Improved asymmetric SNAr reaction of β-dicarbonyl compounds catalyzed by quaternary ammonium salts derived from cinchona alkaloids

Article abstract of DOI:10.1021/jo060627l

The scope and limitation of the asymmetric nucleophilic aromatic substitution reaction of α-substituted 1,3-dicarbonyl compounds and activated aromatic systems catalyzed by N-benzyl-O-benzoylcinchoninium or cinchonidinium salts are presented. Several nove

Full text of DOI:10.1021/jo060627l

Improved Asymmetric SNAr Reaction of â-Dicarbonyl Compounds
Catalyzed by Quaternary Ammonium Salts Derived from Cinchona
Alkaloids
Sara Kobbelgaard, Marco Bella, and Karl Anker Jørgensen*
Danish National Research Foundation: Center for Catalysis, Department of Chemistry, Aarhus UniVersity,
DK-8000 Aarhus C, Denmark
kaj@chem.au.dk
ReceiVed March 22, 2006
The scope and limitation of the asymmetric nucleophilic aromatic substitution reaction of R-substituted
1,3-dicarbonyl compounds and activated aromatic systems catalyzed by N-benzyl-O-benzoylcinchoninium
or cinchonidinium salts are presented. Several novel O-benzoylcinchona alkaloid derived salts have been
prepared and evaluated as catalysts in this reaction, which can proceed with enantioselectivites up to
96% ee. Various 1,3-dicarbonyl compounds and activated aromatic systems are evaluated for the aromatic
nucleophilic substitution reaction, and it has been found that the yield and enantioselectivity are very
dependent on the substrate and reagent. The scope of the functionalization of the products to, e.g., spiro-
oxoindole, a ring-opening reaction of 1,3 R,R-disubstituted dicarbonyl compounds with several
nucleophiles, and the diastereoselective reduction of the keto functionality in the optically active SNAr
product are reported.
Introduction
which the stereochemistry of the newly formed chiral center
can be controlled.3 In a recent paper,4 we reported the first
organocatalytic SNAr reaction of â-ketoesters with activated
aromatic compounds to generate chiral quaternary stereocenters5
using a quaternary ammonium salt derived from cinchona
A fundamental reaction in organic chemistry is the nucleo-
philic aromatic substitution reaction (SNAr) which has been
known for more than 100 years.1 The SNAr reaction normally
requires aromatic compounds having electron-withdrawing
substituents in order to activate the “electron-rich” aromatic ring
for the nucleophilic attack. A variety of different nucleophiles
can be used, including carbon-, oxygen-, and amine-type
nucleophiles.2
(2) See, e.g.: (a) Selvakumar, N.; Yadi Reddy, B.; Sunil Kumar, G.;
Iqbal J. Tetrahedron Lett. 2001, 42, 8398. (b) Snow, R. J.; Butz, T.;
Hammach, A.; Kapadia, S.; Morowick, T. M.; Prokopowicz, A. S.;
Takahashi, H.; Tan, J. D.; Tschantz, M. A.; Wang, X.-J. Tetrahedron 2002,
43, 7553. (c) Lawrence, N. J.; Davies, C. A.; Gray, M. Org. Lett. 2004, 26,
4957.
(3) (a) Nicolaou, K. C.; Li, H.; Boddy, C. N. C.; Ramanjulu, J. M.; Yue
T. Y.; Natarajan, S.; Chu, X. J.; Bra¨se, S.; Ru¨bsam, F. Chem. Eur. J. 1999,
5, 2584. (b) Snyder, S. E.; Shvets, A. B.; Pirkle, W. H. HelV. Chim. Acta
2002, 85, 3605. (c) Islas-Gonzales, G.; Bois-Choussy, M.; Zhu, J. Org.
Biomol. Chem. 2003, 1, 30.
A challenge for the SNAr reaction is to control the stereo-
chemistry of the formed chiral center when carbon nucleophiles
are used, and there are only a limited number of examples in
(1) See, e.g.: (a) Smith, M. B.; March, J. AdVanced Organic Chemistry,
5th ed.; Wiley-Interscience: New York, 2001; Chapter 13, p 850. (b) Buncel,
E.; Dust, J. M.; Terrier, F. Chem. ReV. 1995, 95, 2261.
(4) Bella, M.; Kobbelgaard, S.; Jørgensen, K. A. J. Am. Chem. Soc. 2005,
127, 3670.
10.1021/jo060627l CCC: $33.50 © 2006 American Chemical Society
Published on Web 06/01/2006
4980
J. Org. Chem. 2006, 71, 4980-4987

Organocatalytic Asymmetric SNAr Reactions
FIGURE 1. Concept of the organocatalytic direct SNAr reaction of â-keto ester showing the formation of a chiral ion pair by the interaction of a
cinchona alkaloid quaternary ammonium salt and the â-keto ester generating a chiral nucleophile in a “chiral pocket”.
alkaloids as the catalyst.6 In the present paper, we present the
scope and limitation of this novel asymmetric reaction, together
with various transformations of the optically active products
such as the synthesis of optically active spiro[pyrrolidone-3,3′-
oxoindoles], diastereoselective reductions, and an unusual
nucleophilic ring opening reaction of the products.
The organocatalytic direct SNAr reaction of â-ketoesters is
based on the concept outlined in Figure 1 with a cinchona
alkaloid quaternary ammonium salt as the chiral catalyst.7 A
base removes the acidic proton in the â-ketoester generating
FIGURE 2. Influence of the substituents R′, R′′, and R′′′ on the
an ambident nucleophile, which interacts with the chiral
quaternary ammonium salt forming a chiral ion pair. The idea
behind this chiral ion-pair formation is that the chiral cinchona
alkaloid salt and the â-ketoester generate a nucleophile in a
“chiral pocket” (Figure 1) in which one face is shielded by the
chiral cinchona alkaloid salt leading to an enantioselective
nucleophilic approach to the aromatic compound.
reaction course of the organocatalytic enantioselective SNAr reaction.
biphasic system consisting of toluene as the solvent and CsOH-
monohydrate as the base in the presence of an extensive number
of cinchona alkaloid quaternary ammonium salts 4 having a
diversity in substitution patterns (eq 1).8,9 The results from the
screening process are presented in Table 1.
It has been found that the regio- and stereoselectivity are very
dependent on the substituents R′ and R′′. For R′ ) Bn or 4-CF3-
Bn and R′′ ) allyl of the catalysts 4a and 4b, respectively, the
ambident nature of the nucleophile determines the C- vs
O-arylation selectivity, and mixtures of the C- and O-arylated
products are obtained in ratios of 1.5:1 to 1.0:1 and with very
low enantiomeric excess (entries 1 and 2). For the cinchona
alkaloid catalyst 4c having R′ ) R′′ ) Bn, a 1.0:1 ratio between
the C- and O-arylated products is also obtained and the desired
product 3a is formed as a racemate (entry 3). The functional-
ization of the catalyst turned out to solve both the regio- and
enantioselectivity problem of the SNAr reaction; an exchange
of the R′′-substituent from benzyl to benzoyl (catalyst 4d) leads
to a dramatic change in both regio- and enantioselectivity; at
room temperature, the desired product 3a is now formed as the
major product (C-/O-arylating ratio 4:1) and with an enantio-
meric excess of 46% ee (entry 4). Lowering the reaction
temperature to -40 °C leads to a significant improvement to
>20:1 in favor of 3a and now with 87% ee in toluene as the
solvent (entry 5). Although we lack a rigorous model to explain
the increase in enantiomeric excess, we can hypothesize that a
coordinating group near the 9-position (the R′′-position) is
Results and Discussion
Due to the ambident nature of the nucleophile formed in the
reaction, both a carbon and oxygen nucleophile can perform
the SNAr reaction, and the reaction course is very dependent
on the R′, R′′, and R′′′ substituents as outlined in Figure 2.
Our starting point is the reaction of 2-carbethoxycyclopen-
tanone 1a with 2,4-dinitrofluorobenzene (2,4-DNF) 2a in a
(5) For synthesis of asymmetric quaternary stereocenters starting from
R-dicarbonyl compounds, see, e.g.: Conjugate addition: (a) Hamashima,
Y.; Hotta, D.; Sodeoka, M. J. Am. Chem. Soc. 2002, 124, 11240. (b) Harada,
S.; Kumagai, N.; Kinoshita, T.; Matsunaga, S.; Shibasaki, M. J. Am. Chem.
