Enantioselective construction of quaternary stereogenic carbon atoms by the lewis base catalyzed additions of silyl ketene imines to aldehydes

Article abstract of DOI:10.1002/chem.201403342

Silyl ketene imines derived from a variety of α-branched nitriles have been developed as highly useful reagents for the construction of quaternary stereogenic centers via the aldol addition reaction. In the presence of SiCl₄ and the catalytic a

Full text of DOI:10.1002/chem.201403342

DOI: 10.1002/chem.201403342
Full Paper
&
Asymmetric Synthesis
Enantioselective Construction of Quaternary Stereogenic Carbon
Atoms by the Lewis Base Catalyzed Additions of Silyl Ketene
Imines to Aldehydes
Scott E. Denmark,* Tyler W. Wilson, and Matthew T. Burk[a]
Chem. Eur. J. 2014, 20, 9268 – 9279
9268
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Abstract: Silyl ketene imines derived from a variety of a-
branched nitriles have been developed as highly useful re-
agents for the construction of quaternary stereogenic cen-
ters via the aldol addition reaction. In the presence of SiCl4
and the catalytic action of a chiral phosphoramide, silyl
ketene imines undergo extremely rapid and high yielding
addition to a wide variety of aromatic aldehydes with excel-
lent diastereo- and enantioselectivity. Of particular note are
the high yields and selectivities obtained from electron-rich,
electron-poor, and hindered aldehydes. Linear aliphatic alde-
hydes did react with good diastereo- and enantioselectivity
in the presence of nBu4N+IÀ, but branched aldehydes were
much less reactive. Semiempirical calculations provided a ra-
tionalization of the observed diastereo- and enantioselectivi-
ty via open transitions states.
Introduction
key CÀC bond-forming event. Furthermore, achieving high
levels of enantiotopic face selectivity is difficult because of the
relatively similar steric environments presented by the non-hy-
drogen substituents. Although a number of different catalytic,
enantioselective reactions have been reported, very few exhibit
generality over a wide range of architectures. Setting quaterna-
ry stereogenic centers in acyclic systems presents an even
greater challenge, due to the additional degrees of freedom
associated with these motifs.[3] The need for more general
methods to prepare quaternary carbon atoms is underscored
by the large number of biologically active natural products
and pharmaceutical targets that possess quaternary stereogen-
ic carbon atoms.[4]
The aldol reaction is one of the most well-studied, powerful,
and reliable tools a synthetic chemist has for the stereoselec-
tive construction of carbon–carbon bonds. In the classic reac-
tion an aldehyde or ketone is rendered nucleophilic at the a-
carbon by pretreatment with a Brønsted or Lewis base. The re-
sulting activated enolate species then undergoes addition to
a second carbonyl compound giving rise to a b-hydroxy car-
bonyl product. A stereochemical analysis of this process reveals
that when the reacting partners are both prochiral substrates,
four possible stereoisomers can form (Figure 1).
The aldol addition of a,a-disubstituted enolates to alde-
hydes would provide a powerful method for generating qua-
ternary centers; however, this approach is limited by the need
for and inability to prepare geometrically defined a,a-disubsti-
tuted enolate (1) or enolate equivalents (Scheme 1). Therefore,
to successfully utilize the aldol reaction for the asymmetric
synthesis of quaternary centers, one must either address this
deficiency or develop other nucleophile classes.
Figure 1. Stereochemical dyad resulting from aldol reaction of prochiral
enolate and aldehyde.
Indeed, the primary focus over the past 30 years of research
in aldol methodology has been on the development of catalyt-
ic, enantioselective methods for selectively obtaining single
stereoisomers from this tetrad of compounds. Largely these
challenges have been met by very inspired and elegant solu-
tions and the aldol reaction has provided a useful testing
ground for the development of modern asymmetric catalysis.[1]
Despite these successes a remaining obstacle in aldol method-
ology is the synthesis of quaternary stereogenic centers.
Scheme 1. Aldol reaction for the synthesis of quaternary carbon atoms.
Background
Stereoselective synthesis of E- and Z-substituted enolates
The development of catalytic, enantioselective methods for
the construction of quaternary stereogenic centers represents
an ongoing challenge to organic chemists.[2] The difficulty in
forming these centers arises from the high degree of steric re-
pulsion that is encountered in the transition state during the
The ability to selectively control enolate geometry is para-
mount to the success of achieving stereocontrol in the aldol
reaction. The additions proceed through either chair-like (Zim-
merman–Traxler)[5] or open transition structures depending on
the nature of the enolate and the reaction conditions em-
ployed (Figure 2). In both cases, the transition structures are
well ordered and insufficient stereocontrol in the enolate ge-
ometry will be directly reflected in poor anti/syn diastereose-
lectivity in the product. For this reason, much work has been
dedicated to developing reliable and robust methods for ach-
ieving highly selective enolizations. Pioneering studies by Ire-
land and co-workers showed that for monosubstituted ketones
[a] Prof. Dr. S. E. Denmark, Dr. T. W. Wilson, Dr. M. T. Burk
Department of Chemistry, University of Illinois
Urbana, Illinois 61801 (USA)
Fax: (+1)217-333-3984
E-mail: sdenmark@illinois.edu
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201403342.
Chem. Eur. J. 2014, 20, 9268 – 9279
www.chemeurj.org
9269
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Marek and co-workers have de-
veloped alternative strategies for
the synthesis of trisubstituted
enolates that involve syn-carbo-
metalation of oxazolidinone-
modified ynamides (10) with or-
ganocuprates followed by either
Figure 2. Transition structures for aldol addition reactions with various enolates.
homologation with a zinc-carbe-
and esters, the E/Z selectivity can be dramatically influenced
by the choice of base, temperature, solvent, and additives,
such as HMPA.[6] Subsequently, these observations led to the
development of protocols for achieving either E- or Z-mono-
substituted enolates by judicious choice of reaction condi-
tions.[7] Despite this important body of work, very few methods
have been realized for achieving control with trisubstituted
enolates.
noid,[11] or stereoselective oxidation with tert-butyl hydroperox-
ide.[12] The latter method provides a facile route to stereode-
fined trisubstituted copper enolates (12), which can subse-
quently participate in diastereoselective quaternary aldol reac-
tions (Scheme 4). The key to success in this methodology is
the stereoselective oxidation of the vinyl copper species (11),
which was confirmed by trapping experiments with TMSCl and
characterization of the resulting silyl enol ether. Importantly
this methodology has also proved effective in Mannich addi-
tion reactions with N-sulfonyl imines.
Stereoselective synthesis of trisubstituted enolates and their
application in diastereoselective aldol reactions
A simple and effective method for controlling enolate geom-
etry in tetrasubstituted enol borinates was recently described
by Roush and co-workers (Scheme 5).[13] The enol borinates
(18) are generated in situ by the 1,4-hydroboration of a,b-un-
saturated morpholine carboxamides (17) with (diisopinocam-
pheyl)borane ((lIpc)2BH). Subsequent aldol addition reactions of
Gleason and Manthorpe reported an innovative method for
the synthesis of both E- and Z-trisubstituted amide enolates by
the reduction of bicyclic thioglycolate lactams.[8] The design
relies on the rigidity present in the bicyclic lactams, such that
the sulfur atom is constrained to reside on only one face of
the carbonyl plane. Upon two-electron reduction of the disub-
stituted thioglycolate lactam (4) with lithium di-tert-butylbiphe-
nylide (LiDBB), the carbon–sulfur bond is cleaved and an eno-
late dianion is formed. Trapping of this intermediate with two
equivalents of trimethylsilyl chloride (TMSCl) gave the silyl
ketene aminals (5) in good yield and high diastereoselectivity
(Scheme 2). Importantly, either E- or Z-enolates can be made
using this method by simply changing the order in which the
R groups are introduced onto the thioglycolate lactam (3).
Scheme 3. Aldol reaction of chiral, non-racemic bicyclic thiolactams for the
preparation of quaternary stereogenic centers.
Scheme 2. Stereodefined trisubstituted enolates from bicyclic thioglycolate
lactams.
With the ability to selectively access trisubstituted amide
enolates, Gleason then developed a diastereoselective aldol re-
action for the synthesis of quaternary stereogenic centers.[9]
The success of this reaction relies on the ability to convert the
lithium enolate to a boron enolate using dicyclohexylboron
bromide. The in situ formed boron enolate then undergoes
diastereoselective syn aldol additions with aromatic and olefin-
ic aldehydes in good yield and excellent stereoselectivities
(Scheme 3). Gleason and co-workers have also reported an
enantioselective Mannich-type addition with benzenesulfonyl-
imines.[10]
Traditionally, methods for enolate formation have relied on
the deprotonation of preexisting carbon skeletons. Recently,
Scheme 4. Preparation of stereodefined trisubstituted copper enolates and
their application to diastereoselective aldol reactions.
Chem. Eur. J. 2014, 20, 9268 – 9279
www.chemeurj.org
9270
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ing that the amine catalyst is unable to discriminate between
the two possible enamine intermediates. Similar work has also
been described by the Barbas group for the enantioselective
Mannich reactions of protected a-imino ethyl glycolates, cata-
lyzed by l-proline.[17]
A mechanistically distinct, direct aldol reaction for the asym-
metric synthesis of quaternary carbon atoms using catalytic
amounts of an organo–transition-metal complex has been de-
scribed by Ito and co-workers.[18] The aldol addition between
ethyl 2-cyanopropionate and acetaldehyde is catalyzed by sub-
stoichiometric amounts of a rhodium complex generated in
situ, to afford the corresponding a-cyano-b-hydroxy carboxy-
lates (31) in good yield, but low diastereoselectivity
(Scheme 7). On the basis of this preliminary result, the authors
next evaluated chiral non-racemic phosphine ligands in an at-
tempt to develop a catalytic, enantioselective process. The
highest levels of enantioselectivity are achieved with a trans-
chelating biferrocene ligand 30 and a bulky diisopropylmethyl-
substituted ester group. Under these reaction conditions nitrile
products containing a a-stereogenic quaternary centers could
be prepared in good yield and with moderate to good dia-
stereo- and enantioselectivities.
Scheme 5. Aldol reaction of tetrasubstituted enol borinates derived from the
1,4-hydroboration of acrylamides with (diisopinocampheyl)borane ((lIpc)2BH.
the enol borinates with a panel of aldehydes provided b-hy-
droxy amides (19) containing a stereogenic, a-quaternary
center in excellent diastereo- and enantioselectivity.
Other methods for controlling tetrasubstituted enolate ge-
ometry based on the use of either chiral auxiliaries or stoichio-
metric chiral bases have also been reported. Myers described
the use pseudoephedrine- and pseudoephenamine-derived
amides for selective enolizations with LDA and subsequent dia-
stereoselective enolate alkylation.[14] A chiral non-racemic lithi-
um amide was utilized by Zakarian to prepare geometrically
defined a,a-disubstituted silyl ketene acetals, which participate
in highly selective Ireland–Claisen rearrangements.[15]
Catalytic, enantioselective aldol reactions for the synthesis of
quaternary stereogenic centers
Each of the previous methods relies on the use of a chiral aux-
iliary as the stereocontrolling unit. Two initial investigations on
catalytic, enantioselective aldol reactions for setting quaternary
centers have been reported that employ distinct modes of cat-
alysis.
Scheme 7. Rhodium-catalyzed asymmetric direct aldol reaction for the
An amine-catalyzed, direct aldol addition of a,a-dialkyl alde-
hydes donors with aromatic aldehyde acceptors was identified
by Barbas and co-workers.[16] The authors found that with
10 mol% each of chiral diamine 24 and trifluoroacetic acid,
quaternary carbon containing aldol products (25) could be ob-
tained in good yield and enantioselectivity (Scheme 6). The
products are isolated with poor diastereomeric ratios, suggest-
synthesis of quaternary carbon atoms.
Despite these promising initial results for enantioselective
aldol reactions for synthesizing quaternary carbon atoms, the
generality of these systems remains limited. An analysis of the
results from the previous researchers reveals two central chal-
lenges that continue to impede the development of this reac-
tion: 1) controlling the enolate geometry of a,a-disubstituted
enolates and 2) achieving high enough reactivity in the eno-
late to overcome the intrinsic steric repulsions encountered in
the formation of quaternary carbon atoms. The development
of a successful method for the synthesis of quaternary carbon
atoms using the aldol addition must address these two chal-
lenges.
Research objectives
Silyl ketene imines (34) are a class of a,a-disubstituted nucleo-
philes that avoid the issues associated with enolate geometry.
The key structural feature in these species is the pair of orthog-
onal substituent planes, which imparts an axis of chirality
Scheme 6. Lewis base catalyzed asymmetric direct aldol reaction for the
synthesis of quaternary carbon atoms.
Chem. Eur. J. 2014, 20, 9268 – 9279
www.chemeurj.org
9271
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
when R1 and R2 are dissimilar. This unique geometry also
places a significant portion of the steric bulk in a plane per-
pendicular to and distal from the nucleophilic carbon atom,
which should alleviate some of the steric interactions encoun-
tered in the transition state for quaternary carbon formation.
Aldol-type reactions of these nucleophiles generate syntheti-
cally useful b-hydroxy nitriles (35) containing a a-quaternary
stereogenic center (Scheme 8).
Scheme 9. Proof of principle experiments for the Lewis base catalyzed, SiCl4-
mediated, aldol addition of silyl ketene imines with aromatic and aliphatic
aldehydes.
Scheme 8. Synthesis of quaternary carbon atoms by aldol addition of silyl
ketene imines.
observed for the addition of silyl ketene imine 36a to benzal-
dehyde, the addition to aliphatic aldehyde proved more chal-
lenging and would require more detailed optimization studies.
Encouraged by the proof of principle experiment with benzal-
dehyde further studies were undertaken to establish the rates
of both the background and catalyzed reaction.
Although the preparation and characterization of silyl
ketene imines (SKIs) are well known, only a few reports have
documented their use as nucleophiles in catalytic asymmetric
reactions.[19] Early work by Frainnett and co-workers establish
that silyl ketene imines will undergo exothermic additions to
aldehydes; however, at the onset of this work only a single
study by Fu and co-workers had documented the use of SKIs
in a catalytic, enantioselective acylation reaction.[19a] Previous
studies in these laboratories on the Lewis base catalyzed, sili-
con tetrachloride mediated additions of silyl ketene acetals
have attested to the sensitivity of this catalyst system to minor
changes in nucleophile structure.[20] These studies suggested
that the asymmetric environment provided by this catalyst
system would be well suited for discriminating the two carbon
substituents of a silyl ketene imine. Despite these promising
attributes very little was known about the stability and reactivi-
ty of silyl ketene imines toward the SiCl4/Lewis base catalyst
system. Therefore the first goal of this study was to test the
compatibility and background reactions of silyl ketene imines
with simple aromatic aldehydes. The long-term goals of the
study were to develop general, and highly selective carbonyl
addition reactions for setting quaternary stereogenic cen-
ters.[19b–d]
In situ IR monitoring of the background and catalyzed
reaction rates
The reaction rate was determined by monitoring the loss of
the aldehyde absorption by in situ IR kinetic analysis for the
addition of silyl ketene imine 36a to 1-naphthaldehyde in the
presence of 5 mol% of (R,R)-41a and 1.1 equivalents of SiCl4 at
À658C. A slower-reacting, more-hindered aromatic aldehyde
was chosen for this study to allow for the maximum spectral
resolution. The plot of the aldehyde absorbance at 1700 cmÀ1
versus reaction time for the catalyzed reaction (red line) is
shown in Figure 3, and indicates that the reaction is complete
within 3 min of the addition of the SKI. Furthermore, upon
quenching the reaction into a saturated aqueous solution of
KF/NaHCO3, and following aqueous workup and purification by
column chromatography, the nitrile product 39ab was isolated
Results
Proof of principle studies on the addition of silyl ketene
imines to aldehydes
The initial investigations tested the compatibility and reactivity
of silyl ketene imines under the standard reaction conditions
previously developed for the additions of silyl ketene acetals
to aldehydes. Reaction of silyl ketene imine 36a with both an
aromatic and an aliphatic aldehyde were examined (Scheme 9).
For the addition to benzaldehyde, the results were very prom-
ising, showing not only that the silyl ketene imine was stable
under the reaction conditions, but also that the addition prod-
uct was isolated in high yield. Moreover, the b-hydroxy nitrile
product (39aa) was isolated with high diastereo- and enantio-
selectivity. Although good reactivity and stereoselectivity was
Figure 3. In situ IR kinetic data for catalyzed and background reactions of
SKI 36a addition to 1-naphthaldehyde.
Chem. Eur. J. 2014, 20, 9268 – 9279
www.chemeurj.org
9272
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
in 70% yield and with excellent diastereo- and enantioselectivi-
ty.
Having established the reaction rate for the catalyzed pro-
cess, the achiral background reaction was next examined. To
monitor this rate the same experimental procedure was fol-
lowed, except the Lewis base catalyst (R,R)-41a was not
added. Surprisingly, a significant background reaction was ob-
served showing about 70% conversion of the aldehyde in only
10 min (blue line, Figure 3). Nevertheless, the achiral reaction
does not interfere with the catalyzed process since the nitrile
product is obtained with excellent asymmetric induction.
In situ IR rate comparison of silyl ketene imines and silyl
ketene acetals
The extremely facile reaction rates observed in both the cata-
lyzed and uncatalyzed addition of SKI 36a to 1-naphthalde-
hyde was unexpected and prompted further study. To eluci-
date how the unique geometry of the SKI may be affecting the
observed reaction rates, a comparative rate study between silyl
ketene imines and analogously substituted ketene acetals was
conducted. Silyl ketene acetals (SKAs) were chosen for this
study because they are the ester analogues of ketene imines
and have the same oxidation state at the a and b carbon
atoms. Structurally though, ketene acetals differ from ketene
imines because they lack an orthogonal substituent plane, and
a majority of the steric bulk resides in the same plane as the
reacting carbon. Additionally, monosubstituted SKAs have
been extensively studied within this catalyst system and are
known to be excellent substrates for the addition to aromatic
aldehydes.
Figure 4. In situ IR study comparing reaction rate of silyl ketene acetal (42a)
and silyl ketene imine (36a) in the addition to 1-naphthaldehyde.
and electron-rich aromatic aldehydes reacted with similar rates
and selectivities to benzaldehyde (Table 1, entries 3–5). Finally,
only a slight decrease in the enantioselectivity was observed
for reaction with the electron-rich heteroaromatic aldehyde, 2-
furaldehyde (Table 1, entry 8).
Table 1. Lewis base catalyzed aldol addition of a-phenylpropionitrile-de-
rived silyl ketene imine 36a with aromatic aldehydes.
To test the rate difference, SKA 42a was prepared as a mix-
ture of E/Z isomers (by lithiation of methyl 2-phenylpropionate
followed by trapping with TBSCl) and its reactivity was assayed
by in situ IR monitoring in the addition to 1-naphthaldehyde
using the same experimental conditions as described for SKI
36a. The IR data for each nucleophile is plotted in Figure 4
and the observed difference in reaction rates is dramatic.
Under identical reaction conditions, SKA 42a was nearly un-
reactive, whereas SKI 36a reacted to completion in less than
3 min.
Entry
Aryl
Product
Yield
[%][b]
d.r.[c]
e.r.[c]
1
2
3
4
5
6
7
8
C6H5 (37a)
39aa
39ab
39ac
39ad
39ae
39af
39ag
39ah
87
76
88
93
93
78
84
92
95:5
98.5:1.5
98.4:1.6
99.3:0.7
98.9:1.1
98.6:1.4
96.6:3.4
99.2:0.8
94.9:5.1
1-naphthyl (37b)
4-CF3C6H4 (37c)
4-BrC6H4 (37d)
4-CH3CO2C6H4 (37e)
4-CH3OC6H4 (37 f)
2-CH3C6H4 (37g)
2-furyl (37h)
>99:1
>99:1[d]
99:1
>99:1
96:4
>99:1
99:1
Survey of aldehyde structure in the addition of a silyl
ketene imine
[a] Reactions employed 1.1 equivalents of SiCl4, 1.2 equiv of silyl ketene
imine, 0.05 equivalents of (R,R)-41a at 0.25m in CH2Cl2 at À788C for 2 h.
[b] Yield of analytically pure material. [c] Determined by CSP-SFC. [d] De-
Motivated by the promising results obtained from the in situ
IR rate studies, a more thorough study of the scope of this pro-
cess with respect to the aldehyde structure was conducted. A
wide range of aromatic aldehydes, including electron-neutral,
electron-rich, electron-deficient, and heteroaromatic, were sur-
veyed in the addition of SKI 36a and overall consistently high
selectivities and yields were observed (Table 1). Electron-neu-
tral aromatic aldehydes: benzaldehyde, 1-naphthaldehyde, and
2-naphthaldehyde all reacted with comparable rates and selec-
tivities. Furthermore, only a slight drop in the enantioselectivity
was observed for addition to the more sterically encumbered
aldehyde, 1-naphthaldehyde (Table 1, entry 2). Electron-poor
1
termined by H NMR analysis.
Synthesis and survey of silyl ketene imine additions to
aromatic aldehydes
To further probe the scope of the Lewis base catalyzed aldol
additions of SKIs, a thorough survey of the ketene imine struc-
ture was conducted. In formulating the parameters to evaluate
in such a survey, a number of issues were considered: 1) what
degree of steric differentiation in the alkyl groups of the
Chem. Eur. J. 2014, 20, 9268 – 9279
www.chemeurj.org
9273
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ketene imine is required to retain high levels of diastereoselec-
tivity in the aldol addition and 2) how will the reactivity and
stability of aryl- versus dialkyl-substituted ketene imines com-
pare under the current reaction conditions? The latter question
is especially relevant, since very few dialkyl silyl ketene imines
have been reported. To address these questions, a number of
different silyl ketene imines were prepared from the corre-
sponding a,a-disubstituted nitriles.
Table 3. Lewis base catalyzed aldol addition reaction of a,a-disubstituted
silyl ketene imines with 2-naphthaldehyde.[a]
Entry
SKI
R1
R2
Product
Yield
[%][b]
d.r.[c]
e.r.[d]
The preparation of silyl ketene imines is straightforward and
although modified protocols have been developed all basically
follow the same general procedure first reported by Watt and
co-workers[21] and subsequently modified by Fu.[19a] The modi-
fied procedure involves lithiation and trapping of the nitrile
(43) with TBSCl, and following evaporation of the reaction sol-
vent under vacuum, the LiCl byproduct was removed by pre-
cipitation with pentanes and anhydrous filtration. The product
ketene imines (36a–g) were then obtained in high yield and
purity by simply concentrating the pentane solution under re-
duced pressure. Following this general protocol, a number of
silyl ketene imines (36a–g) were prepared from their respec-
tive nitrile precursors (Table 2). Both a-phenyl (Table 2, en-
tries 1–5) and a-alkyl (Table 2, entries 6 and 7) nitriles reacted
similarly, producing silyl ketene imines in high yields and great-
1
2
3
4
5
6
7
36a
36b
36c
36d
36e
36 f
36g
Me
Et
iBu
iPr
Ph
Ph
Ph
Ph
Ph
39ai
39bi
39ai
39di
39ei
90
78
90
73[c]
79
85
92
98:2
97:3
99:1
61:39
94:6
N/A
98.7:1.3
92.7:7.3
99.6:0.4
78.9:21.1[e]
97.5:2.5
91.2:8.8
92.1:7.9[f]
allyl
À(CH2)5À 39 fi
Me iPr 39gi
60:40
[a] Reactions employed 1.1 equivalents of SiCl4, 1.2 equivalents of silyl
ketene imine, 0.05 equivalents of (R,R)-41a at 0.25m in DCM at À788C for
2 h. [b] Yield of analytically pure material. [c] Yield of chromatographically
homogenous material. [d] Determined by CSP-SFC. [e] Enantiomeric ratio
of the minor diastereomer was 71.4:28.6. [f] Enantiomeric ratio of the
minor diastereomer was 96.6:3.4.
1
er than 98% purity as judged by H NMR analysis of the crude
36a–g to 2-naphthaldehyde (37i) were surveyed in the pres-
ence of 5 mol% of (R,R)-41a and 1.1 equiv of SiCl4 (Table 3).
First, a-alkylbenzylnitrile-derived ketene imines 36a–e were
tested in the aldol addition. The results show that although
steric bulk can be well tolerated at this position, the presence
of an a-branched substituent leads to a drop in both the dia-
stereomeric and enantiomeric purity of the product (Table 3,
compare entries 1–3 to entry 4). The synthetically useful allyl-
substituted silyl ketene imine 36e was well tolerated in the re-
action providing a nitrile product in good yield and high selec-
tivity (Table 3, entry 5). To further expand the nucleophile
scope, two dialkyl-substituted SKIs that do not contain an aryl
ring were prepared and tested in the addition to 2-naphthal-
dehyde. The cyclohexane-derived ketene imine 36 f provided
an aldol product with a non-stereogenic quaternary carbon in
good yield and enantioselectivity (Table 3, entry 6). Silyl ketene
imine 36g, containing disparate alkyl groups, reacted to give
a 60:40 mixture of enantiomerically enriched diastereomers in
good yield (Table 3, entry 7). The high enantiomeric ratio ob-
served within each diastereomer suggests that source of low
d.r. was insufficient steric differentiation in the alkyl substitu-
ents of the ketene imine.
reaction mixtures. Whereas some of these nucleophiles are
stable to distillation (Table 2, entry 1), others have a tendency
to isomerize to the C-silyl nitrile with prolonged heating. Im-
portantly, no significant differences in the yield or selectivity of
the aldol addition reactions have been observed when crude
silyl ketene imines are employed. In addition, Fu and co-work-
er’s have demonstrated the ability of disubstituted C-silyl ni-
triles to participate in enantioselective acylation reactions cata-
lyzed by Lewis bases.[19a] Preliminary experiments using the
conditions described in Table 1 have shown that C-silyl nitriles
are unreactive in this catalyst system.
To test what role the nucleophile structure plays on the se-
lectivity of the reaction, the additions of silyl ketene imines
Table 2. Synthesis of tert-butyldimethylsilyl ketene imines from a,a-di-
substituted nitriles.
Next, a survey of silyl ketene imines with electron-rich, elec-
tron-poor, and sterically encumbered aryl substituents was
conducted using benzaldehyde as the electrophile under the
same conditions as the previous series (Table 4).
Entry
R1
R2
Product
Yield [%][b]
1
2
3
4
5
6
7
Ph
Ph
Ph
Ph
Ph
À(CH2)5À
Me
Me
Et
iPr
iBu
allyl
36a
36b
36c
36d
36e
36 f
35g
95 (73[c])
96
98
98
96
90
92
All members of this series reacted at indistinguishably rapid
rates, and provided the products in good yields. Electron-with-
drawing (Table 4, entry 1) and electron-donating (Table 4,
entry 2) substituents in the para-position had no significant ef-
fects on the stereoselectivity of the reactions, while adding
steric encumbrance at the ortho-position (Table 4, entry 3) re-
sulted in a small reduction in enantioselectivity and a moderate
reduction in diastereoselectivity.
iPr
[a] All reactions employed 1.0 equivalents of diisopropyl amine, 1.0 equiv-
alents of nBuLi and 1.2 equivalents of TBSCl. [b] Yield of crude material,
purity judged to be >95% by H NMR (500 MHz) analysis of the crude re-
action mixture. [c] Yield reported after short-path distillation under re-
duced pressure.
1
Chem. Eur. J. 2014, 20, 9268 – 9279
www.chemeurj.org
9274
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(Table 5, entry 11). Gratifyingly, a small increase in enantioselec-
tivity was observed when a full equivalent of nBu4NI was used
(Table 5, entry 12). Interestingly, there was no corresponding
increase in rate or yield. Finally, by combining previous positive
results, a set of optimal conditions was estimated (Table 5,
entry 13). Under these conditions, high yield as well as enantio-
and diastereoselectivities were achieved.
Table 4. Scope of Lewis base catalyzed aldol reaction with aryl-substitut-
ed silyl ketene imines.
Entry Aryl
R
SKI
Product Yield d.r.[b]
[%][a]
e.r.[b]
Several additional phosphoramide catalysts (41c–e) were
prepared to explore the suitability of dimethylbiphenyldiamine
as an alternative to 1,1’-binapthyl-2,2’-diamine, as well as the
effects of substitution at the 3,3’- and 4,4’-positions. Substan-
tially lower enantioselectivity was observed with biphenyl cata-
lyst (S,S)-41c compared to (R,R)-41a (Table 5, entry 14) howev-
er the diastereoselectivity was unchanged. The 5,5’-substituted
catalyst (R,R)-41d provided the same selectivity as (S,S)-41c,
albeit in opposite absolute configuration (Table 5, entry 15).
Disappointingly, catalyst (S,S)-41e, which is substituted at the
3,3’-position, (Table 5, entry 16) produced only racemic product
with low diastereoselectivity.
1
2
3
4-CF3C6H4
4-CH3OC6H4 Me 36i 39ia
2-CH3C6H4 Me 36j 39ja
Et
36h 39ha
73
90
74
97:3[c]
98.7:1.3
87:13[c] 94.2:5.8[d]
99.5:0.5
99.1:0.9
[a] Reactions employed 1.1 equivalents of SiCl4, 1.2 equivalents of silyl
ketene imine, 0.05 equivalents of (R,R)-41a at 0.25m in DCM at À788C for
2 h. [b] Yield of analytically pure material. [c] Determined by CSP-SFC.
1
[c] Determined by H NMR analysis. [d] Determined by CSP-HPLC after de-
rivatization with 3,5-dinitrobenzoyl chloride.
Addition of silyl ketene imines to aliphatic aldehydes
The reactions of aliphatic aldehydes with enoxysilane nucleo-
philes under Lewis base/silicon tetrachloride activation poses
a challenge due to the formation of a parasitic trichlorosilyl
chlorohydrin species II (Scheme 10). This difficulty can be over-
come with a sufficiently reactive nucleophile. It was hypothe-
sized that the high reactivity of silyl ketene imines could be ap-
plied to address the problem of addition to aliphatic alde-
hydes.
Scheme 10. Chloride trapping of activated aldehydes results in chlorohydrin
formation.
Next, a survey of simple aliphatic aldehydes was undertaken
to explore the scope of this reaction. Five aliphatic aldehydes
were selected to evaluate the effects of substitution patterns
on reactivity and selectivity (Table 6). The reactions of silyl
ketene imine 36k with 38b, which bears an n-pentyl chain,
(Table 6, entry 1) and benzyloxy-functionalized aldehyde 38c
(Table 6, entry 2) proceeded in good yields although the enan-
tioselectivities were somewhat lower than was observed with
38a. The reaction with b-branched aldehyde 38d also pro-
ceeded in good yield; however, the enantioselectivity was still
lower. The silyl ketene imine addition was found to be com-
pletely intolerant of a-branching at the aldehyde (Table 6,
entry 4), which lead to no reaction, as well as a-aryl substitu-
tion (Table 6, entry 5), which lead to a complex mixture. The
latter is likely due to the increased acidity of the aldehyde and
consequent increased self-condensation.
A survey of reaction conditions was undertaken to optimize
the addition of silyl ketene imines to an aliphatic aldehyde
(Table 5). Monitoring of the course of the reaction was conven-
iently accomplished by in situ FT-IR monitoring, observing the
very strong silyl ketene imine stretch (2030 cmÀ1). Cooling to
08C or below was required to control side reactions (Table 5,
entries 1–3), and À208C provided optimal selectivity (Table 5,
entry 5). nBu4NI, which had previously been applied to the ad-
dition of silyl ketene acetals,[20a] proved superior to other hal-
ides, whereas nBu4NOTf did not affect the rate (Table 5,
entry 8) and nBu4NCl (Table 5, entry 6) induced rapid desilyla-
tion of the ketene imine. The use of TIPS-protected ketene
imine 36k at 08C in place of 36a (Table 5, entry 9) improved
selectivity to a level comparable to that obtained with 36a at
À208C (Table 5, entry 5), with a rate similar to that observed at
À108C (Table 5, entry 4). Further increases in the size of the
silyl group reduced enantioselectivity and yield (Table 5,
entry 10). The omission of nBu4NI led to lower enantio- and
diastereoselectivity as well as decreased rate, which suggested
that an increased loading of this reagent might be beneficial
Determination of relative and absolute configurations of the
aldol products
The b-hydroxy nitrile products prepared in this study had not
been previously reported and consequently the absolute con-
Chem. Eur. J. 2014, 20, 9268 – 9279
www.chemeurj.org
9275
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 5. Optimization of silyl ketene imine addition to aliphatic aldehydes.
Entry
catalyst
(equiv)
SiR3
nBu4NX
(equiv)
T
[8C]
t
[h]
Yield[a]
[%]
e.r.[b]
d.r.[c]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
41b (0.2)
41b (0.2)
41a (0.05)
41a (0.05)
41a (0.05)
41a (0.05)
41a (0.05)
41a (0.05)
41a (0.05)
41a (0.05)
41a (0.05)
41a (0.05)
41a (0.05)
41c (0.05)
41d (0.05)
41e (0.05)
TBS
TBS
TBS
TBS
TBS
TBS
TBS
TBS
nBu4NI (0.5)
nBu4NI (0.5)
nBu4NI (0.5)
nBu4NI (0.5)
nBu4NI (0.5)
nBu4NCl (0.5)
nBu4NBr (0.5)
nBu4NOTf (0.5)
nBu4NI (0.5)
nBu4NI (0.5)
–
nBu4NI (1.0)
nBu4NI (1.0)
nBu4NI (1.0)
nBu4NI (1.0)
nBu4NI (1.0)
23
0–14
0
À10
À20
0
0
0
0
0
0
0
À20
À20
À20
À20
17
17
24
18
17
1
17
17
17
17
17
17
22
24
24
24
0
33
66
74
66
trace
56
trace
63
20
34
–
–
–
–
88:12
90:10
93:7
—
87:13
–
90:10
90:10
91:9
–
82:18
–
93:7
TIPS (36k)
TBDPS (36l)
TBS
92:8
78:22
71:29
90:10
94.5:5.5
24:76
76:24
50:50
93:7
80:20
90:10
98.5:1.5
98.5:1.5
98.8:1.2
75:25
TBS
67
84
95
69
TIPS (36k)
TIPS (36k)
TIPS (36k)
TIPS (36k)
84
1
[a] Yield after chromatography. [b] Determined by CSP-SFC. [c] Determined by H NMR analysis.
Discussion
Table 6. Substrate scope in the addition of a SKI 36k to aliphatic alde-
hydes.
Trends in reactivity with respect to silyl ketene imine
structure
The extremely facile addition rate exhibited by silyl ketene
imines under the SiCl4/bisphosphoramide catalyst system is
truly remarkable and has allowed for the synthesis of quaterna-
ry carbon containing aldol products in high yields and stereo-
selectivities for the first time. To probe how the structure of
silyl ketene imines could be accounting for the observed reac-
tion rates a comparative study of silyl ketene acetals and silyl
ketene imines was conducted. The results of these experiments
showed that silyl ketene imines were significantly more reac-
tive than silyl ketene acetals in this catalyst system. This large
disparity in observed reactivity between silyl ketene imines
and ketene acetals most likely results from both steric and
electronic differences that exist between these two nucleophile
classes. However, the steric component may play a more domi-
nant role in these reactions due to the congestion associated
with the formation of a quaternary center. This is especially ap-
parent when comparing the open transition-state models,
which have been proposed to explain the stereochemical out-
come for addition of silyl ketene acetals to aldehydes with the
(R,R)-41a/SiCl4 catalyst system (Figure 5). Previous mechanistic
and computational studies have suggested that the active cat-
alytic species in these reactions is a phosphoramide-bound tri-
chlorosilyl cation.[20a,23] This highly electrophilic chiral Lewis
acid activates the aldehyde through coordination to the lone
pair of the carbonyl and then controls the relative and abso-
lute topicity for the combination of two prochiral reactants.
The high diastereoselectivity observed for these additions can
Entry
Product
R
Yield[a] [%]
e.r.[b]
d.r.[c]
98:2
98.5:1.5
>95:5
1
2
3
4
5
40ab
40ac
40ad
40ae
40af
CH3(CH2)4
BnO(CH2)5
iBu
iPr
PhCH2
94
83
85
0
~91:9[d]
90:10
<85:15[d]
–
–
–
–
0
[a] Yield after chromatography [b] Determined by CSP-SFC [c] Determined
by 1H NMR analysis [d] Incomplete separation of stereoisomers.
figurations could not be determined by comparison of the op-
tical rotations to known compounds. Furthermore, given the
current challenges for preparing aldol-type products contain-
ing a quaternary stereogenic center it became difficult to even
find suitable literature compounds that could be arrived at by
functional group manipulation of the nitrile. Therefore, the ab-
solute and relative configurations of the products were estab-
lished by single-crystal X-ray crystallography on nitrile product
39ad.[22] The S configuration at the alcohol center (C3) con-
firms that the nucleophile adds to the Re-face of the aldehyde
and is in agreement with the sense of asymmetric induction
observed in other reaction manifolds reported for this catalyst
system.
Chem. Eur. J. 2014, 20, 9268 – 9279
www.chemeurj.org
9276
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 5. Open transition structures comparing silyl ketene imines and
acetals.
then be rationalized by minimizing steric interactions between
the approaching nucleophile and the sizable trichlorosilyl
cation. Comparing the antiperiplanar open transition structures
leading to the anti-aldol product for the addition of disubsti-
tuted silyl ketene acetals and silyl ketene imines show dramatic
differences. For example, the transition structure for the addi-
tion of either E or Z silyl ketene acetal 42a to benzaldehyde
both suffer from unfavorable steric interactions with the alkoxy
substituents (Figure 5). This space is completely open in the
transition state proposed for the addition of silyl ketene imine
36a to benzaldehyde. Furthermore, the bulky TBS substituent
of the silyl ketene imine can occupy a position in space per-
pendicular to the plane of the aldehyde and away from the
congestion of the newly forming quaternary center.
Figure 6. a) Calculated transition structure energies for addition of 36a to
benzaldehyde. b) Calculated transition structures for Àscanti and +scsyn addi-
tion of 36a to benzaldehyde.
Reactivity trends with respect to aldehyde structure
Excellent selectivities were observed in the additions of silyl
ketene imines to aromatic aldehydes. This outcome can be at-
tributed largely to the extensive catalyst optimization studies
that had previously been performed on Lewis base catalyzed,
SiCl4-promoted aldol and allylation reactions, culminating in
the development of phosphoramide catalyst 41a.[24]
Silyl ketene imine additions to aliphatic aldehydes were ach-
ieved at temperatures higher than those required for aromatic
aldehydes. The levels of enantioselectivity observed in these
reactions were more variable than those observed in additions
to aromatic aldehydes, ranging from 94.5:5.5 to <85:15. Some
of this reduction in selectivity may be the result of the reaction
temperature. Aliphatic aldehydes are more conformationally
flexible and structurally diverse than aromatic aldehydes, either
of which may contribute to increased variability in substrate
catalyst interactions.
Second, the magnitude of steric interactions of the silyl group
will depend on the degree of rehybridization occurring in the
transition state. Although the silyl groups in Figure 6a are
drawn as pointing out of the page, away from major interac-
tions with anything, this is merely one set of limiting, ground-
state like structures. Later in the reaction coordinate, the Si-N-
C angles should approach 1808 as the nitrogen of ketene
imine 36a becomes sp-hybridized, potentially bringing the
silyl group into close interaction with sections of the transition
structure some distance away from the forming bond.
In order to better understand the factors effecting diastereo-
selectivity in silyl ketene imine additions, semiempirical calcula-
tions (PM6, MOPAC2007 and MOPAC 2009)[25] were performed
to locate transition structures of the reaction of 36a, SiCl4, cat-
alyst (R,R)-41a, and benzaldehyde (Figure 6b). The lowest
energy transition structure (46b Àscanti) corresponds to the
major diastereomer obtained experimentally.
Rationalization of diastereoselectivity in silyl ketene imine
additions
Several aspects of these transition structures are of particular
interest. Most notably, the two lowest energy transition struc-
tures are open synclinal (46b Àscanti and 46 f +scsyn, Fig-
ure 6b). The unfavorable dipole–dipole interactions, which dis-
favor open synclinal transition states in aldol reactions of non-
chelating alkali metal enolates and naked enolates,[26] are only
modestly strong in the more closely analogous Mukaiyama
type aldol reactions[27] and may be outweighed by steric fac-
tors in the case of silyl ketene imine additions.
The diastereoselectivity of silyl ketene imine additions to alde-
hydes poses several interesting questions. The assumption of
an open transition structure is justified by the difficulty of co-
ordinating a nucleophile to an already hexacoordinate silicon
complex. This assumption leads to two enantiomeric sets of six
possible transition structures (46a–f), of which three (46a–c)
are consistent with the observed diastereoselectivity (Fig-
ure 6a). Which of these dominates depends on several issues
that cannot be answered from a simple diagram. First, by anal-
ogy to the chemistry of silyl ketene acetals, an electronic pref-
erence may operate in the anti transition-state structures.
In the lowest energy structure, 46b Àscanti, the ketene imine
axis faces the bulky silicon Lewis acid, indicating that the
ketene imine is effectively the smallest substituent, despite the
Chem. Eur. J. 2014, 20, 9268 – 9279
www.chemeurj.org
9277
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
possibility of steric interactions with the TBS group (Figure 6b).
The next smallest substituent, the methyl group, is positioned
to the right as drawn, in a pocket formed by one of the naph-
thyl groups of the catalyst. This interaction may be how the
catalyst reinforces the inherent diastereoselectivity of the reac-
tion, as the far wall of the pocket is positioned to encounter
the TBS group. This interaction may be seen in structure
46 f +scsyn, in which the O-C-C-C dihedral angle of the forming
bond is distorted to 928, compared to the ideal 608, to avoid
this interaction (Figure 6b). The largest substituent, the phenyl
group, points downward, a direction which appears essentially
open. No attempt was made to locate the six additional transi-
tion structures that are isomeric at the ketene imine chiral axis,
however, the C=NÀSi angles in the six evaluated transition
structures are 172–1748, which suggests that this omission is
of little or no consequence, since the chiral axis has essentially
been lost in the transition state. Furthermore, it is likely that
silyl ketene imines undergo rapid stereoinversion under the re-
action conditions, as no splitting of diastereotopic signals can
The N-methyl group on the binaphthyl ring of the catalyst pro-
trudes far into the binding pocket, effectively shielding the Si-
face of the aldehyde and leaving the Re-face exposed for nu-
cleophilic attack.
The absolute configuration for the nitrile products derived
from silyl ketene imine additions to aromatic aldehydes cata-
lyzed by (R,R)-41a are also consistent with this stereochemical
model. The S configuration of the alcohol center confirmed
that SKI underwent addition to the Re-face of the aldehyde
and open transition-state models based on minimizing interac-
tions between the silyl cation and phenyl substituents of the
SKI are consistent with the observed relative configuration.
These findings provide further support for the current stereo-
chemical model.
Conclusion
The addition of silyl ketene imines to aromatic aldehydes is an
efficient process for the enantioselective construction of qua-
ternary stereocenters. The addition can accommodate a variety
of substituents on both the aldehyde and the ketene imine,
and products are generally isolated in high yields, as well as
excellent diastereomeric and enantiomeric ratios. In situ IR
monitoring for the Lewis base catalyzed addition of silyl
ketene imines to aromatic aldehydes reveal an extremely facile
reaction rate. Directly comparing silyl ketene imines to silyl
ketene acetals in this reaction system suggests that the unique
structure of the SKI accounts for the dramatic rate difference
observed between these two nucleophile classes.
1
be observed by H NMR spectroscopy even at À608C. Previous
studies have found low inversion barriers in analogous alkyl
and aryl ketene imine systems.[28]
Rationale for the observed enantioselectivity with aromatic
aldehydes
The structure of the Lewis base activated complex between
(R,R)-41a and SiCl4 has been of much interest, and kinetic stud-
ies have suggested that the catalytically active species involves
a hexacoordinate silicate bound by two phosphoramide moie-
ties.[23b] Direct evidence for the hexacoordinate silicon complex
through X-ray crystallography has been elusive due to the
transient nature of the phosphoramide–silicon bond. However,
hexacoordinate complexes of SiCl4 with hexamethylphosphor-
amide[23a] and SnCl4 with bisphosphoramides[29] have been ob-
served and characterized by X-ray crystallography in these lab-
oratories. These results have aided in the development of
a working computational model of the trichlorosilyl cation, in
which both the chiral bisphosphoramide and a substrate alde-
hyde are bound to the silicon center (Figure 7).[20a] In the mini-
mized structure, the aldehyde binds trans to one of the phos-
phoramides, owing to the nature of the hypervalent bonds in
the ligand field around silicon. This geometry places the alde-
hyde close to one of the binaphthyl rings of the Lewis base
catalyst, possibly stabilized by an edge-to-face p–p interaction.
The S-absolute configuration for the nitrile product as deter-
mined by single-crystal x-ray crystallography is consistent with
Re-face addition of the nucleophile to the aldehyde. Computa-
tional modeling was utilized to further examine the relative
topicity for the approach of the ketene imine to the activated
aldehyde/Lewis acid complex. On the basis of these studies,
the major stereoisomer for the addition of silyl ketene imines
is seen to arise from a synclinal transition structure in which
the Re-face of the aldehyde is the most accessible.
The reduced reaction rate observed for addition of silyl
ketene imines to aliphatic aldehydes could be overcome by in-
creasing the reaction temperature and employing nBu4NI as an
additive. Under the optimized reaction conditions, the addition
of silyl ketene imines to aliphatic aldehydes was achieved in
high yield with un-hindered linear aliphatic aldehydes. The re-
action rate of this process was strongly dependent on alde-
hyde structure, for which more hindered aliphatic aldehydes
were unreactive. The enantioselectivity for the addition of silyl
ketene imines to aliphatic aldehydes ranged from good to ex-
cellent. Efforts to identify a catalyst structure that would allow
for increased reactivity with a broader class of aliphatic alde-
hydes and/or better enantioselectivity with linear aliphatic al-
dehydes were unsuccessful. The current binaphthylamine de-
rived Lewis base, (R,R)-41a, appears to represent at least
a local optimum as all accessible modifications produced less
selective catalysts. Future directions will focus on applying silyl
ketene imines to new classes of electrophiles.
Figure 7. Calculated model of the benzaldehyde–silyl cation complex opti-
mized with PM3 basis set using GAMESS(UC) QC package and visualized
using Chem3Dꢁ.
Chem. Eur. J. 2014, 20, 9268 – 9279
www.chemeurj.org
9278
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[13] C. Allais, A. S. Tsai, P. Nuhant, W. R. Roush, Angew. Chem. 2013, 125,
13126–13129; Angew. Chem. Int. Ed. 2013, 52, 12888–12891.
[14] a) D. A. Kummer, W. J. Chain, M. R. Morales, O. Quiroga, A. G. Myers, J.
Am. Chem. Soc. 2008, 130, 13231–13233; b) M. R. Morales, K. T. Mellem,
A. G. Myers, Angew. Chem. 2012, 124, 4646–4649; Angew. Chem. Int. Ed.
2012, 51, 4568–4571.
[15] a) Y.-c. Qin, C. E. Stivala, A. Zakarian, Angew. Chem. 2007, 119, 7610–
7613; Angew. Chem. Int. Ed. 2007, 46, 7466–7469; b) C. E. Stivala, A. Za-
karian, J. Am. Chem. Soc. 2008, 130, 3774–3776.
Acknowledgements
Financial support was provided by the National Science Foun-
dation (NSF CHE-0717989 and CHE-1012663). M.T.B. thanks the
University of Illinois for a Seemon H. Pines Graduate Fellow-
ship.
[16] N. Mase, F. Tanaka, C. F. Barbas, III, Angew. Chem. 2004, 116, 2474–
2477; Angew. Chem. Int. Ed. 2004, 43, 2420–2423.
Keywords: aldol reaction · asymmetric synthesis · quaternary
stereogenic center · semiempirical calculations · stereoselective
catalysis
[17] N. S. Chowdari, J. T. Suri, C. F. Barbas, III, Org. Lett. 2004, 6, 2507–2510.
[18] R. Kuwano, H. Miyazaki, Y. Ito, J. Organomet. Chem. 2000, 603, 18–29.
[19] a) A. H. Mermerian, G. C. Fu, Angew. Chem. 2005, 117, 971–974; Angew.
Chem. Int. Ed. 2005, 44, 949–952; b) S. E. Denmark, T. W. Wilson, M. T.
Burk, J. R. Heemstra, Jr., J. Am. Chem. Soc. 2007, 129, 14864–14865;
c) S. E. Denmark, T. W. Wilson, Synlett 2010, 1723–1728; d) S. E. Den-
mark, T. W. Wilson, Nat. Chem. 2010, 2, 937–943; e) G. T. Notte, V. J. M.
Baxter, J. L. Leighton, Org. Lett. 2011, 13, 816–818; f) S. E. Denmark,
T. W. Wilson, Angew. Chem. 2012, 124, 3290–3293; Angew. Chem. Int. Ed.
2012, 51, 3236–3239; g) S. E. Denmark, T. W. Wilson, Angew. Chem.
2012, 124, 10120–10132; Angew. Chem. Int. Ed. 2012, 51, 9980–9992;
h) J. Guin, G. Varseev, B. List, J. Am. Chem. Soc. 2013, 135, 2100–2103.
[20] a) S. E. Denmark, G. L. Beutner, T. Wynn, M. D. Eastgate, J. Am. Chem.
Soc. 2005, 127, 3774–3789; b) S. E. Denmark, W.-j. Chung, J. Org. Chem.
2008, 73, 4582–4595; c) S. E. Denmark, J. R. Heemstra, J. Org. Chem.
2007, 72, 5668–5688.
[1] a) D. A. Evans, J. V. Nelson, T. R. Taber, Top. Stereochem. 1982, 13, 1–116;
b) C. H. Heathcock, in Asymmetric Synthesis, Vol. 3 (Ed.: J. D. Morrison),
Academic Press, New York, 1984, pp. 111–212; c) I. Paterson, in Compre-
hensive Organic Synthesis Part 2 (Ed.: C. H. Heathcock), Pergamon,
Oxford, 1991, pp. 301–320; d) M. Braun, in Stereoselective Synthesis
Methods of Organic Chemisty, Vol. 3 (Eds.: G. Helmchen, R. Hoffman, J.
Mulzer, E. Schaumann), Houben-Weyl, Stuttgart, 1996, pp. 1603–1666;
e) R. Mahrwald, Modern Aldol Reactions, Vol. 1 and 2, Wiley-VCH, Wein-
heim, Germany, 2004; f) E. M. Carreira, A. Fettes, C. Marti, Org. React.
2006, 67, 1–216.
[2] a) E. J. Corey, A. Guzman-Perez, Angew. Chem. 1998, 110, 2092–2118;
Angew. Chem. Int. Ed. 1998, 37, 388–401; b) J. Christoffers, A. Mann,
Angew. Chem. 2001, 113, 4725–4732; Angew. Chem. Int. Ed. 2001, 40,
4591–4597; c) I. Denissova, L. Barriault, Tetrahedron 2003, 59, 10105–
10146; d) C. J. Douglas, L. E. Overman, Proc. Natl. Acad. Sci. USA 2004,
101, 5363–5367; e) J. Christoffers, A. Baro, Wiley-VCH, Weinheim, Ger-
many, 2005; f) B. M. Trost, C. Jiang, Synthesis 2006, 369–396; g) M.
Bella, T. Gasperi, Synthesis 2009, 1583–1614; h) C. Hawner, A. Alexakis,
Chem. Commun. 2010, 46, 7295–7306.
[3] a) I. Marek, G. Sklute, Chem. Commun. 2007, 1683–1691; b) J. P. Das, I.
Marek, Chem. Commun. 2011, 47, 4593–4623; c) I. Marek, Y. Minko, M.
Pasco, T. Mejuch, N. Gilboa, H. Chechik, J. P. Das, J. Am. Chem. Soc.
2014, 136, 2682–2694.
[4] a) E. A. Peterson, L. E. Overman, Proc. Natl. Acad. Sci. USA 2004, 101,
11943–11948; b) A. Y. Hong, B. M. Stoltz, Eur. J. Org. Chem. 2013, 2745–
2759.
[5] H. E. Zimmerman, M. D. Traxler, J. Am. Chem. Soc. 1957, 79, 1920–1923.
[6] R. E. Ireland, R. H. Mueller, A. K. Willard, J. Am. Chem. Soc. 1976, 98,
2868–2877.
[7] a) Z. A. Fataftah, I. E. Kopka, M. W. Rathke, J. Am. Chem. Soc. 1980, 102,
3959–3960; b) E. J. Corey, A. W. Gross, Tetrahedron Lett. 1984, 25, 495–
498; c) R. E. Ireland, P. Wipf, J. D. Armstrong, J. Org. Chem. 1991, 56,
650–657.
[21] D. S. Watt, Synth. Commun. 1974, 4, 127–131.
[22] CCDC-659734 (39ad) contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
[23] a) S. E. Denmark, B. M. Eklov, Chem. Eur. J. 2008, 14, 234–239; b) S. E.
Denmark, B. M. Eklov, P. J. Yao, M. D. Eastgate, J. Am. Chem. Soc. 2009,
131, 11770–11787.
[24] S. E. Denmark, T. Wynn, J. Am. Chem. Soc. 2001, 123, 6199–6200.
[25] a) J. Stewart, J. Mol. Model. 2007, 13, 1173–1213; b) J. J. P. Stewart,
MOPAC2007 ed., Stewart Computational Chemistry, Colorado Springs,
CO, USA, http://OpenMOPAC.net, 2007.
[26] a) C. H. Heathcock, S. K. Davidsen, K. T. Hug, L. A. Flippin, J. Org. Chem.
1986, 51, 3027–3037; b) S. E. Denmark, B. R. Henke, J. Am. Chem. Soc.
1989, 111, 8032–8034; c) S. E. Denmark, B. D. Griedel, D. M. Coe, J. Org.
Chem. 1993, 58, 988–990; d) S. E. Denmark, B. R. Henke, J. Am. Chem.
Soc. 1991, 113, 2177–2194.
[27] S. E. Denmark, W. Lee, Chem. Asian J. 2008, 3, 327–341.
[28] a) J. C. Jochims, J. Lambrecht, U. Burkert, L. Zsolnai, G. Huttner, Tetrahe-
dron 1984, 40, 893–903; b) J. Lambrecht, B. Gambke, J. Vonseyerl, G.
Huttner, G. K. Nell, S. Herzberger, J. C. Jochims, Chem. Ber. 1981, 114,
3751–3771.
[8] J. M. Manthorpe, J. L. Gleason, J. Am. Chem. Soc. 2001, 123, 2091–2092.
[9] E. D. Burke, J. L. Gleason, Org. Lett. 2004, 6, 405–407.
[10] E. A. Tiong, J. L. Gleason, Org. Lett. 2009, 11, 1725–1728.
[11] J. P. Das, H. Chechik, I. Marek, Nat. Chem. 2009, 1, 128–132.
[12] Y. Minko, M. Pasco, L. Lercher, M. Botoshansky, I. Marek, Nature 2012,
490, 522–526.
[29] S. E. Denmark, X. P. Su, Tetrahedron 1999, 55, 8727–8738.
Received: April 30, 2014
Published online on July 7, 2014
Chem. Eur. J. 2014, 20, 9268 – 9279
www.chemeurj.org
9279
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim