Highly enantioselective aza-Baylis-Hillman reactions catalyzed by chiral thiourea derivatives

Article abstract of DOI:10.1002/adsc.200505230

We report the discovery of asymmetric aza-Baylis-Hillman (ABH) reactions of N-p-nitrobenzenesulfonylimines with methyl acrylate catalyzed by chiral thiourea derivatives. A series of aromatic imines was found to undergo coupling with methyl acrylate with u

Full text of DOI:10.1002/adsc.200505230

FULL PAPERS
DOI: 10.1002/adsc.200505230
Highly Enantioselective Aza-Baylis--Hillman Reactions Catalyzed
by Chiral Thiourea Derivatives
Izzat T. Raheem, Eric N. Jacobsen*
Department of Chemistry and Chemical Biology, Harvard University, 12 Oxford Street, Cambridge, MA 02138, USA
Fax: (þ1)-617-496-1880, e-mail: jacobsen@chemistry.harvard.edu
Received: June 2, 2005; Accepted: August 18, 2005
Supporting Information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: We report the discovery of asymmetric aza- vide a mechanistic analysis based on the identity of
Baylis–Hillman (ABH) reactions of N-p-nitrobenz- this intermediate as well as kinetic investigations
enesulfonylimines with methyl acrylate catalyzed by and isotope studies, and propose a rationale for the
chiral thiourea derivatives. A series of aromatic observed limitations in yield. Synthetic applications
imines was found to undergo coupling with methyl of the ABHproducts are also described.
acrylate with unprecedented levels of enantioselectiv-
ity (87–99% ee), albeit only in modest (25–49%) Keywords: asymmetric catalysis; aza-Baylis–Hillman
yields. A DABCO-acrylate-imine adduct was isolated reaction; kinetic isotope effect; organocatalysis; thio-
as a key intermediate in the ABHreaction. We pro-
urea catalyst
Results and Discussion
Introduction
The nucleophile-catalyzed coupling of imines with elec-
tron-deficient olefins, known as the aza-Baylis–Hillman
(ABH) reaction, provides direct access to highly func-
tionalized chiral amines from readily available achiral
Methodology Development and Reaction
Optimization
starting materials (Scheme 1).[1] While important ad- Over the past several years, our group[4] and others[5]
vances have been made recently in asymmetric catalysis have developed chiral thiourea derivatives (e.g., 1–5,
of the ABHreaction with enone substrates (EWG ¼ Figure 1) as an important new class of asymmetric cata-
COR),[2a–e] no general and highly enantioselective var- lysts. Given the demonstrated ability of these com-
iants employing simple acrylate derivatives (EWG¼ pounds to activate imines and imine derivatives toward
CO2R) have been identified,[2f–i] despite the clear utility a variety of enantioselective additions,[4a–g] and the es-
that such processes might hold. In this paper we report tablished utility of weak acids in catalyzing asymmetric
ABHreactions of methyl acrylate and p-nitrobenzene- Baylis–Hillman (BH) reactions,[6,7] we considered their
sulfonyl (nosyl) imines catalyzed by chiral thiourea de- possible application to the ABHreaction. In an early
rivatives, in a method that provides highly enantioen- screening study, ABHadduct 7a was generated from
riched nosyl-protected amines[3] albeit in low-to-moder- methyl acrylate and N-nosylbenzaldimine (6a) in 30%
ate yield. Our investigations into the mechanism of this ee and 60% conversion using the known Strecker thiour-
reaction have led to the isolation of a late-stage inter- ea catalyst 1a[4c] in THF, with Ph2PMe as a nucleophilic
mediate, and provided insight into the basis for the ob- additive [Eq. (1)]. This promising lead result prompted
served trade-off between enantioselectivity and yield.
us to undertake systematic optimization studies.
ð1Þ
Scheme 1. The aza-Baylis–Hillman reaction.
Adv. Synth. Catal. 2005, 347, 1701 – 1708
ꢀ 2005 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
1701

Izzat T. Raheem, Eric N. Jacobsen
FULL PAPERS
Figure 1. Representative thiourea catalysts.
Among the various imine N-protecting groups exam- verse correlation between enantiomeric excess and nu-
ined, the nosyl group was found to be optimal, as Boc-, cleophile loading was seen with Ph2PMe (Table 1, en-
Moc-, phosphonyl-, p-toluenesulfonyl-, and alkyl-pro- tries 6–8), reactions employing DABCO exhibited the
tected imines underwent the ABHreaction with methyl
opposite trend (Table 1, entries 9–11), and addition of
acrylate with only marginal enantioselectivities. Cata- exactly one equivalent of DABCO relative to methyl
lyst design strategies involving covalent tethering of ter- acrylate led to highest enantioselectivity (Table 1, en-
tiary amine or phosphine units to the chiral thiourea tries 11–13). An extensive screen of thiourea catalysts
framework were all unsuccessful, with greatly diminish- led to the identification of 1a and 1c as optimal (en-
ed catalytic activity observed. However, markedly high- tries 11 and 12), with 1c selected for further study due
er reactivity and enantioselectivities were obtained by to its greater synthetic accessibility.[8] Pronounced sol-
careful selection of reaction solvent and nucleophilic ad- vent and concentration effects were observed, with reac-
ditive. Improved enantioselectivities were obtained by tions carried out in non-polar solvents (toluene, Et2O,
replacing Ph2PMe with 1,4-diazabicyclo[2.2.2]octane xylenes) affording highest ees, but producing highly het-
(DABCO) as the nucleophilic additive. Whereas an in- erogeneous mixtures and generally low yields of 7a.
Table 1. Effect of solvent and nucleophilic additive in the thiourea-catalyzed ABHreaction [Eq. (1)].
Entry[a]
Catalyst
Solvent
Additive (equivs.)[b]
Conv. [%][c]
ee [%][d]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1c
1c[e]
1c
1c
1c
1c
THF
THF
THF
EtOAc
Toluene
Et2O
Et2O
Et2O
Et2O
Et2O
Et2O
Et2O
Xylenes (0.15 M)
Xylenes (0.03 M)
Toluene
EtOAc
CH2Cl2
PPh3 (0.2)
PMe3 (0.2)
0
–
–
Decomp.
80
86
90
41
62
79
16
30
61
53
55
90
57
65
95
Ph2PMe (0.2)
Ph2PMe (0.2)
Ph2PMe (0.2)
Ph2PMe (0.2)
Ph2PMe (0.5)
Ph2PMe (1.0)
DABCO (0.2)
DABCO (0.5)
DABCO (1.0)
DABCO (1.0)
DABCO (1.0)
DABCO (1.0)
DABCO (1.0)
DABCO (1.0)
DABCO (1.0)
28
17
41
60
52
40
76
87
93
94
97
53
90
84
12
[a]
Unless noted otherwise, reactions were carried out on 0.05 mmol scale with nucleophile, imine (1 equiv.), acrylate
(4 equivs.), and catalyst (20 mol %) in freshly distilled solvent for 24–36 h.
See Supporting Information for a listing of other additives examined.
Determined by GC analysis.
Determined by HPLC analysis using commercial chiral columns.
10 mol% catalyst.
[b]
[c]
[d]
[e]
1702
asc.wiley-vch.de
ꢀ 2005 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Adv. Synth. Catal. 2005, 347, 1701 – 1708

Aza-Baylis–Hillman Reactions Catalyzed by Chiral Thiourea Derivatives
FULL PAPERS
Table 2. Thiourea-catalyzed aza-Baylis–Hillman reaction.
Entry
Ar
Time [h]
Product
Yield [%]
ee [%][a]
1
2
3
4
5
6
7
8
9
C6H5 (6a)
36
24
24
16
16
16
24
36
36
7a
7b
7c
7d
7e
7f
7g
7h
7i
49
40
42
36
33
39
27
30
25
95
93
96
87
94
92
91
99
98
3-CH3C6H4 (6b)
3-(OCH3)C6H4 (6c)
4-ClC6H4 (6d)
3-ClC6H4 (6e)
3-BrC6H4 (6f)
1-Naphthyl (6g)
2-Thiophenyl (6h)
3-Furyl (6i)
[a]
Determined by HPLC analysis using commercial chiral columns (see Supporting Information).
More polar solvents (CH2Cl2, EtOAc) or more dilute
Under the optimized reaction conditions,[9] a variety
conditions with xylenes led to a greater degree of solubi- of aromatic nosylimine derivatives were found to be
lization and improved yields (entries 14, 16 and 17), but suitable coupling partners with methyl acrylate (Ta-
at the expense of enantioselectivity. The pronounced in- ble 2). Electron-donating, electron-withdrawing, and
verse relationship between ee and conversion indicated heteroaromatic substrates underwent reaction with
that a deeper mechanistic investigation was warranted high enantioselectivities in all cases, but with isolated
(vide infra).
yields ranging only from 25 to 49%. The synthetic utility
Scheme 2. Some representative synthetic transformations of aza-Baylis–Hillman adducts. Conditions: (a) Cs2CO3, ArCH2Br,
DMF; (b) Pfaltz catalyst,[10] H2, CH2Cl2, 1 atm; (c) OsO4 (5 mol %), NMO, H2O/acetone; (d) tributylvinyltin, Pd(OAc)2, S-
Phos,[11] DMF; (e) chlorobenzaldoxime, Et3N , CH2Cl2; (f) t-BuOOH, Triton B, THF; (g) See Supporting Information; (h)
20% aqueous HCl, reflux. See Supporting Information for full details.
Adv. Synth. Catal. 2005, 347, 1701 – 1708
ꢀ 2005 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
asc.wiley-vch.de
1703

Izzat T. Raheem, Eric N. Jacobsen
FULL PAPERS
of the densely functionalized ABHadducts is illustrated
in the variety of transformations shown in Scheme 2.
tion of 6a and methyl acrylate catalyzed by 1c and DAB-
CO was chosen as a representative system. The physical
properties of the reaction mixtures were observed to
change within a few hours of initiation, with the system
becoming more viscous and the appearance of a yellow
precipitate. Initially, we suspected that the precipitate
corresponded to undissolved imine and product, both
Isolation of Reaction Intermediate and Mechanistic
Investigation
The Baylis–Hillman (BH) reaction and the analogous of which display low solubility in xylenes. Instead, ESI
ABHreaction have been the focus of considerable
mass spectralanalysis of the precipitate isolated from re-
mechanistic analysis over the past several years,[1,12] actions of 6a revealed a prominent peak with an m/z val-
and compelling evidence has been gathered very recent- ue of 489.2, consistent with a structure corresponding to
ly indicating that the deprotonation step in BHreactions
zwitterionic 9a (Scheme 4), corresponding to intermedi-
(I’’ to I’’’ in Scheme 3) can be rate-limiting.[13] With this ate I’’ in Scheme 2. Further, addition of excess 4 N HCl
framework in mind, we sought to glean insight into the to the crude reaction mixture, followed by aqueous ex-
mechanism of the thiourea-catalyzed ABHreaction,
traction, afforded a glassy, off-white solid isolated
with the practical goal of elucidating the basis for the in- from the aqueous layer that was characterized by ESI-
verse relationship between ee and conversion. The reac- MS, 1HNMR, 13C NMR, and IR as the dihydrochloride
salt 10a.[14] The analogous salts 10f and 10i were obtained
in a similar manner from 3-bromobenzaldimine 6f and
furfuraldimine 6i, respectively. In every case, 10 was iso-
lated in high diastereomericpurity, with the relative ster-
eochemistry of the major isomer assigned as anti by
comparison of 1HNMR spectral properties to those of
closely related literature compounds.[15] This represents,
to our knowledge, the first isolation and characteriza-
tion of such intermediates in BHor ABHreactions.
The dihydrochloride salt 10a displayed a high level of
stability, with no detectable decomposition after storage
for over 5 months at room temperature. However, upon
addition of two equivalents of DBU to a solution of 10a
in CD3OD or DMSO-d6, the resonance due to the sulfon-
ꢀ
amide N Hproton disappeared within 4 min and the
benzylic C Hproton resonance a to the sulfonamide
ꢀ
collapsed from a doublet of doublets to a simple doublet.
Furthermore, the solution acquired the characteristic
bright yellow color attributable to a sulfonamide anion.
We therefore assign this species as the zwitterionic inter-
mediate (9a). Spectra of 9a generated in this manner
were always accompanied by resonances due to methyl
acrylate and imine in varying equilibrium concentra-
Scheme 3. Generally accepted mechanism of the Baylis–Hill-
man and aza-Baylis–Hillman reactions (Y¼alkyl or alkoxy,
X¼O or NR’’).
Scheme 4. Preparation of 10a.
1704
asc.wiley-vch.de
ꢀ 2005 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Adv. Synth. Catal. 2005, 347, 1701 – 1708

Aza-Baylis–Hillman Reactions Catalyzed by Chiral Thiourea Derivatives
FULL PAPERS
The heterogeneous nature of the ABHreaction cata-
lyzed by 1c rendered kinetic studies difficult; however,
the uncatalyzed reaction of methyl acrylate and 6a pro-
moted by DABCO proceeded homogeneously in CHCl3
and proved amenable to kinetic analysis. The reaction
was monitored by GC analysis, and was found to display
a first-order kinetic dependence on both DABCO and
methyl acrylate (Figure 2). In contrast to the BHreac-
tion,[13a] rate saturation with respect to the imine electro-
phile was observed.
Kinetic isotope effect studies were also carried out
based on initial rate measurements, by comparison of re-
actions of methyl acrylate with separate reactions of a-
deuterio-methyl acrylate. A prominent primary kinetic
isotope effect was observed (kH/kD ¼3.81), strongly sug-
gesting that deprotonation of the a-H(D) (I’’ to I’’’ in
Scheme 2) is rate-limiting.
The DBU-mediated elimination of 10a in MeOHwas
monitored by ReactIRꢁ and was found to obey a first-or-
der reaction rate profile over>4 half lives. This is consis-
tent with rapid and irreversible deprotonation of 10a to
9a, and intramolecular proton transfer of 9a to the corre-
sponding enolate 11a (Scheme 5). Added imine had no
effect on the rate of elimination, indicating that, in con-
trast to BHreactions, [13a] the electrophile does not play a
role in mediating the deprotonation step.
The mechanism outlined in Scheme 6 is consistent
with the experimental observations summarized above.
Zwitterionic species 9 exists as both syn and anti diaster-
eomers, but the anti diastereomer may undergo precip-
itation selectively. We propose that under the conditions
of the ABHreaction both diastereomers of 9 are
formed, but that 9syn is generated in high ee and decom-
poses relatively rapidly by intramolecular proton trans-
fer/elimination to generate 7 in high ee. In contrast, 9anti
may undergo relatively slow elimination to 7 due to less
favorable steric interactions in the requisite eclipsed (or
nearly eclipsed) conformation, and therefore its concen-
tration builds up during the course of the reaction lead-
Figure 2. Kinetic data and the derived rate law of the uncata-
lyzed ABHreaction obtained by measurement of initial rates.
tions, indicating that 9a undergoes reversion to its pre-
cursors. This constitutes a racemization pathway for 9a
in the absence of catalyst. This reversion pathway
proved impossible to suppress despite extensive screens
of solvent, base, and temperature in the elimination re-
action. Compound 9a generated in this manner under-
went clean elimination to provide product 7a, consistent
with the notion that 9a is indeed an intermediate in the
catalytic cycle. However, 7a obtained under these condi-
tions was generated in lower ees than material produced
via the direct ABHreaction.
Scheme 5.
Adv. Synth. Catal. 2005, 347, 1701 – 1708
ꢀ 2005 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
asc.wiley-vch.de
1705

Izzat T. Raheem, Eric N. Jacobsen
FULL PAPERS
Scheme 6. Proposed pathway for the enantioselective aza-Baylis–Hillman reaction of nosylimines. The unfavorable steric in-
teractions in 9anti presumably responsible for slow elimination of this intermediate to 7 are highlighted.
ing to precipitation. The fact that 7 is produced in low ee thiourea derivatives in asymmetric catalysis are under-
by base-induced elimination from 10a suggests that 9anti way.
may be generated in low ee. This would serve to explain
why solvent systems that effectively solubilize both dia-
stereomers of 9 lead to formation of 7 in depressed ee.
Experimental Section
Unfortunately, all efforts toward direct determination
of the ee of 9a, 10a or any of their analogues were unsuc-
cessful, so we have not secured definitive support for this
hypothesis.
General Procedure for Preparation of Nosylimines
N-Benzylidene-4-nitro-benzenesulfonamide (6a): To a flame-
dried round-bottom flask containing activated 4 Š MS, Am-
berlyst 15 (cat.), and p-nitrobenzenesulfonamide (1 equiv.)
were added freshly distilled toluene (0.25 M) and benzalde-
hyde (1.1 equivs.). A Dean–Stark trap equipped with reflux
condenser was attached, all joints sealed with teflon tape,
and the reaction mixture heated to a vigorous reflux for 20 h.
The mixture was allowed to cool to room temperature, and
Conclusion
We have developed a highly enantioselective catalytic
aza-Baylis–Hillman reaction of nosylimines with methyl
acrylate. While product yields are very moderate
(<50%), the enantioselectivities for a variety of aromat- was then filtered through a pad of Celite. The pad was washed
with toluene, and the filtrate concentrated under vacuum to a
solid. The solids were washed with 3ꢁhexanes, and the solid
collected on a sintered glass funnel. This material was suffi-
ciently pure to use in the ABHreactions, but could be further
recrystallized in high yield from EtOAc/hexanes. The product
was isolated as a light tan powder. FT-IR (CH2Cl2 thin film):
n¼1606 (s), 1569 (s), 1528 (s), 1450 (w), 1350 (m), 1333 (w),
1310 (m), 1224 (w), 1161 (s), 1088 (m), 856 (m), 796 (s),
ic imines are high and unprecedented for ABHreactions
with acrylate derivatives. Highest ees are obtained un-
der conditions where a zwitterionic intermediate 9anti
undergoes precipitation from the reaction mixture, pre-
sumablyas a result of its slow rate of elimination to prod-
uct. Thesestudies suggest thatbothhighenantio-and dia-
stereoselectivity in the addition to the imine-catalyst
complex may be crucial for obtaining high ees and yields
in the overall reaction. Efforts directed toward this goal,
1
739 cmꢀ1 (m); HNMR (500 MHz, CDCl 3): d¼9.13 (s, 1H),
8.39 (d, J¼8.8 Hz, 2H), 8.21 (d, J¼8.8 Hz, 2H), 7.96 (d, J¼
¼
and toward the further exploration of the utility of chiral 7.2 Hz, 2H), 7.67 (t, J¼7.5 Hz, 1H), 7.53 (t, J 7 .6 Hz, 2H);
1706
asc.wiley-vch.de
ꢀ 2005 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Adv. Synth. Catal. 2005, 347, 1701 – 1708

Aza-Baylis–Hillman Reactions Catalyzed by Chiral Thiourea Derivatives
FULL PAPERS
13C NMR (100 MHz, CDCl3): d¼172.62, 150.84, 144.43,
136.06, 131.96, 129.97, 129.62, 129.41, 124.59; LR-MS (ApCI):
m/z¼291.0 (5%) [MþH]þ, 261.0 (100%) [MꢀNOþ]þ.
(d, J¼10.3 Hz 1H), 8.07 (d, J¼8.3 Hz, 2H), 7.79 (d, J¼
8.3 Hz, 2H), 7.18 (d, J¼4.9 Hz, 2H), 6.95–7.10 (m, 3H), 4.93
(dd, J¼6.0, 10.5 Hz, 1H), 4.16 (d, J¼13.2 Hz, 1H), 3.72–3.91
(m, 7H), 3.70 (s, 3H), 3.42–3.63 (m, 6H), 2.90 (m, 1H); 13C
NMR (100 MHz, DMSO-d6): d¼170.97, 149.69, 146.80,
136.04, 128.93, 128.80, 128.62, 127.95, 124.56, 61.69, 59.24,
53.62, 51.10, 46.24, 43.45; LR-MS (ESI): m/z¼489.2 (100%)
General Procedure for the Asymmetric Aza-Baylis–
Hillman Reaction
þ
[Mꢀ2 H Cl] .
2-[(4-Nitrobenzenesulfonylamino)-(S)-phenylmethyl]-
acrylic acid methyl ester (7a): An oven-dried vial was charged
Base-Mediated Elimination of 10a
with
N-benzylidene-4-nitro-benzenesulfonamide
(6a,
1 equiv.), catalyst 1c (10 mol %), DABCO (1 equiv.), and acti-
vated 3 Š MS. The vial was evacuated and purged with N2. Pre-
cooled, freshly distilled, anhydrous xylenes (0.15 M) and meth-
yl acrylate (8 equivs.) were added via syringe at 48C, and the
mixture stirred for 36 h. The mixture was diluted with anhy-
drous MeOHand then quenched immediately with 4 N HCl
in dioxane. The crude adduct was purified by flash chromatog-
raphy (100% CH2Cl2) to afford the pure aza-Baylis–Hillman
adduct as a white solid. The ee of this material was determined
to be 95% [ChiralPak AS, 1.0 mL/min, 280 nm, 40% IPA/hex-
anes, tr(major)¼11.566 min, tr(minor)¼17.618 min]; [a]2D3:
þ27.78 (c 2, EtOH); FT-IR (CH2Cl2 thin film): n¼3293 (br),
3108 (w), 3068 (w), 1720 (s), 1632 (w), 1607 (w), 1531 (s),
1440 (m), 1350 (s), 1312 (m), 1166 (s), 1091 (m), 1062 (m),
2-[(4-Nitrobenzenesulfonylamino)-(S)-phenylmethyl]-
acrylic acid methyl ester (7a, from10a): A flame-dried, 5-mL
round-bottom flask was charged with dihydrochloride salt
(10a from above, 150 mg, 0.267 mmol) which was dissolved in
anhydrous DMSO (2 mL) or MeOH(2 mL) at room tempera-
ture. Freshly distilled DBU (84 mL, 0.560 mmol) was added via
syringe, and the reaction mixture stirred for 16 h. The mixture
was then diluted with 5 mL CH2Cl2, 1 N HCl (5 mL) was added
with vigorous stirring, the layers were allowed to separate, and
the organic layer removed. The aqueous layer was extracted
with 3ꢁ10 mL CH2Cl2, dried over Na2SO4, and concentrated
under vacuum to afford the crude aza-Baylis–Hillman adduct
(7a), which matched the 1HNMR and 13C NMR spectra as pre-
viously reported (vide supra). As discussed, ees of this product
were substantially depressed, and generally obtained in a range
of 30–40%.
1
855 (m), 738 cmꢀ1 (s); HNMR (500 MHz, CDCl 3): d¼8.22
(d, J¼8.8 Hz, 2H), 7.92 (d, J¼8.8, 3 Hz, 2H), 7.14–7.27 (m,
5H), 6.23 (s, 1H), 6.15 (d, J¼9.3 Hz, 1H), 5.82 (s, 1H), 5.40
(d, J¼9.3 Hz, 1H), 3.64 (s, 3H); 13C NMR (100 MHz, CDCl3):
d¼165.96, 150.08, 146.74, 138.49, 138.01, 128.94, 128.50,
128.41, 128.32, 126.64, 124.30, 59.86, 52.50; LR-MS (ApCI):
m/z¼377.0 (2%) [MþH]þ, 347.0 (30%) [MꢀNOþ]þ, 330.0
(32%) [MꢀNO2]þ, 175.0 (100%) [Mꢀsulfonamide]þ.
Acknowledgements
This work was supported by the NIGMS through GM-43214
and P50 GM069721 and by predoctoral fellowship support to
I. T. R. from the NSF. The authors also gratefully acknowledge
the Harvard University Mass Spectrometry facility for assis-
tance in the characterization of 10a.
Isolation of Aza-Baylis–Hillman Intermediate as the
Dihydrochloride Salt (10a)
A flame-dried, 10-mL round-bottom flask was charged with a
stir bar, imine 6a (75 mg, 0.259 mmol), thiourea catalyst 1c
(15.6 mg, 0.026 mmol), and DABCO (29 mg, 0.259 mmol).
The flask was cooled to 48C, and charged with pre-cooled,
freshly distilled xylenes (1.75 mL) and methyl acrylate
(187 mL, 2.07 mmol). The mixture was allowed to stir for 12 h
at 48C, over which time a bright yellow precipitate formed.
Distilled H2O (1.75 mL) was added, and the biphasic mixture
was stirred vigorously for an additional 3–6 h at 48C. HCl
(200 mL, 4 N in dioxane) was added. Additional 5 mL H2O
were added, and the reaction mixture was stirred vigorously
for 5 min. The layers were allowed to separate, and the organic
layer removed. The aqueous layer was extracted with CH2Cl2
(3ꢁ20 mL) and the combined organic layers were back-ex-
tracted with 3ꢁ5 mL H2O. The combined aqueous layers
were concentrated directly under vacuum at 608C. The result-
ing clear oil was further subjected to high vacuum for 24 h to
remove residual H2O, affording the dihydrochloride salt of
the aza-Baylis–Hillman intermediate (10a) as a glassy solid;
yield: 61.2 mg (42%); FTIR (CH2Cl2 thin film): n¼1734 (s),
1630 (w), 1530 (s), 1459 (w), 1438 (w), 1351 (s), 1313 (w),
1166 (s), 1089 (w), 853 (w), 739 cmꢀ1 (m); 1HNMR
(500 MHz, DMSO-d6, 10–16:1 mixture of diastereomers,
with signals corresponding to major form indicated): d¼9.72
References and Notes
[1] For reviews, see: a) D. Basavaiah, A. J. Rao, T. Satyanar-
ayana, Chem. Rev. 2003, 103, 811–892; b) E. Ciganek, in:
Organic Reactions, Vol. 51, (Ed.: L. A. Paquette), Wiley,
New York, 1997, pp. 201–350; c) D. Basavaiah, P. D.
Rao, R. S. Hyma, Tetrahedron 1996, 52, 8001–8062;
b) S. E. Drewes, G. H. P. Roos, Tetrahedron 1988, 44,
4653–4670.
[2] a) K. Matsui, S. Takizawa, H. Sasai, J. Am. Chem. Soc.
2005, 127, 3680–3681; b) M. Shi, L.-H. Chen, C.-Q. Li,
J. Am. Chem. Soc. 2005, 127, 3790–3800; c) M. Shi, L.-
H. Chen, C.-Q. Li, Tetrahedron: Asymmetry 2005, 16,
1385–1391; d) M. Shi, G.-L. Zhao, Adv. Synth. Catal.
2004, 346, 1205–1219; e) M. Shi, Y.-M. Xu, Eur. J. Org.
Chem. 2002, 696–701; f) M. Shi, L.-H. Chen, Chem.
Commun. 2003, 1310; g) M. Shi, Y.-M. Xu, Angew.
Chem. Int. Ed. 2002, 41, 4507–4510; h) D. Balan, H.
Adolfsson, Tetrahedron Lett. 2003, 44, 2521; i) S. Kawa-
hara, A. Nakano, T. Esumi, Y. Iwabuchi, S. Hataketya-
ma, Org. Lett. 2003, 5, 3103.
Adv. Synth. Catal. 2005, 347, 1701 – 1708
ꢀ 2005 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
asc.wiley-vch.de
1707

Izzat T. Raheem, Eric N. Jacobsen
FULL PAPERS
[3] T. Fukuyama, C.-K. Jow, M. Cheung, Tetrahedron Lett.
1995, 36, 6373–6374.
Chen, J. Org. Chem. 2003, 68, 915–919; c) M. Shi, L.-
H. Chen, Chem. Commun. 2003, 1310–1311; d) J. E. Im-
briglio, M. M. Vasbinder, S. J. Miller, Org. Lett. 2003, 5,
3741–3743.
[4] a) M. S. Sigman, E. N. Jacobsen, J. Am. Chem. Soc. 1998,
120, 4901–4902; b) M. S. Sigman, P. Vachal, E. N. Jacob-
sen, Angew. Chem. Int. Ed. 2000, 39, 1279–1281; c) P. Va-
chal, E. N. Jacobsen, J. Am. Chem. Soc. 2002, 124,
10012–10014; d) G. D. Joly, E. N. Jacobsen, G. D. Joly,
J. Am. Chem. Soc. 2004, 126, 4102–4103; e) A. G. Wen-
zel, E. N. Jacobsen, J. Am. Chem. Soc. 2002, 124,
12964–12965; f) T. P. Yoon, E. N. Jacobsen, Angew.
Chem. Int. Ed. 2005, 44, 466–468; g) M. S. Taylor, E. N.
Jacobsen, J. Am. Chem. Soc. 2004, 126, 10558–10559;
h) D. E. Fuerst, E. N. Jacobsen, J. Am. Chem. Soc.
2005, 127, 8964–8965.
[5] a) T. Okino, Y. Hoashi, Y. Takemoto, J. Am. Chem. Soc.
2003, 125, 12672–12673; b) T. Okino, S. Nakamura, T.
Furukawa, T. Takemoto, Org. Lett. 2004, 6, 625–627;
c) A. Berkessel, F. Cleemann, S. Mukherjee, T. N. Mul-
ler, J. Lex. Angew. Chem. Int. Ed. 2005, 44, 807–811;
d) B.-J. Li, L. Jiang, M. Liu, Y.-C. Chen, L.-S. Ding, Y.
Wu, Synlett 2005, 603–606.
[6] a) I. E. Marko, P. R. Giles, N. J. Hindley, Tetrahedron
1997, 53, 1015–1024; b) Y. M. A. Yamada, S. Ikegami,
Tetrahedron Lett. 2000, 41, 2165–2169; c) N. T. McDou-
gal, W. L. Trevellini, S. A. Rodgen, L. T. Kliman, S. E.
Schaus, Adv. Synth. Catal. 2004, 346, 1231–1240.
[7] Examples of asymmetric Baylis–Hillman reactions cata-
lyzed by alternative mechanisms (e. g., chiral nucleo-
philes, Lewis acids): a) Y. Iwabuchi, M. Nakatani, N. Yo-
koyama, S. Hatakeyama, J. Am. Chem. Soc. 1999, 121,
10219–10220; b) K.-S. Yang, W.-D. Lee, J.-F. Pan, K.
[8] Catalyst 1c is prepared in 5 steps and 84% overall yield
from commercially available materials. See Supporting
Information for full experimental details.
[9] A beneficial effect of added molecular sieves was ob-
served, likely due to minimization of imine hydrolysis.
[10] F. Menges, A. Pfaltz, Adv. Synth. Catal. 2002, 344, 40–44.
[11] S. D. Walker, T. E. Barder, J. R. Martinelli, S. L. Buch-
wald, Angew. Chem. Int. Ed. 2004, 43, 1871–1876.
[12] For mechanistic investigations of the Baylis–Hillman re-
action see: a) P. Langer, Angew. Chem. Int. Ed. 2000,
#39#38, 3049–3052; b) J. S. Hill, N. S. Isaacs, J. Phys.
Org. Chem. 1990, 3, 285–288; c) M. L. Bode, P. T.
Kaye, Tetrahedron Lett. 1991, 32, 5611–5614; d) Y.
Fort, M. C. Berthe, P. Caubere, Tetrahedron 1992, 48,
6371–6384.
[13] a) K. E. Price, S. J. Broadwater, H. M. Jung, D. T.
McQuade, Org. Lett. 2005, 7, 147–150; b) V. K. Aggar-
wal, S. Y. Fulford, G. C. Lloyd-Jones, Angew. Chem.
Int. Ed. 2005, 44, 1706–1708.
[14] Reactions carried out in xylenes in the absence of cata-
lyst 1c also led to precipitation of 9a. The diastereomeric
composition of this material, as determined by analysis
of the corresponding salt 10a, was ca. 5:1 favoring the
anti isomer.
[15] E. P. Kundig, L.-H. Xu, P. Romanens, Tetrahedron Lett.
1995, 36, 4047.
1708
asc.wiley-vch.de
ꢀ 2005 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Adv. Synth. Catal. 2005, 347, 1701 – 1708