Soc. 2003, 125, 2582. (c) Fanghui, W.; Hongming, L.; Hong, L.; Deng, L.
Angew. Chem., Int. Ed. 2006, 45, 947. Conjugate addition to alkynones:
(d) Bella, M.; Jørgensen, K. A. J. Am. Chem. Soc. 2004, 126, 5672.
Amination: (e) Saaby, S.; Bella, M.; Jørgensen, K. A. J. Am. Chem. Soc.
2004, 126, 8120. Alkylation: (f) Park, E. J.; Kim, M. H.; Kim, D. Y. J.
Org. Chem. 2004, 69, 6897. (g) Ooi, T.; Miki, T.; Taniguchi, M.; Shiraishi,
M.; Takeuchi, M.; Maruoka K. Angew. Chem., Int. Ed. 2003, 42, 3796.
Halogenation: (h) Bartoli, G.; Bosco, M.; Carlone, A.; Locatelli, M.;
Melchiorre, P.; Sambri, L. Angew. Chem., Int. Ed. 2005, 44, 6219. (i)
Marigo, M.; Kumaragurubaran, N.; Jørgensen, K. A. Chem. Eur. J. 2004,
10, 2133. Fluorination: (j) Shibata, N.; Suzuki, E.; Asahi, T.; Shiro, M. J.
Am. Chem. Soc. 2001, 123, 7001. Pd-mediated arylation: (k) Ahman, J.;
Wolfe, J. P.; Troutman, M. V.; Palucki, M.; Buchwald, S. L. J. Am. Chem.
Soc. 1998, 120, 1918. (l) Hamada, T.; Chieffi, A.; A° hman, J.; Buchwald,
S. L. J. Am. Chem. Soc. 2002, 124, 1261. (m) Spielvogel, D. J.; Buchwald,
S. L. J. Am. Chem. Soc. 2002, 124, 3500.
(8) (a) O’Donnel, M. J.; Wu, S.; Huffman, J. C. Tetrahedron 1994, 50,
4507. (b) Corey, E. J.; Bo, Y.; Bush-Pedersen, J. J. Am. Chem. Soc. 1998,
120, 13000. (c) Dolling, U.-H.; Davis, P.; Grabowski, E. J. J. J. Am. Chem.
Soc. 1984, 106, 6; 446. For the modifications of cinchona alkaloid-derived
PTC catalysts, see: Kacprzak, K.; Gawronski, J. Synthesis 2001, 961.
(9) For the use of benzoylated quinine derivatives in asymmetric
synthesis, see, e.g.: France S.; Shah, M. H.; Weatherwax, A.; Wack, H.;
Roth, J. P.; Lectka, T. J. Am. Chem. Soc. 2005, 127, 1206 and references
therein.
(6) See, e.g.: (a) Halpern, M. E. Phase-Transfer Catalysis. Mechanism
and Synthesis; American Chemical Society: Washington, DC, 1997. (b)
Sasson Y.; Neumann, R. Handbook of Phase-Transfer Catalysis; Blackie
A&M: London, 1997.
(7) (a) Ooi, T.; Maruoka, K. Chem. ReV. 2003, 103, 3013. (b) Lygo, B.;
Andrews, B. I. Acc. Chem. Res. 2004, 37, 518. (c) O’Donnell, M. J. Acc.
Chem. Res. 2004, 37, 506.
J. Org. Chem, Vol. 71, No. 13, 2006 4981

Kobbelgaard et al.
TABLE 1. Screening of Catalysts for the Organocatalytic (15 mol %) SNAr Addition of 2-Carbethoxycyclopentanone 1a to 2,4-DNF 2aa
a All reactions were preformed with 0.03 mmol of 4, 0.20 mmol of 1a, and 0.25 mmol of 2a in 4 mL of solvent and 150 mg of CsOH to >90% conversion
(<2 h). b Ratio determinated by 1H NMR spectroscopy. c Enantiomeric excess determinated by HPLC. d N-Trifluorobenzyl-O-allylcinchonium bromide was
used. e Low conversion: less than 20% isolated yield. f Solvent is PhMe/CHCl3 9:1. g KOH is used as base.
4982 J. Org. Chem., Vol. 71, No. 13, 2006

Organocatalytic Asymmetric SNAr Reactions
SCHEME 1. Hydrolysis of the Benzoate Ester to the Commercially Available N-Benzylcinchonidinium Salt, which Consequently
Reacts with 2,4-DNF 2a Present in the Reaction
SCHEME 2. SNAr Addition of 2-Carbethoxycyclopentanone
1a to 2,4-DNF 2a Preformed by the Quasienantiomers 4d
and 4t
important to achieve high enantioselectivities. Catalyst 4c (R′′
) CH2Ph), which differs from catalyst 4d (R′′ ) C(dO)Ph)
for a subtle oxygen atom, gives 3a as a racemate and 1:1 C-/
O-arylation ratio. It is noteworthy that hydrolysis of the benzoate
group is observed with prolonged reaction times (Scheme 1),
giving the cinchonidinium salt having a free hydroxyl group,
which under the reaction conditions applied quickly reacts with
2,4-DNF 2a to form the new catalyst 4s. We have observed
that the N-benzylcinchonidinium salt with the free hydroxyl
group can also catalyze the SNAr reaction presented in eq 1
and using 50 mol % of the catalyst up to 70% ee of 3a was
obtained. Unfortunately, this result is not reproducible; the high
enantioselectivity varies apparently with the rate of the addition
of catalyst. More disappointing, the C- vs O-arylation ratio is
1:9; therefore, even if a reliable protocol for performing this
reaction could have been developed, it would have little
synthetic value considering the low yield of 3a. The phase
transfer catalyst having the 2,4-dinitro phenyl group in the
9-position (catalyst 4s) is also catalyzing the SNAr reaction;
however, with low enantioselectivity, 44% ee, but with some
regioselectivity, in favor of the desired product C-/O-arylation
positions (4p) leads to a reaction with 65% ee (entry 17). An
exchange of the chloro substituent in the meta-position (4n)
with a nitro substituentscatalyst 4psleads to a significant
lowering in the enantiomeric excess of 3a to only 15% ee (entry
ratio 8:1 at -40 °C. The success of this SNAr reaction is thus
18). Placing other functionalities in the 9-O-position of the
related to avoiding the hydrolysis of the ester group in the
catalyst, such as a sulfone (4r), provided catalysts which are
much less effective, compared to those with ester functionalities
(entry 19) under the reaction conditions tested.
catalyst.
With the successful SNAr reaction using 4d as the catalyst,
we decided to test a number of related catalysts by varying the
R′ and ester group R′′ (Table 1, entries 6-19). With catalyst
4f, lower enantiomeric excess is obtained compared to the
catalyst having the hydrogen atom in the R′′′-position (entry
7).
Virtually all of the catalysts (4a,c-r) we have presented so
far are based on the cinchonidinium salt. The opposite enantio-
mer of the product is obtained when the quasienantiomer 4t is
applied as the catalyst (Scheme 2). This catalyst (4t) affords
up to 45% ee of 3a in the SNAr reaction of 2-carbethoxycy-
The enantioselectivity of the SNAr reaction is eroded when
clopentanone 1a and 2,4-DNF 2a, which is a lower enantio-
the R′ benzyl group is exchanged with a methyl group (catalyst
selectivity compared to the cinchonidinium salt 4d (64% ee).
4e); however, the regioselectivity is maintained to some extent
The solvent is of importance for the organocatalytic direct
SNAr reaction, and we have previously used a 1:9 mixture of
CHCl3 and toluene as the reaction solvent, achieving up to 64%
ee of 3a at -20 °C. With prolonged reaction times, precipitation
of catalyst 4d is observed, and moreover some hydrolysis of
the benzoate group in the 9-position was detected. After
extensive solvents and base screening, we have found that a
mixture of CH2Cl2/PhMe 1:4, solid KOH as the base, and
performing the reaction at -40 °C were the optimal reaction
conditions (Table 2). In this case, little or no hydrolysis of the
catalyst can be detected. Furthermore, with this new reaction
protocol, virtually all of the catalyst is soluble, which might
account for the increased enantioselectivity and Table 2 shows
the results for the asymmetric SNAr reaction using various
catalysts 4d,i,k-m,r. The effect of changing the inorganic base
with the C-arylated product 3a as the major compound 9:1 C-/
O-arylation ratio (entry 6). Further investigations revealed that
the naphthyl or biphenyl catalysts, 4g and 4h, respectively, gave
a slight increase in enantiomeric excess of 3a to 65% ee, while
maintaining the high C- to O-arylation selectivity (entries 8 and
9). Under the present reaction conditions, substitution in the
para position with alkyl groups has a large influence on the
enantioselectivity of the reaction: Me (4i), 67% ee; Et (4j),
61% ee; t-Bu (4k), 13% ee; and OMe (4l), 58% ee (entries 10-
13). The enantioselectivity is also very dependent on the
substitution position in the phenyl ring, moving the chloro
substituent from the ortho- (4m) to the meta- (4n) and to the
para-position (4o) gives the following results: 64% ee, 17%
ee, and 58% ee, respectively (entries 14-16). It is remarkable
though that having two chloro substituents in the two meta-
J. Org. Chem, Vol. 71, No. 13, 2006 4983

Kobbelgaard et al.
TABLE 2. Screening of Conditions and Catalysts (15 mol %) in the SNAr Reaction of 2-Carbethoxycyclopentanone 1a to 2,4-DNF 2aa
a All reactions were performed with 0.03 mmol of 4, 0.20 mmol of 1a, and 0.25 mmol of 2a in 4 mL of solvent and 150 mg of solid base to >90%
conversion (<2 h). b Enantiomeric excess determined by HPLC.
is very modest, and KOH was preferred because (i) it increased
the enantiomeric excess slightly and (ii) it is less hygroscopic.
The results for the 9-O-benzoate catalysts 4d,i,k-m,r tested
in the SNAr reaction of 2-carbethoxycyclopentanone 1a and 2,4-
DNF 2a in Table 2 show some interesting changes; while several
N-benzyl benzoate catalysts afforded a range of different
enantioselectivities of 3a, varying from 5 to 87% ee, with the
previously reported conditions, every catalyst tested with the
4984 J. Org. Chem., Vol. 71, No. 13, 2006

Organocatalytic Asymmetric SNAr Reactions
TABLE 3. Regio- and Enantioselective SNAr Reaction of Various
TABLE 4. Organocatalyzed Asymmetric C-Selective SNAr
Reaction of â-Ketoester 1a with Different Electrophiles 2a-d
Catalyzed by 4d (15 mol %)a
Nucleophiles 1a-e to 2,4-DNF 2a Catalyzed by 4d (15 mol %)a
Entry
nucleophile
time (h)
yield (%)b
ee (%)c
entry
electrophile
yieldb (%)
eec (%)
1
2
3
4
5
1a
1b
1c
1d
1e
2
24
5
21
5
3a, 78
3b, 93d
3c, 16
3d, 36
3e, 20e
85
85
60
35
38
1
2
3
4
2a
2b
2c
2d
3a, 78
3f, 76
3g, 68
3h, 76
85
90
90
94
a Experimental conditions: 0.05 mmol of 4d, 0.30 mmol of 1a, and 0.38
mmol of 2 in 6 mL of solvent (CH2Cl2/PhMe 1:9) and 225 mg of KOH at
-40 °C. b Isolated yield. c Enantiomeric excess determined by HPLC.
a Experimenal conditions: 0.06 mmol of 4d, 0.40 mmol of 1, and 0.50
mmol of 2a in 8 mL of solvent (CH2Cl2/PhMe 1:9) and 300 mg of KOH
at -40 °C. b Isolated yield. c Enantiomeric excess determined by HPLC.
d Reaction mixture PhMe/CHCl3 9:1. e 16% O-arylated product is also
obtained (5e).
these products. We have previously shown4 that the products
are, e.g., precursors of oxoindoles bearing a chiral quaternary
carbon center a motif present in numerous natural substances.10
The enantioselective approach toward this class of molecules
was presented by the one-pot synthesis (four step reaction:
N-Boc-removal, reduction of two nitro groups and ring closure)
of the spiro[pyrrolidone-3,3′-oxoindole] 6b in 70% yield from
compound 3b (eq 4).
new protocol performed significantly better, within 85-96%
ee (Table 2, entries 1-10), and in one casescatalyst 4rsthe
enantiomeric excess increased from 5 to 92% ee (entries 11
and 12). Previously, the chloride-substituted catalyst gave very
different enantioselectivities depending on the position of the
chloride: the ortho-substituted catalyst 4l gave 64% ee, whereas
the meta-substituted catalyst 4m provided 3a in only 17% ee,
while under the new conditions they afforded 95 and 96% ee,
respectively (entries 7-10).
