Highly enantioselective proline-catalysed direct aldol reaction of chloroacetone and aromatic aldehydes

Article abstract of DOI:10.1002/chem.201103667

Ready salted proline: The combination of proline and an achiral triazabicyclodecene-derived guanidinium salt permits, for the first time, the direct aldol reaction of chloroacetone and aromatic aldehydes (see scheme). The resulting chlorohydrins are formed with high regio-, diastereo- and enantioselectivity. This procedure is experimentally simple and green, working without solvent, in test tubes placed inside a standard laboratory fridge without agitation or mechanical stirring.

Full text of DOI:10.1002/chem.201103667

DOI: 10.1002/chem.201103667
Highly Enantioselective Proline-Catalysed Direct Aldol Reaction of
ChloACTHNUTRGENUGrN oacetone and Aromatic Aldehydes
ꢀngel Martꢁnez-CastaÇeda, Belꢂn Poladura, Humberto Rodrꢁguez-Solla,
Carmen Concellꢃn,* and Vicente del Amo*[a]
The discovery of the first proline-catalysed intermolecular
aldol reaction in 2000 by List, Lerner, and Barbas[1] intro-
duced a general interest in organocatalysed reactions. In the
years to come, many hundreds of papers were produced in-
spired to some degree by this phenomenal discovery.[2] Pro-
line (1) and other readily available natural amino acids were
initially investigated as organocatalysts. With their limita-
tions exposed, a breathtaking research rush took off, moti-
vated by preparing novel tailor-made catalysts exhibiting im-
proved reactivity and capable of mediating novel chemical
processes, otherwise ineffective with proline. Impressive cat-
alyst structures were found in this way. However, the huge
effort required for their preparation makes most of these
compounds unattractive in industry, which is unable to
afford lengthy multistep syntheses of catalysts to be tested
on a particular reaction. As an alternative, our research
group[3] and others[4] have demonstrated the methods by
which, in classical reactions, the addition of simple additives
can enhance the reactivity and selectivity of an off-the-
bench catalyst: proline. Herein, we report the first[5] proline-
catalysed asymmetric synthesis of chlorohydrins, through the
intermolecular aldol reaction between chloroacetone and ar-
omatic aldehydes, made feasible by the participation of
Scheme 1. Organocatalysts employed in the direct aldol reaction between
chloroacetone and aromatic aldehydes to give chlorohydrins 6 and 7.
a guanidinium salt as an additive.
The organocatalytic stereoselective construction of halo-
genated chiral carbon centres is a challenging synthetic op-
eration.[6] A collection of organocatalysts, 2,[7] 3,[8] 4[9] and
of catalysts 2–5 requires elaborate synthetic manipulations
and catalysts 3 and 4, in particular, contain expensive chiral
motifs, (S)-2-amino-2’-hydroxy-1,1’-binaphthyl ((S)-NOBIM)
and (S)-2,2’-diamino-1,1’-binaphthyl ((S)-BINAM), respec-
tively.
5,[10] has been tested on the direct aldol reaction of chloro-
ACHTUNGTRENNUNGacetone and aromatic aldehydes to give chlorohydrins 6 and
7, as shown in Scheme 1. Although the enantioselectivities
achieved are acceptable, the regioselectivity, 6/7, and diaste-
reoselectivity, anti-7/syn-7, of the reaction is modest, except
in selected examples. It must be noted that the preparation
In a previous contribution,[3] we disclosed that the partici-
pation of catalytic amounts (10 mol%) of achiral triazabicy-
clodecene (TBD)-derived guanidinium tetrafluoroborate 8
improved the proline-catalysed intermolecular direct aldol
reaction of aromatic and heteroaromatic aldehydes with var-
ious ketones. The desired aldol products were obtained in
excellent yield, with enantiomeric excesses (ee values) peak-
ing at 99% and diastereoselectivities up to 97:3, favouring
the anti configuration. This protocol is remarkably simple,
taking place under solvent-free conditions, with no stirring,
inside test tubes placed in a standard laboratory fridge.
Motivated by these results, we decided to explore the be-
haviour of our catalytic guanidinium salt/proline system to-
wards the reaction illustrated in Scheme 1. 4-Nitrobenzalde-
[a] ꢀ. Martꢁnez-CastaÇeda, B. Poladura, Dr. H. Rodrꢁguez-Solla,
Dr. C. Concellꢂn, Dr. V. del Amo
Departamento de Quꢁmica Orgꢃnica e Inorgꢃnica
Universidad de Oviedo, C/Juliꢃn Claverꢁa 8, 33006, Oviedo (Spain)
Fax : (+34)985-10-3448
E-mail: ccf@uniovi.es
vdelamo@uniovi.es
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201103667. It contains detailed
synthetic protocols, spectroscopic characterisation of chlorohydrins
7a–k and epoxides 10a, c, d, f and k and HPLC chromatograms.
5188
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 5188 – 5190

COMMUNICATION
hyde 9a was selected as the model substrate for preliminary
reactions. Satisfyingly, in accordance with our previous reac-
and high regio-, diastereo- and enantioselectivity, giving
over 90% ee in all cases, independent of the nature of the
aldehyde employed, thus, confirming the reproducibility and
robustness of this transformation. Moreover, aldol reactions
performed without additive 8 showed significantly poorer
regio- and diastereoisomeric ratios for chlorohydrins 7, as
well as poorer enantiomeric excesses (see the Supporting In-
formation for details).
tion conditions, when
a suspension of (S)-proline (1;
15 mol%), 8 (10 mol%) and 4-nitrobenzaldehyde 9a in
chloroacetone was left to stand for 20 days[11] inside a stan-
dard laboratory fridge (temperature set up and checked to
be 0–38C), without requiring any sort of stirring or mechani-
cal agitation, a mixture of chlorohydrins 6a and 7a was pro-
duced with good regio- (96:4, 7a/6a), diastereo- (anti-7a/
syn-7a, d.r. 91:9) and enantioselectivity (98% ee for anti-7a;
Table 1, entry 1). Aiming to shorten the reaction time, we
Chlorohydrins 7a–k[12] were characterised from the crude
reaction mixtures as a consequence of their instability to-
wards chromatography, either on silica gel or alumina. Fur-
thermore, we observed that storing products 7 at room tem-
perature for several days caused epimerisation of the stereo-
genic centres with a consequent decrease in the selectivity
figures collected in Table 1. Intending to transform these a-
chloro-b-hydroxy ketones into easily isolable compounds,
we contemplated the possibility of acetylating 6 and 7 under
classical conditions.[13] Unfortunately, all efforts to per-
form this reaction proved unfruitful and afforded com-
plex reaction mixtures.
Table 1. (S)-Proline/guanidinium salt co-catalysed synthesis of chlorohy-
drins 7a–k.[a]
In view of these results, we decided to prepare a,b-
epoxy ketones 10 from representative crude a-chloro-b-
hydroxy ketones through an intramolecular SN2 reaction
by following a literature procedure,[9a] or by treatment with
N,N-diisopropylethylamine (DIPEA; Table 2, entries 1–6).
Epoxides 10 were readily isolated after flash column chro-
matography as single trans-diastereoisomers of absolute con-
figuration (3R,4S).[14] In our hands, the enantioselectivity of
epoxides 10 obtained in this manner was lower than that of
the corresponding anti-a-chloro-b-hydroxy ketones 7 out-
lined in Table 1. A simple epimerisation process of the C3
ArCHO
Conversion Regioselectivity[b] d.r.[c] ee
[%][b]
[%][d]
1
9a 4-NO2-C6H4
99
99
96:4
99:1
91:9 98
83:17 95
92:8 97
93:7 97
91:9 97
90:10 98
94:6 94
94:6 95
93:7 96
93:7 93
90:10 92
93:7 94
2[e] 9a 4-NO2-C6H4
3
4
5
6
9b 3-NO2-C6H4
9c 2-NO2-C6H4
9d 4-CO2Me-C6H4
9e 4-CN-C6H4
97
96:4
98
96
>99:1
99:1
>99
95
96:4
92:8
7[f] 9 f 3-F-C6H4
8
9
10
11
12
9g 4-Cl-C6H4
9h 3-Cl-C6H4
9i 4-Br-C6H4
9j 2-Br-C6H4
9k C6H5
79
95:5
98
96:4
77
97:3
Table 2. Epoxidation reaction of representative a-chloro-b-hydroxy ke-
tones 6 and 7.[a]
90
99
>99:1
98:2
[a] Reaction conditions: chloroacetone (4.0 mmol), ArCHO (0.4 mmol),
(S)-proline (1; 15 mol%), 8 (10 mol%), no solvent. The reaction mixture
was left to stand for 20 days inside a fridge (0–38C) with no stirring.
1
[b] Determined by H NMR spectroscopy of the crude reaction mixtures.
Conversion of aldehydes 9 (limiting reagent) into chlorohydrins 6 and 7.
1
[c] Diastereoisomeric ratio of anti- to syn-7 determined by H NMR spec-
Ar
Product
Yield
[%][b]
ee
[%][c]
troscopy of the crude reaction mixtures and identified by comparison
with similar compounds previously described in the literature (anti-dia-
stereoisomers: references [7 and 9a]; syn-diastereoisomers: refer-
ence [10]). [d] Enantiomeric excess of chlorohydrins anti-7 as determined
by chiral HPLC of the crude reaction mixtures. [e] Reaction carried out
at 258C and stirred for 6 days. [f] The reaction mixture was left to stand
for 14 days inside a fridge (0–38C) with no stirring.
1
4-NO2-C6H4
4-NO2-C6H4
2-NO2-C6H4
4-CO2Me-C6H4
3-F-C6H4
C6H5
2-NO2-C6H4
3-F-C6H4
10a
10a
10c
10d
10 f
10k
10c
10 f
59
63
52
66
49
61
55
33
70 (98)
73 (95)
90 (97)
81 (98)
86 (94)
87 (94)
85
2[d]
3
4
5
6
7[e]
8[e]
79
decided to follow an analogous reaction at room tempera-
ture until the disappearance of aldehyde 9a. Under this set
of conditions, the reaction was completed in 6 days, although
at the expense of a severe and unacceptable drop in the se-
lectivity figures, particularly in terms of diastereoselectivity
(Table 1, entry 2).
Without further optimisation, a selection of aldehydes,
9b–k, containing different functional groups and substitution
patterns, were reacted with chloroacetone under our reac-
tion conditions (Table 1, entries 3–12). All of the reactions
shown in Table 1 proceeded smoothly, with good conversion
[a] Reaction conditions: a mixture of chlorohydrins 6 and 7 (0.4 mmol,
ratio as obtained in the reactions presented in Table 1), dry NEt3
(0.56 mmol), dry CH2Cl2 (1 mL). [b] Isolated yield of analytically pure
products calculated from the corresponding aldehydes 9. [c] Enantiomer-
ic excess of pure trans-epoxides 10 as determined by chiral HPLC. Enan-
tiomeric excess of the corresponding anti-a-chloro-b-hydroxy ketone 7 is
given in brackets. [d] DIPEA was used instead of dry NEt3. [e] Reaction
conditions: chloroacetone (4.0 mmol), ArCHO (0.4 mmol), (S)-proline
(1; 15 mol%), 8 (10 mol%), no solvent. The reaction mixture was left to
stand for 20 days inside a fridge (0–38C) with no stirring. The resulting
mixture was allowed to warm to RT, then diluted by pouring into dry
CH2Cl2 (1 mL), treated with dry NEt3 (0.56 mmol) and stirred for 48 h.
Chem. Eur. J. 2012, 18, 5188 – 5190
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5189

V. del Amo, C. Concellꢂn et al.
À
stereocentre (C Cl bond) cannot justify the reduction in ee
[1] B. List, R. A. Lerner, C. F. Barbas III, J. Am. Chem. Soc. 2000, 122,
2395.
upon isolation of trans-epoxides 10. It must be noted that
epimerisation of anti- to syn-chlorohydrins 7 and subsequent
SN2 reaction would render the corresponding cis-epoxides,
which are not detected experimentally in the epoxidation re-
actions investigated herein. Based on this observation, it can
be postulated that both anti and syn diastereoisomers of
chlorohydrins 7, of varying ee for each diastereoisomer,
afford, upon treatment with the amine base, the most stable
trans-epoxides 10 through SN2, or alternatively SN1, mecha-
nisms. With this concession, the overall ee for isolated trans-
epoxides 10 reflects the stereochemical integrity of the C4
[2] For some authoritative reviews on organocatalysis, see: a) P. I.
Dalko, L. Moisan, Angew. Chem. 2001, 113, 3840; Angew. Chem.
Int. Ed. 2001, 40, 3726; b) B. List, Synlett 2001, 1675; c) B. List, Tet-
rahedron 2002, 58, 5573; d) P. I. Dalko, L. Moisan, Angew. Chem.
2004, 116, 5248; Angew. Chem. Int. Ed. 2004, 43, 5138; e) W. Notz,
F. Tanaka, C. F. Barbas, Acc. Chem. Res. 2004, 37, 580; f) J. Seayad,
B. List, Org. Biomol. Chem. 2005, 3, 719; g) B. List, Chem.
Commun. 2006, 819; h) G. Lelais, D. W. C. MacMillan, Aldrichimica
Acta 2006, 39, 79; i) S. Mukherjee, J. W. Yang, S. Hoffmann, B. List,
Chem. Rev. 2007, 107, 5471; j) A. Dondoni, A. Massi, Angew. Chem.
2008, 120, 4716; Angew. Chem. Int. Ed. 2008, 47, 4638; k) P. Mel-
chiorre, M. Marigo, A. Carlone, G. Bartoli, Angew. Chem. 2008, 120,
6232; Angew. Chem. Int. Ed. 2008, 47, 6138; l) C. F. Barbas, Angew.
Chem. 2008, 120, 44; Angew. Chem. Int. Ed. 2008, 47, 42; m) Z.
Shao, H. Zhang, Chem. Soc. Rev. 2009, 38, 2745; n) S. Bertelsen,
K. A. Jørgensen, Chem. Soc. Rev. 2009, 38, 2178; o) X. Liu, L. Lin,
X. Feng, Chem. Commun. 2009, 6145.
À
stereocentre (C OH) from both anti and syn chlorohydrins
7. Nonetheless, previous studies on the formation of related
a,b-epoxy ketones[15] and esters,[16] from their corresponding
a-halo-b-hydroxy derivatives, propose the participation of
these intermediates in a base-promoted aldol/retro-aldol dy-
namic equilibrium coupled with irreversible intramolecular
SN2 steps, which provides an alternative explanation for the
observed selectivity for the trans-epoxides.
[3] A. Martꢁnez-CastaÇeda, B. Poladura, H. Rodrꢁguez-Solla, C. Concel-
lꢂn, V. del Amo, Org. Lett. 2011, 13, 3032.
[4] a) Y. Zhou, Z. Shan, J. Org. Chem. 2006, 71, 9510; b) ꢅ. Reis, S.
Eymur, B. Reis, A. S. Demir, Chem. Commun. 2009, 1088; c) X.
Companyꢂ, G. Valero, L. Crovetto, A. Moyano, R. Rios, Chem. Eur.
J. 2009, 15, 6564; d) N. El-Hamdouni, X. Companyꢂ, R. Rios, A.
Moyano, Chem. Eur. J. 2010, 16, 1142; e) G. Ma, A. Bartoszewicz, I.
Ibrahem, A. Cꢂrdova, Adv. Synth. Catal. 2011, 353, 3114.
[5] The aldol reaction between chloroacetone and aromatic aldehydes
has only been performed by using 50 mol% proline in ionic liquids
as the solvent. Only two examples are reported. Enantiomeric ex-
cesses of the desired chlorohydrins are not described: T. Kitazume,
Z. Jiang, K. Kasai, Y. Mihara, M. Suzuki, J. Fluorine Chem. 2003,
121, 205.
In either case and working towards improving this trans-
formation, we studied the chances of carrying out the aldol
and subsequent epoxidation reactions in a one-pot sequence,
that is, with no isolation or analysis of the intermediate
chlorohydrins[17] (Table 2, entries 7 and 8). The correspond-
ing trans-epoxides were isolated cleanly, after flash column
chromatography, in comparable yield and enantioselectivity.
In conclusion, we have developed a methodology for the
direct intermolecular aldol reaction of chloroacetone and ar-
omatic aldehydes to provide a-chloro-b-hydroxy ketones.
This protocol employs for the first time catalytic amounts of
(S)-proline, aided by the participation of a TBD-derived tet-
rafluoroborate guanidinium salt. The results obtained are
better, in terms of yield and regio-, diastereo- and enantiose-
lectivity than those reported with other sophisticated, non-
commercially available organocatalysts. This procedure is
experimentally simple and green, working without solvent,
in test tubes placed inside a standard laboratory fridge with-
out agitation or mechanical stirring. Experiments searching
for new additives that allow proline, or other simple organo-
catalysts of its type, to promote organic reactions, which are
not otherwise possible, are currently being investigated in
our laboratory.
[6] K. Shibatomi, Synthesis 2010, 2679.
[7] L. He, Z. Tang, L. F. Cun, A. Q. Mi, Y. Z. Jiang, L.- Z. Gong, Tetra-
hedron 2006, 62, 346.
[8] A. Russo, G. Botta, A. Lattanzi, Tetrahedron 2007, 63, 11886.
[9] a) G. Guillena, M. d. C. Hita, C. Nꢃjera, Tetrahedron: Asymmetry
2007, 18, 1272; b) G. Guillena, M. d. C. Hita, C. Nꢃjera, S. F. Viꢂ-
zquez, J. Org. Chem. 2008, 73, 5933.
[10] X.-Y. Xu, Y.-Z. Wang, L.-Z. Gong, Org. Lett. 2007, 9, 4247.
[11] The course of the reaction was carefully monitored by TLC. The 20
days were required for the disappearance of aldehyde 9a from the
reaction media to be observed.
[12] The absolute configuration of chlorohydrins anti-7 was assigned to
be (3S,4S) by comparison of the chiral HPLC trace for compound
anti-7a to that previously reported in the literature (reference [8]).
[13] AcCl (or Ac2O) and pyridine (or NEt3) in a dry organic solvent.
[14] The absolute configuration of epoxides 10 was assigned to be
(3R,4S) by comparison of the specific rotation values obtained for
compound 10k, ([a]2D0 =À80.0 (c=1.0 in CHCl3, 62% ee)) with those
previously reported in the literature (reference [5] and T. Nemoto,
T. Ohshima, K. Yamaguchi, M. Shibasaki, J. Am. Chem. Soc. 2001,
123, 2725). Furthermore, a comparison of the HPLC trace of epox-
ide 10k (Chiralpak AD-H column, hexane/iPrOH (98:2),
1.25 mLminÀ1, l=220 nm, tr =8.8 (major), 10.9 min (minor)), with
that previously reported in the literature (reference [9a]) authenti-
cates its absolute configuration.
Acknowledgements
[15] a) S. Arai, Y. Shirai, T. Ishida, T. Shioiri, Tetrahedron 1999, 55, 6375;
b) C. Peppe, R. P. das Chagas, Synlett 2006, 0605.
[16] M. E. Krafft, S. J. R. Twiddle, J. W. Cran, Tetrahedron Lett. 2011, 52,
1277.
[17] Experimental procedures are detailed in the Supporting Informa-
tion.
We thank MICINN (CTQ2010-14959) for financial support. B.P., C.C.
and V.d.A. thank MICINN for an FPI pre-doctoral fellowship, a Juan de
la Cierva contract and a Ramꢂn y Cajal contract, respectively.
Received: November 21, 2011
Revised: February 24, 2012
Keywords: aldol
reaction
·
guanidinium
salts
·
organocatalysis · proline
Published online: March 20, 2012
5190
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 5188 – 5190