst was used. Consequently, we examined the use of mixtures
of solvents in which the catalyst is insoluble and small amounts (1 – 5%) of methanol. Yields are very
good using a 1,2-dichloroethane-methanol mixture. Among other potential solvents, THF was also
studied. Yields are low and reactions are slow when pure THF is used; however, addition of 1% methanol
rises the yield, being quantitative in 40 min. In all the experiments 1% catalyst (moles/ mole of substrate)
was used.