Expanding the system to other nucleophiles and electrophiles
has been attempted. Various nucleophiles have been tested
(Table 3), and it turns out that the reaction proceeds best for
nucleophiles containing a five-membered ring such as 1a-c
where enantioselectivities in the range of 60-85% ee are
obtained (Table 3, entries 1-3). However, it should be noted
that an exchange of the ester group in 1a with a cyano group
(1c) leads to a significant drop in the yield of the reaction to
only 16%; however, the enantiomeric excess of product 3d was
still satisfactory (60% ee, entry 3). Furthermore, it seems there
is a limit to the size of the nucleophiles in the case of 1d, a
bulky nucleophile, that reduces the enantiomeric excess to only
35% ee (entry 4). Expanding to a six-membered ring (1e), the
enantiomeric excess declines to 38% ee and the reaction is less
selective regarding the C-/O-arylation, as a ratio of 1.3:1 is
found.
Another class of interesting compounds is pyrazoles, and the
â-ketoesters might be ideal candidates for creating chiral
pyrazoles (Scheme 3a).11
SCHEME 3 a
Variation within the electrophiles is also possible as presented
in Table 4, and the reaction proceeds with high stereoselectivity.
It is important to have a strong electron-withdrawing group in
the para-position relative to the substitution site and in all case
this is achieved via a nitro group. In the case where there are
two dissimilar fluorides (2c), the reaction is regioselective and
the reaction proceeds with substituting only the fluoride, para
to the nitro group (Table 4, entry 3). It appears also from the
results in Table 4 that different groups can be attached to the
aromatic ring (2a-d), still maintaining high regioselectivity and
enantioselectivities in the range 85-94% ee.
a The expected reaction of â-ketoesters of the SNAr reaction with
b
hydrazine in a ring-closing fashion creating a pyrazole. The unpredicted
formation of 2-(2,4-dinitrophenyl)-5-hydrazinocarbonylpentanoic acid ethyl
ester 7a.
We expected that the chiral R,R-disubstituted â-ketoesters
should be able to condensate with hydrazine, thereby giving
access to chiral pyrazoles via hydrazone formation and a ring-
closure reaction. When the chiral R,R-disubstituted-â-ketoester
The products formed via SNAr reaction have a variety of
applications, and we examined possible functionalization of
J. Org. Chem, Vol. 71, No. 13, 2006 4985

Kobbelgaard et al.
TABLE 5. Opening of r,r-Disubstituted Aryl â-Ketoesters with a Selection of Nucleophiles
1
a Isolated yield. b Estimated by H NMR analysis of the crude reaction mixture. Isolated yield not determined due to the small scale of the reaction.
3a was treated with hydrazine hydrate complete conversion
occurred and formation of a highly polar product was observed
using TLC; however, this product was not the expected pyrazole,
but a compound (7a) resulting from an unpredicted nucleophilic
ring opening of the cyclopentanone ring in 3a (Scheme 3b).12
This unexpected reaction prompted us to test several nucleo-
philes for this ring opening, and the results are presented in
Table 5. All nucleophiles afforded the open-chain compound
7. The reaction of 3a with hydrazine hydrate (Table 5, entry 1)
or NaOEt (entry 2) afforded the ring-opened product 7a,b in
moderate to good yields within a few hours. Electron-withdraw-
ing groups on the aromatic ring are not essential (entry 3), since
3i also afford the opened product 2-phenyl-5-hydrazinocarbo-
nylpentanoic acid ethyl ester (7c). The R,R-disubstituted â-ke-
toester 3a was additionally reacted with a chiral amine (8c) to
investigate if the reaction revealed any diastereoselectivity (entry
4); this was sadly not the case. Compounds of class 7 are chiral,
but regardless of the grade of enatiopurity of the starting material
they are obtained as racemates. While in the past few years
numerous protocols for the synthesis of chiral nonracemic R,R-
disubstituted â-ketoesters appeared in the literature,5 these
competing ring-opening reactions constitute a limit to the
possible functionalization of the products, without losing the
enantiomeric excess of the chiral compound just formed.
Another interesting functionalization of the products formed
through SNAr reaction would be the â-hydroxy compounds.
Herein, we present a diastereoselective reduction of optically
active R,R-disubstituted aryl-â-ketoester 3a to R-hydroxylester
9a (eq 6). This anti-diastereoselective reduction of 3a was
achieved with borane dimethyl sulfide complex (BH3-DMS)
in THF to afford a single diastereoisomer 9a, albeit in low yields.
chona alkaloid salts. Good enantiomeric excess can be obtained
if a catalyst capable of two-site binding is employed; however,
so far a catalyst able to achieve vast generality of the nucleophile
has not been found. A large number of chiral cinchona alkaloid
salts can catalyze the model reaction with high selectivity, both
in terms of C- vs O-selectivity and enantiomeric excess;
however, the asymmetric SNAr reaction of different nucleophiles
is dependent on the structure of the nucleophile, and it turns
out that the reaction proceeds best for nucleophiles containing
a five-membered ring. Variation of the electrophile is viable,
and high regioselectivity and enantioselectivities of 85-94%
ee are obtained. The products formed via SNAr reaction, reacts
with different nucleophiles in an unexpected nucleophilic ring
opening reaction, and another possible functionalization of these
products is the diastereoselective reduction of an optically active
R,R-disubstituted aryl-â-ketoester to the â-hydroxyester.
Experimental Section
General Procedure for the Enantioselective SNAr Reaction
of Aromatic Fluorides 2a-d with Nucleophile 1a-e Catalyzed
by 4a-t. In a test tube, the catalyst 4 (0.03 mmol, 15 mol %) was
dissolved in CH2Cl2 (0.8 mL), and nucleophile 1a-e (0.20 mmol,
1.00 equiv), aromatic fluoride 2a-d (0.25 mmol, 1.25 equiv), and
PhMe (3.2 mL) were added. Another test tube contained KOH (150
mg). The test tubes were cooled to the indicated temperature, and
KOH was added to the reaction mixture. The test tube was closed
by a rubber septa; no inert atmosphere was used. The reaction
mixture was stirred at the indicated temperature until no starting
nucleophile could be detected by TLC. The crude mixture was
directly purified by FC to afford pure products 3a-h. The
enantiomeric excess was determined by HPLC.
3a: yield 89%; 1H NMR δ (CDCl3) 8.78 (d, 1H, J 2.4 Hz), 8.34
(dd, 1H, J 8.8 Hz, J 2.0 Hz), 7.41 (d, 1H, J 8.8 Hz), 4.12 (q, 1H,
(10) Marti, C.; Carreira, E. M. Eur. J. Org. Chem. 2003, 42, 2209 and
references therein.
(11) See, e.g.: (a) Sircar, I.; Morrison, G. C.; Burke, S. E.; Skeean, R.;
Weishaar, R. E. J. Med. Chem. 1987, 30, 1724. (b) Jenwitheesuk, E.;
Samudrala, R. Bioorg. Med. Chem. Lett. 2003, 13, 3989. (c) Kees, K. L.;
Fitzgerald, J. J., Jr.; Steiner, K. E.; Mattes, J. F.; Mihan, B.; Tosi, T.;
Mondoro, D.; McCaleb, M. L. J. Med. Chem. 1996, 39, 3920 and references
therein.
In conclusion, we have reported in full our investigation on
the asymmetric SNAr reaction catalyzed by benzoylated cin-
(12) For a related reaction, see: Cort, L. A.; Mahesar, M. A. J. Chem.
Soc., Perkin Trans. 1 1979, 2034.
4986 J. Org. Chem., Vol. 71, No. 13, 2006

Organocatalytic Asymmetric SNAr Reactions
J 7.2 Hz), 4.11 (q, 1H, J 7.2 Hz), 3.2-3.1 (m, 1H), 2.8-2.6 (m,
1H), 2.6-2.2 (m, 3H), 2.1-1.9 (m, 1H), 1.14 (t, 3H, J 7.2 Hz);
13C NMR δ (CDCl3) 211.3, 168.1, 149.4, 147.2, 140.5, 131.5, 127.4,
121.5, 65.4, 63.0, 38.9, 37.0, 19.6, 14.1; HRMS calcd for C14H14N2-
d, to the reaction mixture was added dropwise 1 mL of 1 M HCl
and the solution extracted with Et2O. The organic layer was dried
over MgSO4, filtered, and evaporated. The crude reaction mixture
was purified by FC (Et2O/pentane 2:1) to give the optically active
â-hydroxy ester 9a. The enantiomeric excess was determined by
HPLC.
