A new class of urea-substituted cinchona alkaloids promote highly enantioselective nitroaldol reactions of trifluoromethylketones

Article abstract of DOI:10.1021/ol103089j

The first class of bifunctional cinchona-alkaloid catalysts incorporating a urea moiety at C-5′ has been developed. These materials catalyze the efficient and highly enantioselective 1,2-addition of nitromethane to trifluoromethylketones to form synthetically pliable products incorporating a quaternary stereocenter. Excellent product yields and levels of enantiomeric excess are possible, and the optimum catalyst structure is capable of promoting the Henry reaction involving alkyl trifluoromethylketones with unprecedented enantioselectivity.

Full text of DOI:10.1021/ol103089j

ORGANIC
LETTERS
2011
Vol. 13, No. 6
1298–1301
A New Class of Urea-Substituted
Cinchona Alkaloids Promote Highly
Enantioselective Nitroaldol reactions
of Trifluoromethylketones
Carole Palacio and Stephen J. Connon*
Centre for Synthesis and Chemical Biology, School of Chemistry, University of Dublin,
Trinity College, Dublin 2, Ireland
connons@tcd.ie
Received December 21, 2010
ABSTRACT
The first class of bifunctional cinchona-alkaloid catalysts incorporating a urea moiety at C-50 has been developed. These materials catalyze the
efficient and highly enantioselective 1,2-addition of nitromethane to trifluoromethylketones to form synthetically pliable products incorporating a
quaternary stereocenter. Excellent product yields and levels of enantiomeric excess are possible, and the optimum catalyst structure is capable of
promoting the Henry reaction involving alkyl trifluoromethylketones with unprecedented enantioselectivity.
Over the past 5 years, cinchona alkaloid derivatives
incorporating a C-9-(thio)urea component (i.e., 1-2
Figure 1) have emerged as powerful catalysts for a range
of asymmetric transformations susceptible to the influence
of general acid-base catalysis, including Michael-type
reactions, nonconcerted cycloadditions, and the additions
of pronucleophiles to imines.1-3 While the success of these
systems is undoubtedly related to the proximity of the
mutually compatible bifunctional components in a well-
defined chiral environment, from a design perspective, one
is limited in what can be achieved in terms of tuneability: i.e.,
while the (thio)urea component can be modified/replaced
(as elegantly demonstrated by Rawal,4 for example), the
relative positioning of the nucleophile- and electrophile-
activating components (a key feature of any bifunctional
catalyst) remains relatively fixed in all such systems. There-
fore, if the stereoelectronic demands of a particular general
acid/base catalyzed reaction happen not to overlay neatly
with the distance between the C-9-substituted catalyst’s
bifunctional components, it is unlikely that any amount of
modification of the (thio)urea substituent will result in
efficient catalysis.
(1) Selected reviews: (a) Hiemstra, H.; Marcelli, T. Synthesis 2010,
1229. (b) Connon, S. J. Synlett 2009, 354. (c) Yu, X.; Wang, W. Chem.
Asian J. 2008, 3, 516. (d) Connon, S. J. Chem. Commun. 2008, 2499. (e)
Doyle, A. G.; Jacobsen, E. N Chem. Rev. 2007, 107, 5713. (f) Taylor,
M. S.; Jacobsen, E. N. Angew. Chem., Int. Ed. 2006, 45, 1520. (g)
Connon, S. J. Chem.;Eur. J. 2006, 12, 5419. (h) Marcelli, T.; van
Maarseveen, J. H.; Hiemstra, H. Angew. Chem., Int. Ed. 2006, 45, 7496.
(2) (a) Lee, B. -J.; Jang, L.; Liu, M.; Chen, Y. -C.; Ding, L. -S.; Wu, Y.
With this in mind, we became interested in developing
systems which retain the essential elements of the success-
ful bifunctional system 1-2, yet in which the distance
between the bifunctional components has been varied.
ꢀ
ꢀ
Synlett 2005, 603. (b) Vakulya, B.; Varga, S.; Csampai, A.; Soos, T. Org.
Lett. 2005, 7, 1967. (c) McCooey, S. H.; Connon, S. J. Angew. Chem., Int.
Ed. 2005, 44, 6367. (d) Ye, J.; Dixon, D. J.; Hynes, P. S. Chem. Commun
2005, 4481.
(3) For an account concerning seminal work on non-cinchona alka-
loid-based bifunctional (thio)urea catalysts, see: Miyabe, H.; Takemoto,
Y. Bull. Chem. Soc. Jpn. 2008, 81, 785.
(4) Malerich, J. P; Hagihara, K.; Rawal, V. H. J. Am. Chem. Soc.
2008, 130, 14416.
r
10.1021/ol103089j
Published on Web 02/21/2011
2011 American Chemical Society

while use of 4b resulted in poor yield and low levels of
product enantiomeric excess. No further development of
these types of catalyst have since been reported.
We therefore embarked upon the development and eva-
luation of a small catalyst library of general structure 5.
These systems incorporate variable substituents at C-910,11
and C-50 so that the steric requirement and electronic/
hydrogen bond donating properties of the catalyst can be
systematically modulated. We wished to evaluate the per-
formance of these catalysts in the enantioselective Henry
reaction involving trifluoromethylketone substrates.
This reaction was selected for a number of reasons: (a) it
is synthetically useful, generating trifluoromethylated chiral
products (of general interest due to their unique properties12
and pharmaceutical/agrochemical relevance13) incorpo-
rating a quaternary stereocenter;14 (b) it represented a sig-
nificant challenge—at the outset of this study the only
direct catalytic methodology for the enantioselective catalysis
of this reaction utilized 25 mol % of a chiral La complex
at -40 °C;15,16 (c) as mentioned previously, bifunctional
(thio)ureas have found widespread application in the cata-
lysis of additions to imines and Michael acceptors; however,
relatively few reactions involving additions to carbonyl
compounds promoted by this class of catalyst have been
reported;1,3,17 and (d) since Hiemstra5 reported that 3a
promoted Henry reactions with aldehyde substrates, we were
intrigued as to how a catalyst characterized by bifunctional
components in closer proximity would fare in an analogous,
yet more challenging 1,2-addition reaction.
Figure 1. Representative C-9-, C-60-, and C-50-modified cincho-
na alkaloid catalysts and the proposed novel library 5.
Inspired by the catalytically useful cupreine/cupreidine
family1a,h of natural product derivatives, Hiemstra et al.
previously developed the C-60 thiourea-substituted cinch-
ona alkaloid catalyst 3a and demonstrated it capable of
highly efficient asymmetric nitroaldol reactions involving
aromatic aldehydes,5 while a similar catalyst with a larger
C-9 substituent for catalysis of the addition of thiols to
enones was later developed by Deng.6,7
As this work was in progress, Bandini et al.18 reported
the first organocatalytic variant of this process using a C-9
acylated cupreine derivative. Under optimized conditions,
excellent product yields and enantioselectivities were possible.
Our study began with the synthesis of a small library of
cinchona alkaloids substituted at C-50 with functionality
capable of hydrogen-bond donation (i.e., 5a-j, Table 1).
These were evaluated as promoters of the addition of
nitromethane to R,R,R-trifluoroacetophenone (6) in THF
at ambient temperature.
The installation of catalytically useful functionality at
C-50 is considerably less well explored. Jørgensen8 and
Deng9 have independently developed hydrazide-substi-
tuted catalysts of general type 4a and 4b, respectively,
which exploit the reaction between cupreines and diazodi-
carboxylate esters at C-50. While Jørgensen demonstrated
interesting catalytic activity in Michael additions to acro-
lein and in the amination of 2-naphthols, Deng found that
4a failed to promote the R-amination of a β-keto ester,
(5) (a) Marcelli, T.; van der Haas, R. N. S.; van Maarseveen, J.;
Hiemstra, H. Angew. Chem., Int. Ed. 2006, 45, 929. (b) Hammar, P.;
Marcelli, T.; Hiemstra, H.; Himo, F. Adv. Synth. Catal. 2007, 349, 2537.
(6) Liu, Y.; Sun, B.; Wang, B.; Wakem, M.; Deng, L. J. Am. Chem.
Soc. 2009, 131, 418.
(12) Reviews: (a) Schlosser, M. Angew. Chem., Int. Ed. 1998, 37,
1496. (b) Smart, B. E. J. Fluorine Chem. 2001, 109, 3.
(13) For an example, see: Pierce, M. E.; Parsons, R. L., Jr.; Radesca,
L. A.; Lo, Y. S.; Silverman, S.; Moore, J. R.; Islam, Q.; Choudhury, A.;
Fortunak, J. M. D.; Nguyen, D.; Luo, C.; Morgan, S. J.; Davis, W. P.;
Confalone, P. M.; Chen, C. -Y.; Tillyer, R. D.; Frey, L.; Tan, L.; Xu, F.;
Zhao, D.; Thompson, A. S.; Corley, E. G. J. Org. Chem. 1998, 63, 8536.
(7) Two examples of cinchona alkaloid-based catalysts substituted at
C-60 with phenyl and amide groups, respectively, have been reported;
€
€
see:(a) Waldmann, H.; Khedkar, V.; Duckert, H.; Schurmann, M.;
Oppel, I. M.; Kumar, K. Angew. Chem., Int. Ed. 2008, 47, 6869. (b)
Abermil, N.; Masson, G.; Zhu, J. J. Am. Chem. Soc. 2008, 130, 12596.
(8) (a) Brandes, S.; Bella, M.; Kjoersgaard, A.; Jørgensen, K. A.
Angew. Chem., Int. Ed. 2006, 45, 1147. (b) Brandes, S.; Niess, B.; Bella,
M.; Prieto, A.; Overgaard, J.; Jørgensen, K. A. Chem.;Eur. J. 2006, 12,
6039.
ꢀ
(14) Selected recent reviews: (a) Tur, F; Mansilla, J.; Lillo, V. J.; Saa,
J. M. Synthesis 2010, 1909. (b) Bella, M.; Gasperi, T. Synthesis 2009,
1583.(c) Cozzi, P. G.; Hilgraf, R.; Zimmermann, N. Eur. J. Org. Chem.
2007, 5969. (d) Trost, B. M.; Jiang, C. Synthesis 2006, 369. (e) Christoffers,
J.; Baro, A. Adv. Synth. Catal. 2005, 347, 1473.
ꢀ
(15) Tur, F.; Saa, J. M. Org. Lett. 2007, 9, 5079.
(9) Liu, X.; Sun, B.; Deng, L. Synlett 2009, 1685.
(16) Deng has demonstrated cupreine derivatives to be highly effec-
tive catalysts for the addition of nitromethane to R-ketoesters: Li, H.;
Wang, B.; Deng, L. J. Am. Chem. Soc. 2006, 128, 732.
(17) Our group have had a particular interest in the use of (thio)ureas
to promote 1,2-addition reactions: (a) Maher, D. J.; Connon, S. J.
Tetrahedron Lett. 2004, 45, 1301. (b) Procuranti, B.; Connon, S. J. Chem.
Commun. 2007, 1421. (c) Peschiulli, A.; Gun’ko, Y.; Connon, S. J. J. Org.
Chem. 2008, 73, 2454. (d) Peschiulli, A.; Quigley, C.; Tallon, S.; Gun’ko,
Y. K.; Connon, S. J. J. Org. Chem. 2008, 73, 6409. (e) Kavanagh, S. A.;
Piccinini, A.; Fleming, E. M.; Connon, S. J. Org. Biomol. Chem. 2008, 6,
1339.
(10) For an early example illustrating the importance of the C-9
substituent in controlling catalyst conformation in cinchona alkaloid
systems, see: Cortez, G. S.; Oh, S. H.; Romo, D. Synthesis 2001, 1731.
(11) Deng has found that the C-9 substituent can play a key role in
determining catalyst efficacy in cupreine-type systems; for examples, see:
(a) Li, H.; Wang, Y.; Tang, L.; Wu, F.; Liu, X.; Guo, C.; Foxman, B. M.;
Deng, L. Angew. Chem., Int. Ed. 2005, 44, 105. (b) Li, H.; Song, J.; Liu,
X.; Deng, L. J. Am. Chem. Soc. 2005, 127, 8948. (c) Wang, Y.; Liu, X.;
Deng, L. J. Am. Chem. Soc. 2006, 128, 3928. (d) Wu, F.; Li, H.; Hong, R.;
Deng, L. Angew. Chem., Int. Ed. 2006, 45, 947. (e) Wu, F.; Hong, R.;
Khan, J.; Liu, X.; Deng, L. Angew. Chem., Int. Ed. 2006, 45, 4301. (f) Li,
H.; Wang, B.; Deng, L. J. Am. Chem. Soc. 2006, 128, 732.
(18) Bandini, M.; Sinisi, R.; Umani-Ronchi, A. Chem. Commun 2008,
4360.
Org. Lett., Vol. 13, No. 6, 2011
1299

(ca. 50% ee, entries 3 and 4). 5-Aminodihydroquinine (5a)
catalyzed rapid 1,2-addition relatively unselectively, as did its
C-9 methylated analogue 5b (entries 5 and 6 respectively).
Conversion of the C-50 amino group to a Boc moiety (to
augment its hydrogen-bond donating characteristics) led to
significant improvement in catalyst performance (entries 7
and 8); however, product ee remained modest. Benzoylation
of 5c at C-9 (i.e., catalyst 5e, entry 9) resulted in a dramatic
loss of catalyst activity.
Table 1. Evaluation of C-50-Substituted Catalysts in the Henry
Reaction Involving Trifluoroacetophenone (6)
We were pleased to find that the conversion of catalyst
5b to the corresponding urea derivative 5f19 resulted in a
catalyst capable of promoting the reaction with signifi-
cantly higher enantioselectivity (39% ee, entry 10), while
again installation of a benzoylunitat C-9 proved unhelpful
(i.e., catalyst 5g, entry 11). Given the deleterious effect of
acylation at C-9 on catalyst performance, we were sur-
prised to observe that the C-9 carbamoyl derivative 5h
possessed excellent catalytic activity and could promote
the reaction with superior product enantiomeric excess
(quantitative conversion, 62% ee at 2 mol % catalyst
loading) to that obtained using any of the other catalysts
in this study (entry 12).20
To shed some light on the influence of the carbamoyl
group on catalyst efficacy we prepared 5i, which is devoid
of the C-50 urea. This catalyst was both relatively inactive
and unselective (entry 13). It also promoted the formation of
the opposite dominant enantiomer to that observed in
reactions mediated by 5h. We would therefore suggest that
the superiority of 5h to 5f and the inferiority of 5i to 5h signals
that the C-9 carbamoyl moiety is not directly involved in
catalysis by hydrogen-bond donation. It is obvious, however,
that its presence in conjunction with the C-50 urea is advan-
tageous, most likely because it plays an as yet unidentified
role in controlling catalyst conformation.
Interestingly, the attempted replacement of the C-9
carbamoyl moiety with another carbamate substituent
incapable of hydrogen bond donation led to the formation
of the N-acylurea 5j as the major product.21 This catalyst
promoted the reactions with respectable enantioselectivity,
albeit at a considerably slower rate than 5h (entry 14).
With the identification of the best catalyst complete, we
next proceeded to optimize the reaction conditions: en-
antioselectivity diminished when the reactions were con-
ducted in toluene and dichloromethane, while the use of
MTBE as the reaction medium facilitated faster 1,2-addition
with similar levels of product ee to those observed in THF
(entries 15-17). Lower loadings of the nucleophile led to
improved enantioselectivity at the expense of reaction rate
(entries 18 and 19), while reduced reaction temperature
entry
cat.
X
solv
temp
time
(h)
yielda
(%)
eeb
(%)
(mol %)
(°C)
1
8 (2)
10 THF
10 THF
10 THF
10 THF
10 THF
10 THF
10 THF
10 THF
10 THF
10 THF
10 THF
10 THF
10 THF
10 THF
10 PhMe
10 MTBE
10 CH2Cl2
18
18
4
4
70
96
>98
91
88
75
22
91
22
66
2
6
2
9 (2)
5
-54
52
7
3
1 (2)
18
4
4c
5
2 (2)
18
21
18
18
22
22
64
21
21
21
21
21
21
21
21
21
21
21
72
72
72
5a (2)
5b (2)
5c (2)
5d (2)
5e (2)
5f (2)
5g (2)
5h (2)
5i (2)
5j (2)
5h (2)
5h (2)
5h (2)
5h (2)
5h (2)
5h (2)
5h (20)
5h (20)
5h (10)
18
6
18
13
10
20
7
7
18
8
18
9
18
10c
11
12d
13
14
15
16
17
18
19
20
21
22
23
18
39
18
18
>98
18
50
90
99
80
98
50
77
81
62
72
62
-8
48
12
56
10
66
69
87
88
90
90
18
18
18
18
18
5
3
THF
THF
18
18
10 THF
-25
-25
-25
-25
7
5
7
MTBE
MTBE
MTBE
a Determined by 1H NMR spectroscopy using an internal standard.
b Determined by CSP-HPLC; see the Supporting Information. c 76%
yield after 4 h reaction time. d After 4 h reaction time: 30% yield, 40% ee.
e After 4 h reaction time: 83% yield, 62% ee.
(19) The corresponding C-50 thiourea derivatives proved hydrolyti-
cally unstable. We suspect this is related to general-base catalysis of
thiourea hydrolysis by the proximal quinuclidine ring.
The parent alkaloids, quinine (8) and quinidine (9), proved
efficient catalysts of the reaction at just 2 mol % loading;
however the β-nitro alcohol product 7 was formed with very
low levels of enantiomeric excess (entries 1 and 2). The
benchmark C-9 thiourea-modified catalysts 1 and 2 (X =
S, Y = Et, Z = OMe) also promoted the smooth formation
of 7, with significantly higher levels of enantiomeric excess
(20) Interestingly, nitroethane proved a poor nucleophile in these
reactions. Selected data (rt, THF solvent, nitroethane in 10-fold excess).
Cat. 5f (2 mol %): 89 h, 35% conv, dr 1:7, 33% ee (major), 5% ee
(minor). Cat. 1 (2 mol %): 4 h, 17% conv, dr 1.4:1, 22% ee (major), 5%
ee (minor). Cat. 5h (10 mol %): 63 h, 59% conv, dr 2:1, 12% ee (major),
3% ee (minor).
(21) Catalyst 5j was formed from the reaction of 50-amino-(N,N-
dimethylcarbamoyldihydroquinine with 1 equiv of 3,5-bis-trifluoro-
methylphenyl isocyanate in dichloromethane solvent.
1300
Org. Lett., Vol. 13, No. 6, 2011

The next task was to evaluate the substrate scope: we
were pleased to find that p-fluoro-R,R,R-trifluoroacetophe-
none (8) could be converted to 16 in good yield and excellent
enantiomeric excess (Table 2, entry 1). The corresponding
meta-isomer 9afforded 17 in slightly lower ee (entry 2), as did
the p-bromo and p-methyl analogues 10 and 11 (i.e., products
18 and 19, respectively, entries 3 and 4). The thienyl-sub-
stituted alcohol 20 was formed in 82% ee, which, while lower
than that obtained using nonheterocyclic acetophone deri-
vatives, represents a significant improvement over the pre-
vious literature benchmark for substrate 12 (entry 5).
Gratifyingly, alkyl trifluoromethylketones (i.e., 13-15);
a substrate class which has more proven problematic in
previous studies15,18;proved particularly susceptible to the
influence of catalyst 5h. Good yields and excellent levels of
product ee (g95%) were recorded in all cases (i.e., 21-23,
entries 6-8). To the best of our knowledge this represents the
most enantioselective general process for the catalytic asym-
metric Henry reaction involving this substrate class to date.
In summary, we have developed a new class of cinchona
alkaloid-based organocatalyst characterized by the pre-
sence of a C-50 urea moiety, which an examination of
models suggests is in very close proximity to the quinucli-
dine ring nitrogen atom. In a challenging nitroaldol reac-
tion process, the most selective of these materials (catalyst
5h) outperformed the traditional C-9-thiourea substituted
catalysts 1 and 2 from both catalyst activity and product
enantioselectivity standpoints. Under optimized condi-
tions 5h could promote the formation of the products of
significant potential synthetic utility possessing quaternary
stereocenters in good yields and excellent enantiomeric
excess. Alkyl-substituted trifluoromethylketone substrates
were converted to the corresponding nitroaldol products
with the highest levels of enantiomeric excess recorded in
the literature thus far. The exploration of the applicability
ofthisclass of catalyst inotherreactions isunderwayinour
laboratory.
Table 2. Evaluation of Substrate Scope
a Isolated yield. b Determined by CSP-HPLC. c Reaction at -25 °C.
Acknowledgment. This material was based on work
funded by The Irish Research Foundation for Science
Engineering and Technology (IRCSET).
resulted in the formation of (R)-7 with higher ee (entries
20-22). Combining these findings allowed the generation of
the product in good yield and 90% enantiomeric excess
(entry 23).22
Supporting Information Available. General experimen-
tal procedures, synthesis of catalyst 5h, 1H and 13C NMR
spectra, characterization data, and HPLC assays. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(22) No conversion was observed when acetophenone derivatives
devoid of the -CF3 substituent were utilized in these reactions.
Org. Lett., Vol. 13, No. 6, 2011
1301