4698
A. Yamada, K. Nakata / Tetrahedron Letters 57 (2016) 4697–4701
catalyst loading from 5 to 2 mol % (entry 18). In an attempt to
increase the reaction efficiency, the reaction time was prolonged
to 48 h; however, the conversion was not significantly improved
(entry 19).
With the optimized reaction conditions, three other types of
isothiourea catalysts, (R)-BTM (12)28 and (R)-HBTM (13),28b,29
reported by Birman et al., and HBTM-2.1 (14),30 reported by Smith
et al., that served as efficient acyl transfer catalysts, were applied to
the KR of ( )-9f (Table 3). The reaction of ( )-9f was carried out
with 0.75 equiv of anhydride 8b and a catalytic amount of (R)-
BTM (12) to obtain a moderate s-value (entry 2). When the same
reactions were conducted with (R)-HBTM (13) and HBTM-2.1
(14), both the reactions proceeded far beyond 50% conversion,
affording the corresponding chiral alcohols 9f in >99% ee (entries
3 and 5). Because the exact ee of 9f could not be determined by
HPLC analyses, the s-values were not evaluated under the condi-
tions. Thus, the reactions were reinvestigated by reducing the
amount of anhydride 8b to 0.5 equiv. A moderate s-value was
obtained using (R)-HBTM (13) (entry 4), and almost the same
s-value was obtained in the case of (R)-NMBG (3) using
HBTM-2.1 (14) (entry 6 vs entry 1).
To assess the generality of this novel method and systematically
explore the electronic and steric effects on the aromatic rings of the
substrates, a series of racemic b-hydroxy esters ( )-15a–k with
carboxylic anhydride 8b was investigated under the optimized
reaction conditions (Table 4). The KR of 15a–c with methyl sub-
stituents on the aromatic rings of the substrates smoothly pro-
ceeded in almost 50% conversion with good s-values in all the
cases, irrespective of their positions (entries 1–3). Similarly, the
KR of methoxy-substituted substrates 15d–f was investigated;
the reactions were influenced by the substitution patterns. The
reaction of 15d with an ortho-methoxy group resulted in poor reac-
tivity and a low s-value (entry 4). On the other hand, the reactions
of 15e and 15f with meta- and para-methoxy groups smoothly pro-
ceeded with good s-values, respectively (entries 5 and 6). The same
tendency was observed in the case of chloro-substituted substrates
15g–i, and better results were obtained compared to the corre-
sponding cases of methoxy-substituted substrates (entries 7–11).
Scheme 1. Previous study (1) and this study (2).
propionic anhydride (8c) and isobutyric anhydride (8d), the latter
branched substituent showed better results than the former nor-
mal substituent (entries 3 and 4). In contrast, aromatic carboxylic
anhydride 8e with a very low s-value was not effective for the reac-
tion (entry 5).
Next, to evaluate the effect of the structure of ester moiety on
the substrates, the KR of diverse racemic b-hydroxy esters ( )-
9a–h was investigated using both DPHAA (8a), and cyclohexane
carboxylic anhydride (8b) under the abovementioned conditions
(Table 2). First, the reaction was carried out using DPHAA (8a)
(entries 1–8). The KR of 9a and 9b bearing normal substituents
(R2 = Et and Bn) on the ester moiety was carried out to afford good
s-values (entries 1 and 2). On the other hand, the reaction of 9c–e
bearing disubstituted substituents (R2 = iPr, CHPh2, and cC6H11
)
showed better results (entries 3–5). The reaction of 9f with a
trisubstituted substituent (R2 = tBu) smoothly proceeded with a
high s-value regardless of the steric effect (entry 6). When the
methyl group(s) was replaced with a phenyl group(s) at the R2 sub-
stituent of 9f, the reactivity was enhanced; however, the selectivity
decreased (entries 7 and 8). When anhydride 8b was used, the
same tendency was observed, and the selectivity slightly improved
in the corresponding case of 8a (entries 9–17). Among all the
entries, the highest s-value was obtained in the reaction of 9f with
8b (entry 15). Under the conditions in entry 15, it was found that
both the reactivity and selectivity decreased upon decreasing the
The reactions of
a- and b-naphthyl-substituted substrates 15j
and 15k were also investigated; they afforded a high selectivity
(entries 12–15).
In summary, we achieved the nonenzymatic acylative KR of
b-hydroxy esters with cyclohexane carboxylic anhydride in the
Table 1
(R)-NMBG catalyzed KR of ( )-6 using various carboxylic anhydrides 8a–e
Entry
R1
Yield (7; 6)a (%)
ee (7; 6) (%)
s
1
2
3
4
5
Ph2CH (a)
cC6H11 (b)
Et (c)
43; 46
41; 50
32; 57
44; 33
19; 69
83; 66
83; 80
74; 44
72; 79
14; 2
22
26
10
14
1
iPr (d)
Ph (e)
a
Isolated yield.