NaO7 345.0699, found 345.0699; [R]rt ) -40 (c 4.05, CDCl3,
D
56% ee) (optical rotation performed from another reaction than
reported in Table 1). The ee was determined by HPLC using
Chiralpak AS column (hexane/i-PrOH 80:20); flow rate 1.0 mL/
min; τmajor ) 12.5 min; τminor ) 15.4 min.
1
9a: yield 32%; H NMR δ (CDCl3) 8.70 (d, 1H, J 2.1 Hz),
8.44 (dd, 1H, J 8.8 Hz, J 1.5 Hz), 8.01 (d, 1H, J 8.8 Hz), 4.55 (t,
1H, J 6.9 Hz), 4.20 (dq, 2H, J 7.1 Hz, J 2.9 Hz), 3.12 (bs, 1H),
2.79 (ddd, 1H, J 13.6 Hz, J 9.9 Hz, J 6.5 Hz), 2.1-2.0 (m, 1H),
2.0-1.8 (m, 3H), 1.7-1.6 (m, 1H), 1.19 (t, 3H, J 7.1 Hz); 13C
NMR (CDCl3) 172.6, 149.1, 146.6, 143.9, 131.0, 126.9, 120.6, 79.7,
62.0, 59.7, 33.5, 32.3, 19.7, 13.9; HRMS calcd C14H16N2NaO7
General Procedure for Opening of r,r-Disubstituted Aryl
â-Ketoesters with Various Nucleophiles to 7a,b,d. The â-ketoester
(50 mg (3a), 0.155 mmol, 1.0 equiv), nucleophile (0.165 mmol,
1.1 equiv), and solvent (2 mL) were combined in a test tube with
a magnetic stir bar and maintained at the indicated temperature
until no â-ketoester could be detected by TLC. The crude mixture
was directly purified by FC to afford pure products 7a,b,d.
347.0855, found 347.0858; [R]rt ) +57 (c 2.47, CH2Cl2, 31%
D
ee). The ee was determined by HPLC using a Chiralpak AS column
(hexane/i-PrOH 80:20); flow rate 1.0 mL/min; τmajor ) 6.9 min;
τminor ) 8.0 min.
1
7a: yield 84%; H NMR δ (CDCl3) 8.72 (d, 1H, J 2.1 Hz),
8.41 (dd, 1H, J 8.6 Hz, J 2.0 Hz), 7.78 (d, 1H, J 8.6 Hz), 7.21 (bs,
1H), 4.23 (t, 1H, J 7.3 Hz), 4.2-4.1 (m, 2H), 3.64 (bs, 2H), 2.3-
2.1 (m, 3H), 2.0-1.9 (m, 1H), 1.8-1.5 (m, 2H), 1.71 (t, 3H, J 7.1
Hz); 13C NMR (CDCl3) 172.8, 171.1, 149.3, 146.7, 139.9, 131.5,
127.1, 120.2, 61.8, 46.3, 33.4, 32.0, 23.1, 13.9; HRMS calcd
C14H18N4NaO7 377.1073, found 377.1068.
Aryl â-Ketoester (3a) to â-Hydroxy Ester (9a). To a solution
of â-ketoester (3a) (113 mg, 0.35 mmol, 1.00 equiv) in THF (5
mL) was added 3 equiv of BH3-DMS complex (0.105 mL, 1.05
mmol, 3.00 equiv). After being stirred at room temperature for 7
Acknowledgment. This work was made possible by a grant
from the Danish National Research Foundation. Thanks are
expressed to a reviewer for drawing our attention to ref 12.
Supporting Information Available: Experimental procedures
and characterization. This material is available free of charge via
the Internet at http://pubs.acs.org.
JO060627L
J. Org. Chem, Vol. 71, No. 13, 2006 4987