Angewandte
Chemie
Table 1: Optimization of the reaction conditions.[a]
was obtained under the standard reaction conditions
(entry 15).
With the optimal reaction conditions determined, various
4-alkyl-4-aryl-3-butenoic acids 2 were hydrogenated. The
substituent on the phenyl ring of the substrate had a negligible
influence on the enantioselectivity, and all of the tested
substrates afforded essentially the same enantioselectivity
(92–96% ee; Table 2, entries 2–10). The reactivity of (E)-4-
(naphthalen-2-yl)pent-3-enoic acid (2k) was similar to that of
2a, and an enantioselectivity of 97% ee was obtained
Entry
Catalyst
Additive
pH [atm]
Conv. [%][b]
ee [%][c]
2
1
2
3
4
5
6
7
8
(Sa,S)-3a
(Ra,S)-3a
(Sa,S)-3b
(Sa)-3c
NEt3
NEt3
NEt3
NEt3
NEt3
NEt3
NEt3
NEt3
NEt3
NEt3
NEt3
NEt3
iPr2NEt
Et2NH
Cs2CO3
6
6
6
6
6
6
6
6
30
3
100
45
80
85
100
94
100
100
100
100
100
98
100
100
21
60
rac.
20
15
92
85
95
94
90
96
95
95
95
95
–
(Sa,S)-3d
(Sa,S)-3e
(Sa,S)-3 f
(Sa,S)-3g
(Sa,S)-3 f
(Sa,S)-3 f
(Sa,S)-3 f
(Sa,S)-3 f
(Sa,S)-3 f
(Sa,S)-3 f
(Sa,S)-3 f
Table 2: Asymmetric hydrogenation of 4-alkyl-4-aryl-3-butenoic acids.[a]
9
10
11[d]
12[d,e]
13[d]
14[d]
15[d]
3
3
3
3
Entry
Ar: R (2)
Product
Yield [%]
ee [%]
1
C6H5: Me (2a)
1a
1b
1c
1d
1e
1 f
1g
1h
1i
1j
1k
1l
1m
1n
97
97 (98)
96
96
98
96
97
98
97
98
98
96
96
95
94 (93)
94
96
95
92
95
95
95
95
97
90
92
2[b]
3
4-MeC6H4: Me (2b)
4-MeOC6H4: Me (2c)
4-FC6H4: Me (2d)
3
[a] Reaction conditions: 0.5 mmol scale, [substrate]=0.25 molLÀ1
1.0 equiv additive, 12 h. [b] Determined by NMR spectroscopy.
[c] Determined by chiral-phase HPLC or supercritical fluid chromatog-
raphy (SFC) analysis of the corresponding anilide. [d] Using 0.5 mol%
catalyst. [e] Using 0.5 equiv NEt3.
,
4
5
6
7
8
9
3-BrC6H4: Me (2e)
3-MeC6H4: Me (2 f)
3-MeOC6H4: Me (2g)
3,4-Me2C6H3: Me (2h)
3-MeO-4-MeC6H3: Me (2i)
3,4-(OCH2O)C6H3: Me (2j)
2-naphthyl: Me (2k)
2-thiophenyl: Me (2l)
4-MeOC6H4: Et (2m)
4-MeOC6H4: iPr (2n)
10[c]
11
12[c]
13
14[c]
and racemic product (entry 2). This result demonstrated that
the chiralities in (Ra,S)-3a were unmatched in their ability to
induce enantioselectivity in the hydrogenation reaction. The
effect of the oxazoline ring substituent of the ligand on the
enantioselectivity of the reaction was studied. Changing the
substituent from benzyl to isopropyl [(Sa,S)-3b] or H [(Sa)-3c]
decreased both conversion and enantioselectivity (entries 3
and 4). The substituents on the phosphorous of the ligand
strongly affected the enantioselectivity (entries 1, 5, and 6):
catalyst (Sa,S)-3d, with 3,5-dimethylphenyl substituents on
the phosphorous, exhibited the best chiral induction (92% ee,
entry 5). To further improve the enantioselectivity, we
synthesized two new ligands with an a-naphthylmethyl or
a b-naphthylmethyl group on the oxazoline ring, and the
corresponding catalysts (Sa,S)-3 f and (Sa,S)-3g were prepared
by means of a previously reported method.[6] Both (Sa,S)-3 f
and (Sa,S)-3g showed higher enantioselectivities in the
hydrogenation reaction than did (Sa,S)-3d (entries 7 and 8).
The reaction conditions were optimized with catalyst
(Sa,S)-3 f. Experiments on hydrogen pressure showed that the
reaction gave higher enantioselectivity under lower hydrogen
pressure (Table 1, entry 10 vs. entries 7 and 9), which is similar
to our previous observations in the hydrogenation of a,b-
unsaturated carboxylic acids.[3j] The catalyst loading can be
reduced to 0.5 mol% without diminishing the reactivity and
enantioselectivity (entry 11). Reducing the basic additive
NEt3 to 0.5 equiv slightly decreased the conversion under
standard reaction conditions (entry 12). Aside from NEt3,
other organic bases, such as iPr2NEt and NHEt2, were also
suitable additives for the hydrogenation reaction, affording
identical results to NEt3 (entries 13 and 14). On the other
hand, the use of the inorganic base Cs2CO3 drastically
decreased the reactivity of the catalyst; only 21% conversion
97
88
[a] The reaction conditions and analysis were the same as those in
Table 1, entry 11. Full conversions were obtained in all cases. [b] The data
in parentheses were obtained at 1.52 g scale. [c] Using 1 mol% catalyst.
(entry 11). The Ar group of 2 can also be a heterocycle. For
example, the hydrogenation of (E)-4-(thiophen-2-yl)pent-3-
enoic acid (2l) ran smoothly and afforded the desired product
with 90% ee in the presence of 1 mol% catalyst (entry 12).
Changing the R group of the acid 2 from methyl to ethyl (2m)
slightly reduced the enantioselectivity (entry 13). However,
when R was a bulky isopropyl (2n), 1 mol% of the catalyst
was required for complete conversion, and the ee decreased to
88% (entry 14). This asymmetric hydrogenation reaction
could easily be carried out on a gram scale, which is a benefit
for practical applications. For instance, (E)-4-p-tolylpent-3-
enoic acid (2b) was hydrogenated smoothly on a 1.52 g scale
under the standard hydrogenation conditions to produce the
desired product without compromise of either yield or
enantioselectivity (entry 2).
To further demonstrate the utility of the hydrogenation
reaction, three natural sesquiterpenes, (R)-aristelegone-A,
(R)-curcumene, and (R)-xanthorrhizol were conveniently
synthesized from the products of the asymmetric hydro-
genation (Scheme 4). With the hydrogenation product 1i as
starting material, (R)-aristelegone-A,[7] was obtained through
a Brønsted acid catalyzed Friedel–Crafts acylation and
a subsequent demethylation with Et2NCH2CH2SNa in 68%
overall yield (steps a and b). Note that the present procedure
Angew. Chem. Int. Ed. 2012, 51, 2708 –2711
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim