6858
S. Dong et al. / Tetrahedron Letters 56 (2015) 6857–6859
Scheme 1. Synthesis of HSCH2OSiMe2Bu-t.
NaSH.xH2O, but that gave the desired product heavily contami-
nated (>50%) with other substances. Treatment of 11 with AcSNa
worked well, but we were unable to effect clean hydrolysis to 12.
Oxidation of crude 12 to the disulfide, followed by hydride
reduction was also unsatisfactory. Eventually, we generated HSLi
in THF under anhydrous conditions8 and found that it reacts
cleanly with 11 to afford the desired reagent 12. Material made
in this way (ca 92% yield) contains only minor impurities (1H
NMR, 13C NMR) and is perfectly suitable for direct use; it can be
stored in a closed vessel in a freezer for several (at least 4) weeks.
As expected, the protected thiol can be deprotonated and then
alkylated with bromides, and our results are shown in Table 1.
For compounds 13 to 16 the experimental procedure involved add-
ing the thiol 12 to a mixture of NaH (1 equiv) and the starting bro-
mide in DMF at 0 °C. In the case of 18 and 19,9 the NaH was added
to a solution of the bromide and the thiol in order to avoid prema-
ture deprotonation of the carbonyl compound. This reverse mode
of addition was arbitrarily also used for 17. In a preliminary exper-
iment to prepare 13, n-BuLi was added to a THF solution of the
thiol at ꢀ78 °C, followed by BnBr, but the NaH method was just
as effective.
Scheme 2. Synthesis and alkylation of HSCH2OSiPh2Bu-t.
EtSCH2OSiPh2Bu-t, itself formed very efficiently. However, when
we treated the chloride with HSLi the results were erratic and in
only one out of six attempts did we obtain the desired HSCH2-
OSiPh2Bu-t (32%). We were unable to identify the cause(s) of this
variability, but we found that reaction of the chloride with AcSK
gives the expected thioacetate (Scheme 2), and this reacts with
hydrazine hydrate to liberate HSCH2OSiPh2Bu-t in 57% yield. As
expected, thiol 23 can be alkylated, as shown by the single example
we studied (23 ? 24).
As stated above, our need for a nucleophilic protected thiol
arose during studies on a natural product synthesis, and we were
specifically interested in the generation of
a-(alkylthio) glycine
esters 28. Several of these have been made (Eq. 3) by reaction of
an amine with EtO2CCHO, followed by addition of a thiol (p-
MeOC6H4CH2SH was used in the reported experiments).15 Our
requirements were for a protecting group on sulfur that could be
removed by treatment with fluoride ion or with a sulfenyl chloride;
hence the design of the thiol 12. When we used this thiol in the
manner described by Eq. 3 we obtained 28 (R0 = SCH2OSiMe2Bu-t)
in 86% yield.
As shown by the Table, primary and activated bromides give
yields between 60% and 88%. Cyclopentyl bromide also reacted,
but experiments with cyclohexyl bromide were unsuccessful.
Under our experimental conditions we did not notice the
formation of vinyl sulfides, which are formed readily from RCH
(OSiMe3)SH (R – H), NaH and alkyl halides if the RCH(OSiMe3)
SH/NaH mixture is stirred for 30 min at 0 °C and 10 min at room
temperature.13
ð3Þ
Although our main interest is in the t-butyldimethylsilyl group
for O-protection, we also sought to prepare HSCH2OSiPh2Bu-t in an
exactly analogous manner, using ClCH2OSiPh2Bu-t. This is also a
known14 compound available in high yield (99%) from
Conclusion
The CH2OSiR3 protecting group for bivalent sulfur has proven
useful in model studies1 related to MPC1001; with appropriate
substrates the unit SCH2OSiR3 can be introduced either by use of
the electrophilic reagent 2 or the complementary nucleophilic
reagent 12.
Table 1
S-Protected thiols
Acknowledgments
We thank NSERC for financial support, the China Scholarship
Council for an Award to D. S., and Professor W. B. Motherwell
and Dr. S. T. Hilton for experimental details of their route to
(alkylthio) glycine esters.
a-
Supplementary data
Supplementary data (experimental procedures and copies of 1H
and 13C NMR spectra for new compounds) associated with this
References and notes