Communications
tions developed for our previous work on kinetic resolutions
with triphenylsilyl chloride and catalyst 1 (Table 1, entries 1–3).
These polymers were soluble at À788C in THF at the reaction
concentration, which fulfilled our requirement for a homogene-
ous silyl chloride source. Upon reaction completion, the poly-
mer product, enantioenriched silyl ether 9, was isolated by re-
moval of the reaction solvent, followed by precipitation with
methanol. The solid polymer products were then isolated by
either centrifugation or filtration. Ultimately, the polymer
length had little effect on the selectivity of the kinetic resolu-
tion, and selectivity factors ranging between 6 to 9 for the
three polymers were observed. The higher molecular weight
polymer (Mn =7580 gmolÀ1, n=25) was ultimately chosen for
further studies, owing to the increased efficiency of polymer
precipitation over the shorter polymers. Ultimately, the selec-
tivity of the polymer-supported reactions decreased relative to
that of the unsupported silyl chloride (Ph3SiCl; Table 1, entry 4).
This is presumably due to the different polar environments of
the silyl chloride in the polymer versus in solution.[19] The mi-
croenvironment of the polymer would provide a less polar en-
vironment for the silylation reaction, similar to employing non-
polar solvents, which has been shown to dramatically affect
both the conversion and the selectivity of the kinetic resolu-
tion.[11] The slower diffusion of the larger polymer silyl chloride
versus Ph3SiCl could additionally affect the conversion, such
that additional equivalents of silyl chloride would be needed if
the polymer were employed so that a conversion similar to
that obtained with the use of Ph3SiCl could be achieved (0.8
vs. 0.6 equiv., respectively).
Table 2. Substrate scope for the polymer-supported silylation-based ki-
netic resolution.[a]
Entry Recovered
alcohol
Catalyst er of
er of
Conv.[b] s[b] s[b]
Ph3SiCl[d]
recovered desilylated [%]
7
alcohol
65:35
alcohol
14:86
1
2
1
1
30
36
8
15
72:28
11:89
12 25
3[c]
2
2
63:37
63:37
9:91
7:93
25
24
11 36
16 100
4[c]
[a] Reactions were run at a concentration of 0.2m (entries 1 and 2) and
0.4m (entries 3 and 4) with respect to the alcohol for 48 h. [b] See
ref. [18]. [c] Employed 7 (1 equiv.) and iPr2NEt for 94 h. [d] Data was taken
from ref. [14] (entry 1), ref. [11] (entry 2, iPr2NCHEt2 instead of iPr2NEt), and
ref. [12] (entries 3 and 4); Ph3SiCl was employed as the silyl source. See
references for full experimental details.
Substrates that were successfully employed in previous sily-
lation-based kinetic resolutions were resolved by utilizing silyl
chloride polymer 7. Two cyclic secondary alcohols (Table 2, en-
tries 1 and 2) and two a-hydroxy lactones (Table 2, entries 3
and 4) were resolved by employing 1 and 2 as the catalysts, re-
spectively. Overall, the substrates resulted in synthetically
useful selectivities, ranging from 8 to 16. Although these selec-
tivities are lower than those obtained with triphenylsilyl chlo-
ride (Table 2, last column),[11,12] as discussed above, the results
show that efficient resolution is possible with this polymer-
bound reagent. To the best of our knowledge, this is the most
efficient polymer-supported kinetic resolution with the use of
a small-molecule catalyst.
To show the recyclability of silyl chloride polymer 7, it was
employed in a preparative-scale run with the intention of re-
covering the polymer, and it was then used in a second kinetic
resolution (Scheme 5). The reaction was run under standard re-
action conditions by using (Æ)-8 (0.6 g) and 7 (1 g). The selec-
tivity factor for the first run was 6.5, which is only a minor re-
duction relative to the selectivity of previous smaller scale
runs. The polymer product was isolated, and the derivatized al-
cohol was desilylated by reduction with lithium aluminum hy-
Additionally, the need for chromatography was eliminated
by use of the silyl chloride polymer. After performing an acid
wash to remove the catalyst and amine base, the enantiomers
were effectively separated by dissolving the mixture in metha-
nol and filtering off the insoluble polymer-bound silyl ether
product. The recovered, unreacted alcohol in the filtrate was
generally pure enough to use in further syntheses, if desired.
The other enantiomer was cleanly obtained after the silyl ether
polymer was treated with tetrabutylammonium fluoride and
removal of the silane polymer by precipitation. Even though
multiple cycles of precipitation were needed to effectively re-
cover the relatively short polymer (n=25), on large scale this
could be more advantageous than methodology employing
Ph3SiCl, which requires two columns to isolate both enantio-
mers.
Scheme 5. Recycling of silyl chloride polymer in a subsequent kinetic resolu-
tion. [a] See ref. [18]. [b] Kinetic resolutions were run at a concentration of
0.15m with respect to the alcohol.
ChemCatChem 2015, 7, 1527 – 1530
1529
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim