Enantioselective synthesis of anti- and syn-β-hydroxy-α-phenyl carboxylates via boron-mediated asymmetric aldol reaction

Article abstract of DOI:10.1039/c3cc40860d

A reagent-controlled, diastereo- and enantioselective synthesis of anti- and syn-β-hydroxy-α-phenyl carboxylates has been achieved by the proper choice of solvent, temperature, alkoxy group, and amine for the diisopinocampheylboron-mediated asymmetric enolization-aldolization of phenylacetates. The pure diastereomers can be readily separated by column chromatography.

Full text of DOI:10.1039/c3cc40860d

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Enantioselective synthesis of anti- and syn-b-hydroxy-
a-phenyl carboxylates via boron-mediated asymmetric
aldol reaction†
Cite this: Chem. Commun., 2013,
49, 3152
Received 31st January 2013,
Accepted 25th February 2013
P. Veeraraghavan Ramachandran* and Prem B. Chanda
DOI: 10.1039/c3cc40860d
www.rsc.org/chemcomm
A reagent-controlled, diastereo- and enantioselective synthesis of
anti- and syn-b-hydroxy-a-phenyl carboxylates has been achieved
by the proper choice of solvent, temperature, alkoxy group,
and amine for the diisopinocampheylboron-mediated asymmetric
enolization–aldolization of phenylacetates. The pure diastereomers
can be readily separated by column chromatography.
Scheme 1 Diastereoselective aldol reaction of 20-hydroxy-1,10-binaphthyl ester
The utility of diastereo- and enantioselective aldol reactions is
almost unparalleled in the art of organic synthesis of simple and
complex organic molecules.1 A comparison of the several promoters
enolates with aldehydes.
available for this transformation discloses the advantages offered by designed to examine the enolate formation–aldolization of methyl-
borane-derived reagents.2 Among the carbonyl partners, ketones, (2), ethyl- (3), isopropyl- (4), and tert-butyl- (5) phenylacetates using
thioesters, and imides have been well explored.1,2 Nevertheless, the (ꢀ)-diisopinocampheylboron triflate [(ꢀ)-Ipc2BOTf, (ꢀ)-1]. The
enolboration–aldolization of carboxylate esters, a ubiquitous class of literature suggested that the enolization–aldolization of tert-butyl
carbonyls, has remained relatively unexplored.3,4 The potential phenylacetate (5), bearing a bulky alkoxy group, should provide
synthetic utility of aldol adducts derived from arylacetates, especially high anti-selectivity for the aldol reaction.4f However, we were
when used for the synthesis of aryl analogs of bio-active molecules,5 apprehensive due to our past difficulty in enolizing this ester with
prompted our recent study that revealed an extraordinary effect dicyclohexylboron triflate.6 Indeed, an attempted reaction of 5 with
of solvent and temperature on the diastereoselectivity of racemic (ꢀ)-1 in the presence of Et3N at ꢀ30 1C, followed by aldolization of
b-hydroxy-a-phenyl esters.6 To the best of our knowledge, there is benzaldehyde (6a) at the same temperature showed insignificant
only one report on the preparation of chiral b-hydroxy-a-phenyl product formation. On the basis of the preparation of racemic anti-
esters, which entails a substrate-controlled aldol reaction of lithium aldols from methyl phenylacetate by the appropriate choice of the
enolates of 20-hydroxy-1,10-binaphthyl phenylacetate (Scheme 1).7 solvent and temperature, the effect of the reaction parameters on
This reaction provided 54 : 46 to 98 : 2 selectivity for anti-aldols in the diastereomeric ratio (dr) and enantiomeric ratio (er) using
27–82% de, and lacks a protocol for the complementary syn-aldols. esters bearing less bulky than the t-butoxy group was examined.
The expensive chiral auxiliary and moderate selectivity described Thus, phenylacetates 2–4 were enolized in ‘‘anti-favoring solvent’’,6
therein, as well as the demonstrated success of diisopinocampheyl- carbon tetrachloride, with (ꢀ)-1 in the presence of Et3N at ꢀ30 1C,9
boron-mediated aldol reaction4f,8 encouraged the examination of a followed by aldolization with benzaldehyde (6a) at the same
reagent-controlled asymmetric aldol reaction of phenylacetates. temperature. Hydrogen peroxide oxidation, in the presence
The details of the study and the successful preparation of anti- of pH 7 buffer, provided a crude mixture of diastereomers of the
and syn-aldols in high enantiomeric purities are described herein. b-hydroxy-a,b-diphenyl esters (7a–9a). Analysis of the carbinolic
To optimize the conditions for the selective formation of proton using 1H NMR spectroscopy disclosed the diastereo-
either syn or anti chiral b-hydroxy-a-phenyl esters, a project was selectivity of the reaction. The pure diastereomers were readily
separated by silica gel column chromatography and their
combined yields are presented in Table 1.10 The er was determined
Hebert C. Brown Centre for Borane Research, Purdue University, 560 Oval Drive,
by 1H NMR analysis of the menthyl carbonate derivative of the diols
West Lafayette, IN 47907-2084, USA. E-mail: chandran@purdue.edu
derived from the major diastereomer.11
Esters 2 and 3 provided 85% and 83% anti-isomer of 3-hydroxy-
2,3-diphenylpropanoates (7a, 8a, respectively) in 80% and 81%
† Electronic supplementary information (ESI) available: Synthesis and character-
ization of compounds, 1H and 13C NMR spectra and detailed experimental
procedure. See DOI: 10.1039/c3cc40860d
c
3152 Chem. Commun., 2013, 49, 3152--3154
This journal is The Royal Society of Chemistry 2013

View Article Online
Communication
ChemComm
Table 1 Optimization of anti- and syn-selective asymmetric aldol reaction of
phenylacetates with (ꢀ)-1
Cond.a
Aldol
Scheme 2 Optimum conditions for anti-selective asymmetric aldol reaction of 4
with (ꢀ)-1.
Entry Ester Enol–aldol. Solvent
#
Yieldb (%) syn : antic erd
1
2
3
4
5
6
7
8
2
3
4
5
4
4
4
4
4
4
4
2
3
4
3
3
3
A
A
A
A
B
B
B
B
B
B
B
C
C
C
D
—
C
CCl4
CCl4
CCl4
CCl4
Pentane 9a 83
Et2O 9a 88
Toluene 9a 89
CH2Cl2 9a 83
Pentane 9a 79
Pentane 9a 86
Pentane 9a 84
CH2Cl2 7a 75
CH2Cl2 8a 75
CH2Cl2 9a 75
CH2Cl2 8a 67
CH2Cl2 8a 77
CH2Cl2 8a 60
7a 80
8a 81
9a 78
10a —e
15 : 85
17 : 83
8 : 92
78 :22
79 : 21
79 : 21
—
steric and electronic requirements (hindered and unhindered,
aliphatic, and arylic with electron-donating and electron-
withdrawing groups) was treated with the enolate derived from 4
to furnish the corresponding anti-b-hydroxy-a-phenyl esters in mod-
erate to high yields, very good dr, and moderate to high er. As
summarized in Table 2, electron-donating (6b and 6c) and with-
drawing (6d and 6e) groups on the phenyl ring did not affect the
yields or er of the pure anti-isomers. A representative heteroaromatic
aldehyde, thiophene-2-carbaldehyde (6f), provided the highest dia-
stereoselectivity (98 : 2) of all the aldehydes tested. Both unbranched
aliphatic, n-propyl (6g), and branched aliphatic aldehydes, isobutyr-
aldehyde (6h) and pivalaldehyde (6i) also gave similar results. The
diastereoselectivity was extremely high for the hindered aldehyde 6i.
Attention was then turned to standardizing the conditions to
prepare the syn-aldols with high er. Fortuitously, the initial screen-
ing of solvents for the enolization–aldolization of 4 with (ꢀ)-1 had
revealed 68% syn-selectivity in CH2Cl2, at ꢀ78 1C (Table 1, entry 8).
On the basis of the improved syn-selectivity at higher temperature,
presumably due to the thermodynamic Z-enolate,6 the enolization
temperature was increased to 0 1C, while maintaining the aldoliza-
tion temperature at ꢀ78 1C,13 whereat the diastereoselectivity of
syn-9a showed minor improvement (72% syn, Table 1, entry 14).
On the basis of the effect of the sterics on syn-selectivity, we
examined esters 2 and 3 under the same conditions.
—
8 : 92
8 : 92
7 : 93
87 : 13
80 : 20
78 : 22
f
68 : 32
21 : 79
12 : 88
8 : 92
78 : 22
83 : 17
72 : 28
81 : 19
82 : 18
82 : 18
—
—
9g
10h
11i
12
13
14
15
16
17g
f
79 : 21
86 : 14
f
—
>99 : o1
f
—
—
f
j
>99 : o1
f
—
a
A = enolization: (ꢀ)-1/Et3N, ꢀ30 1C; aldolization: 6a, ꢀ30 1C. B =
enolization: (ꢀ)-1/Et3N, ꢀ78 1C; aldolization: 6a, ꢀ78 1C. C = enoliza-
tion: (ꢀ)-1/Et3N, 0 1C; aldolization: 6a, ꢀ78 1C. D = enolization: (ꢀ)-1/
b
Et3N, 25 1C; aldolization: 6a, ꢀ78 1C. Combined yields of pure syn and
c
anti isomers. Syn and anti ratios were determined using 1H NMR
spectroscopy of the crude product. er was determined by the 1H NMR
d
analysis of menthyl carbonate derivatives of diols derived from aldol
e
f
products. Insignificant
product
formation. Not
determined.
g
h
i-Pr2NEt was used as amine. Concentration of reaction was decreased
to half. Concentration of reaction was doubled. The antipode of the
i
j
reagent, (+)-1 was used under condition C.
combined yields, respectively, with similar er (78 : 22 and 79 : 21
respectively; Table 1, entries 1 and 2). The anti-selectivity improved
to 92% with 4 (Table 1, entry 3), although the er of isopropyl
3-hydroxy-2,3-diphenylpropanoates (9a) remained the same.
Despite the high anti-selectivity with 4, the unsatisfactory er
prompted a full examination of the effects of solvent, amines,
and the mode of addition12 at low temperature. High anti-
selectivity was expected on the basis of our earlier observation
that lower temperatures favor the formation of anti-isomers
during the aldol reaction of 2, irrespective of the solvent.6
Indeed, the reaction of 4 conducted in pentane, ethers, toluene,
and CH2Cl2 at ꢀ78 1C (Table 1, entries 5–8) revealed similar
diastereoselectivity (92–93% anti-) in all of the solvents except
in CH2Cl2. Unexpectedly, the reaction provided only a moderate
68 : 32 diastereoselectivity favoring the syn-isomer in CH2Cl2
(Table 1, entry 8). Pentane was chosen as the optimal solvent
for anti-aldols due to the higher er (87 : 13) (Table 1, entry 5).
The effect of the amine was determined by replacing with
i-Pr2NEt, whereafter the anti-selectivity dropped from 92% to
79% (Table 1, entry 9). Altering the concentration of the
reaction from 0.2 M to either 0.1 M or 0.4 M did not drastically
change the anti-selectivity or er of the anti-isomer (Table 1,
entries 10 and 11).
The ethyl ester (3) gave the best results: 83% syn-selectivity with
an excellent enantiomeric ratio of >99 : o1 (Table 1, entry 13).
While increasing the enolization temperature to 25 1C did not
improve the syn-selectivity, the yield decreased to 67% (Table 1,
entry 15). Replacing Et3N with i-Pr2NEt further decreased the yield
to 60% without any improvement in the dr (Table 1, entry 17).
Table 2 Examination of aldehydes for anti-selection
RCHO
Aldol
#
Entry
6
R
Yielda (%) drb
erc (anti)
1
2
3
6a C6H5
6b 4-MeC6H4
6c 4-MeOC6H4 2S,3R-9c 76
2S,3R-9a 83
2S,3R-9b 90
92 : 8
89 : 11 81 : 19
94 : 6 81 : 19
87 : 13
4
5
6d 4-FC6H4
6e 4-CF3C6H4
2S,3R-9d 80
2S,3R-9e 70
90 : 10 80 : 20
87 : 13 80 : 20
6
6f
2-Thioph
2S,3R-9f
2S,3S-9g 68
2S,3S-9h 68
64
98 : 2
88 : 12 88 : 12
88 : 12 81 : 19
88 : 12
7d
8d
9
6g n-Pr
6h i-Pr
6i
t-Bu
2S,3R-9i
70
97 : 3
78 : 22
a
b
Combined yields of pure syn- and anti-isomers. dr was determined
by 1H NMR analysis of the crude reaction mixture. er was determined
c
by the 1H NMR analysis of mono- or dimenthyl carbonate derivatives of
diols derived from aldol products. dr was determined by 13C NMR
d
With the optimal conditions for the selective formation of
anti-aldols in hand (Scheme 2), a series of aldehydes of varying
analysis of the crude reaction mixture.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 3152--3154 3153

View Article Online
ChemComm
Communication
and amines is key for achieving the aldol products bearing
anti- and syn-b-hydroxy-a-phenyl moieties in high dr and er.
Pure anti- and syn-diastereomers can be readily separated by
silica gel column chromatography. Both antipodes of a-pinene
are naturally available in abundance, making this stoichio-
metric asymmetric process economical and very attractive.
The authors thank the Herbert C. Brown Center for Borane
Research for financial assistance of this project.
Scheme 3 Optimum conditions for syn-selective asymmetric aldol reaction of 3
with (ꢀ)-1.
Table 3 Examination of aldehydes for syn-selection
Notes and references
1 (a) R. Mahrwald, Modern Aldol Reactions, Wiley-VCH Verlag GmbH &
Co., KGaA, Weinheim, 2004; (b) R. Mahrwald, Aldol Reactions,
Springer, Dordrecht Heidelberg, 2009.
RCHO
Aldol
#
Entry
6
R
Yielda (%) drb
erc (syn)
2 C. J. Cowden and I. Paterson, Organic Reactions, John Wiley & Sons,
New York, 1997, vol. 51, pp. 1–200.
3 It was initially reported that di-n-butylboron triflate fails to enolize
methyl propanoate. D. A. Evans, J. V. Nelson, E. Vogel and
T. R. Taber, J. Am. Chem. Soc., 1981, 103, 3099.
1
2
3
4
6a C6H5
6b 4-MeC6H4
6c 4-MeOC6H4 2S,3S-8c
6d 4-FC6H4
2S,3S-8a
75
83 : 17 >99 : o1
82 : 18 93 : 7
86 : 14 93 : 7
2S,3S-8b 73
66
2S,3S-8d 67
2S,3S-8e
2S,3S-8f
2S,3R-8g 72
2S,3R-8h 75
2S,3S-8i
84 : 16 96 : 4
5
6e 4-CF3C6H4
82
81
84 : 16 >99 : o1
4 (a) E. J. Corey and S. S. Kim, J. Am. Chem. Soc., 1990, 112, 4976;
(b) H. C. Brown and K. Ganesan, Tetrahedron Lett., 1992, 33, 3421;
(c) E. J. Corey and D.-H. Lee, Tetrahedron Lett., 1993, 34, 1737;
(d) H. C. Brown and K. Ganesan, J. Org. Chem., 1994, 59, 2336;
(e) A. Abiko, Acc. Chem. Res., 2004, 37, 387; ( f ) P. V. Ramachandran
and D. Pratihar, Org. Lett., 2009, 11, 1467.
5 (a) T. G. van Aardt, P. S. van Heerden and D. Ferreira, Tetrahedron
Lett., 1998, 39, 3881; (b) T. G. van Aardt, H. van Rensburg and
D. Ferreira, Tetrahedron, 2001, 57, 7113; (c) J. McNulty, J. J. Nair,
C. Griffin and S. Pandey, J. Nat. Prod., 2008, 71, 357; (d) J. McNulty,
J. J. Nair, M. Singh, D. J. Crankshaw and A. C. Holloway, Bioorg. Med.
Chem. Lett., 2010, 20, 2335; (e) J. McNulty, J. J. Nair, N. Vurgun,
B. R. DiFrancesco, C. E. Brown, B. Tsoi, D. J. Crankshaw and
A. C. Holloway, Bioorg. Med. Chem. Lett., 2012, 22, 718.
6 P. V. Ramachandran and P. B. Chanda, Org. Lett., 2012, 14, 4346.
7 K. Fuji, A. Mija and K. Tanaka, J. Chem. Soc., Perkin Trans. 1, 1998,
185.
6
6f
2-Thioph
83 : 17 96 : 4
7d
8d
9
6g n-Pr
6h i-Pr
85 : 15 >99 : o1
85 : 15 >99 : o1
e
6i
t-Bu
—
—
—
a
b
Combined yields of pure syn- and anti-isomers. dr was determined
by 1H NMR analysis of the crude reaction mixture. er was determined
c
by the 1H NMR analysis of mono- or dimenthyl carbonate derivatives of
diols derived from aldol products. dr was determined by 13C NMR
d
e
analysis of crude reaction mixture. This reaction afforded a 3 : 2 ratio
of syn : anti aldol product.
The dr of the reaction was determined by analyzing the
1H NMR of the crude mixture; pure syn-isomers were readily
separated by column chromatography. As earlier, reduction and
conversion to the menthyl carbonate derivative allowed for er
determination.
8 (a) I. Paterson, M. A. Lister and C. K. McClure, Tetrahedron Lett.,
1986, 27, 4787; (b) I. Paterson and J. M. Goodman, Tetrahedron Lett.,
1989, 30, 997.
The generality of the reaction (Scheme 3) was validated
under the optimum conditions for the syn-selection with the same
series of aldehydes, wherein moderate to high yields and good syn-
selectivities were achieved. Unlike the anti-isomer, the syn-isomer
was obtained in excellent er. Thus, enolization of 3 with (ꢀ)-1 in the
presence of Et3N at 0 1C, followed by aldolization with the selected
9 The melting point of CCl4, ꢀ23 1C, limits further lowering of
reaction temperature.
10 Unlike the C2–C3-proton coupling constant J of 3–6 Hz for C2-alkyl
substituted syn-aldols (ref. 4f), the C2-phenyl analogs revealed J of
7–8 Hz as reported in the literature. S. Pinheiro, M. B. Lima, C. B. S. S.
Goncalves, S. F. Pedraza and F. M. C. de Farias, Tetrahedron Lett.,
2000, 41, 4033. The 1H NMR spectrum of pure diastereomers over
time proved the stability and lack of epimerization at C2-position.
aldehydes at ꢀ78 1C in CH2Cl2 provided syn-8a–8h in 82 : 18– 11 The menthyl carbonate derivative was prepared by a slight modifi-
cation of the literature procedure: B. Halpern and J. W. Westley,
J. Org. Chem., 1968, 33, 3978 See ESI† for details.
12 The solution of ester 4 was cooled to 0 1C and the cold solution was
86:14 dr with 93:7 to >99:o1 er of the syn-isomer in 66–82%
yields. The results are summarized in Table 3.
The absolute stereochemistry of anti-9 was determined to be
2S,3R by comparing the optical rotation of methyl (2S,3R)-3-
hydroxy-2,3-diphenylpropanoate (7a) with that reported in the
transferred dropwise to the mixture of the reagent (ꢀ)-1 and Et3N at
ꢀ78 1C. After enolization for 2 h at the same temperature, aldehyde
6a was also cooled to ꢀ78 1C and added dropwise to a mixture of
enolate for aldolization.
literature.14 Similarly, the absolute stereochemistry of syn-8 was 13 While optimizing enantioselectivity of the anti-isomer, ꢀ78 1C was
found to be optimal.
determined to be 2S,3S by comparing the optical rotation of the
14 Reported [a]1D8 = ꢀ122.6 (c 1.22, MeOH), 77% ee for methyl (2S,3R)-3-
hydroxy-2,3-diphenylpropanoate (7a) (ref. 7). Observed [a]2D2 = ꢀ75.0
diol derived from ethyl syn-3-hydroxy-2,3-diphenylpropanoate
(8a) with that of the known (1S,2R)-1,2-diphenylpropane-
1,3-diol.15 On the basis of analogy, the 2S,3R configuration
has been assigned for all of the anti-aldols and the 2S,3S
configuration for the syn-aldols.16
(c 1.22, MeOH), 56% ee for the same compound, 7a.
15 Reported [a]2D5
=
ꢀ25 for (1S,2R)-1,2-diphenylpropane-1,3-diol.
H. M. L. Davies, J. Yang and J. Nikolai, J. Organomet. Chem., 2005,
690, 6111. Observed [a]2D5 = ꢀ65 (c 1, CHCl3) for (1S,2R)-1,2-diphenyl-
propane-1,3-diol, prepared by LAH reduction of ethyl 3-hydroxy-2,3-
diphenylpropanoate (8a).
In conclusion, we have developed the first boron-mediated
asymmetric aldol reaction of phenylacetates. An appropriate
matching of the alkoxy group of the ester, solvent, temperature,
16 The 2S,3S configuration has been assigned for anti-9g and 9h and
the 2S,3R configuration has been assigned for syn-8g and 8h on the
basis of Cahn–Ingold–Prelog priority rules.
c
3154 Chem. Commun., 2013, 49, 3152--3154
This journal is The Royal Society of Chemistry 2013