July 1998
SYNLETT
757
Benzyl Trityl Ether and DDQ as New Tritylating Reagents
Masato Oikawa, Hiroaki Yoshizaki, and Shoichi Kusumoto*
Department of Chemistry, Graduate School of Science, Osaka University, Toyonaka, Osaka 560-0043, Japan
Fax 81-6-850-5419; skus@chem.sci.osaka-u.ac.jp
Received 31 March 1998
Abstract: We describe herein a new tritylation procedure of alcohols
using benzyl trityl ether and 2,3-dichloro-5,6-dicyano-1,4-
benzoquinone. The reaction involves oxidative abstraction of one of the
benzylic protons of benzyl trityl ether, followed by transformation of the
generated benzyl trityl ether cation into a complex of benzaldehyde and
trityl cation. The present procedure proceeds under mild neutral
conditions to afford trityl ethers in generally good yields for primary
alcohols, and in acceptable yields for several secondary alcohols.
Tritylation of symmetrical diols was examined next. When ethylene
glycol was treated with 1.2 equiv of BTE and 1.5 equiv of DDQ, a
bistritylated compound was produced in 47% yield (entry 6). No trace of
the monotritylated compound was observed even during the time of
reaction as an intermediate as far as monitored by thin-layer
chromatography. The same result was obtained for tritylation of 1,4-
butanediol wherein the bistrityl ether was the sole product (46%, entry
7). In both entries, the monotritylated compound has higher reactivity
than the starting diol. Yields in entries 6 and 7 were calculated to be
79% and 76%, respectively, based on BTE. In contrast, tritylation of
(1R,2R)-(+)-hydrobenzoin gave a monotrityl ether in 81% yield (entry
8). The bistritylated product was not produced at all in this case,
probably owing to the severe steric hindrance in the monotritylated
product.
The triphenylmethyl (trityl = Tr) group is a useful protecting group for
1
alcohols. Distinctive features of the Tr group are: (i) owing to its
bulkiness, selective protection of polyols as a Tr ether is easily achieved,
and (ii) removal (deprotection) of the Tr group is also easily effected by
2
®
3
the action of mild acids such as formic acid or Amberlyst resin. For
the preparation of Tr ethers, a combination of trityl chloride (TrCl) and
Selective transformation of an unsymmetrical diol, 1-phenylethane-1,2-
diol, into a monotrityl ether can be achieved using 1.1 equiv of BTE
(entry 9). In this reaction, the primary hydroxy group was selectively
derivatized in 73% yield. It is noteworthy that this monotrityl ether was
obtained in rather low yield (64%) under the conventional conditions
(1.1 equiv of TrCl, 3.0 equiv of triethylamine, 0.2 equiv of DMAP,
DCM, 30 °C, 3 h). Increasing the amount of BTE and DDQ to 2.0 equiv
each for the purpose of improving the yield resulted in the production of
the undesired bistrityl ether (25%, entry 10).
4,5
an amine base seems to be the most popular method. Other reagents
for tritylation of alcohols involve a complex of trityl chloride and 4-
6
dimethylaminopyridine (DMAP) prepared beforehand. Tritylation
under Lewis acidic conditions is achieved by employing trityl
7
trimethylsilyl ether and trimethylsilyl trifluoromethanesulfonate. In
this paper we wish to report a new entry for tritylation of alcohols by the
use of benzyl trityl ether (BTE) and 2,3-dichloro-5,6-dicyano-1,4-
benzoquinone (DDQ) (eq 1).
Tritylation of sugar derivatives was then examined. o-Nitrophenyl-β-D-
galactopyranoside containing four hydroxy groups was selectively
converted to the 6-O-Tr product in 55% yield by the use of 1.2 equiv of
BTE (entry 11). In this reaction none of the regioisomers were obtained.
In an attempt to improve the yield, the reaction was next carried out at
50 °C in DCE (entry 12), but the yield did not increase at all. It was also
found that the use of 5.0 equiv of BTE and 6.3 equiv of DDQ gave a
complex mixture of products, and the yield of the 6-O-Tr product was
10% (data not shown). The sterically hindered glucosamine derivative
was also transformed into the 3-O-Tr ether by this method in rather low
The experiment was carried out using more than 1.1 equiv of BTE to the
substrate alcohol, a larger excess amount (more than 1.5 equiv) of DDQ,
and molecular sieves 4A in anhydrous dichloromethane (DCM) or 1,2-
yield (38%, entry 13). Because
a
substantial amount of
8
dichloroethane (DCE) at 30-50 °C. Molecular sieves were used to
triphenylmethanol was produced as a byproduct in all entries in Table 1,
it is obvious that hydrolysis of the Tr cation is a serious side reaction.
The use of molecular sieves was important to suppress this side reaction.
But in the case of carbohydrate derivatives as substrates whose hydroxy
groups are not reactive enough, addition of molecular sieves was less
effective. The trace amount of water whose complete removal was
impossible with molecular sieves may seriously compete with the
substrates. Other side reaction such as the formation of the Tr ether of
generated 2,3-dichloro-5,6-dicyanohydroquinone was not observed at
all in any entries.
remove moisture from the reaction mixture. Results on several types of
alcohols are summarized in Table 1.
Tritylation of (S)-(+)-methyl 3-hydroxy-2-methylpropionate proceeded
at a slow rate using 1.2 equiv of BTE and 1.5 equiv of DDQ at 30 °C
(entries 1 and 2). If the reaction was quenched after 5 h, the product was
obtained in only 54% yield after purification by silica-gel flash
chromatography. The yield reached, however, 99% after 36 h. The
enantiomeric excess of the product was 99.0% indicating that
racemization did not take place during the reaction. Tritylation of the
secondary hydroxy group of ethyl 3-hydroxybutyrate, resulted in lower
yield (81%, entry 3) after 22 h using the same equivalents of the
reagents as those in entries 1 and 2. In this reaction, a longer reaction
time did not improve the yield. The reactivity of ethyl (S)-(-)-lactate
toward tritylation was much lower than that of ethyl 3-hydroxybutyrate
so that the Tr ether was produced in only 56% yield after 24 h (entry 4).
For such a low-reactive substrate, it was found effective to use larger
equivalents of the reagents. Thus, by using 2.0 equiv of BTE and 2.5
equiv of DDQ, the tritylation yield of ethyl lactate reached 78% (entry
5). As shown in entries 1-5, tritylation of structurally simple alcohols
gave the products in moderate to excellent yields.
A plausible reaction mechanism of the present tritylation reaction is
illustrated in Scheme 1. Oxidation of electron-rich BTE by DDQ should
11
occur as an abstraction of one of the benzylic protons. The cation I
thus generated was then transformed into a complex III of benzaldehyde
and Tr cation because the production of benzaldehyde was observed in
all entries in Table 1. To support this assumption, the process from the
BTE cation I to the complex III was calculated by the semiempirical
12
13
level of theory employing the AM1 method (Figure 1). By the
coordinate driving analysis, it was found that, as the distance between
14
the Tr group and oxygen (C -O) changed from 1.4 Å to 3.5 Å, the
8
energy (heat of formation) went down via a rather unstable state II (the
Oikawa, Masato
