A combined DFT and NMR investigation of the zinc organometallic intermediate proposed in the syn-selective tandem chain extension-aldol reaction of β-keto esters

Article abstract of DOI:10.1021/jo3004925

The tandem chain extension-aldol (TCA) reaction of β-keto esters provides an α-substituted γ-keto ester with an average syn:anti selectivity of 10:1. It is proposed that the reaction proceeds via a carbon-zinc bound organometallic intermediate potentially bearing mechanistic similarity to the Reformatsky reaction. Evidence, derived from control Reformatsky reactions and a study of the structure of the TCA intermediate utilizing DFT methods and NMR spectroscopy, suggests the γ-keto group of the TCA intermediate plays a significant role in diastereoselectivity observed in this reaction. Such coordination effects have design implications for future zinc mediated reactions.

Full text of DOI:10.1021/jo3004925

Article
pubs.acs.org/joc
A Combined DFT and NMR Investigation of the Zinc Organometallic
Intermediate Proposed in the Syn-Selective Tandem Chain
Extension−Aldol Reaction of β-Keto Esters
Karelle S. Aiken,† Wilhelm A. Eger,‡ Craig M. Williams,‡ Carley M. Spencer,† and Charles K. Zercher*,†
†Department of Chemistry, University of New Hampshire, Durham, New Hampshire 03824, United States
‡School of Chemistry and Molecular Bioscience, University of Queensland, St. Lucia, Brisbane, QLD 4067, Australia
S
* Supporting Information
ABSTRACT: The tandem chain extension−aldol (TCA) reaction of β-keto esters provides an α-substituted γ-keto ester with an
average syn:anti selectivity of 10:1. It is proposed that the reaction proceeds via a carbon−zinc bound organometallic
intermediate potentially bearing mechanistic similarity to the Reformatsky reaction. Evidence, derived from control Reformatsky
reactions and a study of the structure of the TCA intermediate utilizing DFT methods and NMR spectroscopy, suggests the γ-
keto group of the TCA intermediate plays a significant role in diastereoselectivity observed in this reaction. Such coordination
effects have design implications for future zinc mediated reactions.
reaction operates under kinetic control, consistent with the
observation of increased diastereoselection at colder temper-
atures.7 In addition, these zinc carbenoid-mediated chain
extension reactions provide efficient access to targeted materials
involved in the synthesis of various natural products and natural
product analogues.9 Chain extensions of amino acid-derived β-
keto esters and β-keto amides have provided a number of
ketomethylene isostere-containing dipeptide analogues.10
The intent of this study is to probe the nature of the
common organometallic, reactive intermediate (i.e., 6) involved
in each of these above tandem reactions, with the goal of
understanding the syn-selectivity in the second stage aldol
reaction. Furthermore, insight into the nature of this common
reactive intermediate will be important for future practitioners
utilizing zinc reagents with a desired stereochemical outcome in
mind.
INTRODUCTION
■
A one-pot zinc carbenoid-mediated chain extension reaction in
which β-keto esters 1 are converted to γ-keto esters 2 was first
reported in 1997 (Scheme 1).1 Further studies of this chain
extension methodology have resulted in the chain extension of
other substrates such as β-keto phosphonates and β-keto
amides.2 The originally proposed mechanism for this reaction3
involved formation of the classical donor−acceptor (push−
pull) cyclopropane 4, which fragments and forms intermediate
6 (via 5), which on protonation provides chain-extended
products (i.e., 2). More recently, an extensive DFT study was
performed to gain a better understanding of the mechanistic
process, and an alternative pathway involving a cyclopropane
transition state was also identified,4 which suggests that the
presence of a cyclopropane intermediate (classical) and a
cyclopropane transition state (nonclassical) are both viable.
Taking advantage of this proposed organometallic inter-
mediate (6), a number of tandem reactions were subsequently
developed (Scheme 2). Treatment of intermediate 6 with
excess carbenoid in the presence of catalytic trimethylsilyl-
chloride results in formation of an intermediate homoenolate
(7), which can be quenched to provide α-methylated products
(8) (Path A, Scheme 2).5 Alternatively, trapping the organo-
metallic intermediate 6 with iodine, followed by elimination
with base, provides a one-pot route for the generation of α,β-
unsaturated γ-keto esters (10) (Path B, Scheme 2), which have
been utilized in natural product synthesis.6 A tandem chain
extension−aldol reaction, the primary focus of this report,
provides syn-selective aldol products (11) (average dr 10:1)7 in
equilibrium with the tetrahydrofuranyl hemiacetal forms (12)
(Path C, Scheme 2).8 The tandem chain extension−aldol
RESULTS AND DISCUSSION
■
In our earlier report, we suggested that the syn product of the
TCA reaction results from the reaction of a Z-enolate,7 with the
proposed preference for Z-enolate formation most likely related
to the interplay between the ketone carbonyl and the
organometallic intermediate (i.e., 6). This intermediate, which
functions as an enolate in the aldol reaction, appears to be
similar to the reactive intermediate found in the traditional
zinc-Reformatsky reaction (Scheme 3).11
The zinc-Reformatsky intermediate is generated typically by
zinc insertion into the carbon−halogen bond of a α-haloester
Received: March 13, 2012
Published: June 15, 2012
© 2012 American Chemical Society
5942
dx.doi.org/10.1021/jo3004925 | J. Org. Chem. 2012, 77, 5942−5955

The Journal of Organic Chemistry
Article
Scheme 1. Chain Extension Reaction of β-Keto Carbonyl
Scheme 3. Zinc-Reformatsky Reaction
Compounds
bound species 14 of the Reformatsky intermediate reacts with
the aldehyde through its monomeric form.12 It was further
suggested that the more energetically favorably pathway for the
reaction occurs via isomerization and a metallo-Claisen
rearrangement of the enolate, a reaction which involves a six-
electron, closed transition state 15.
Unless performed in the presence of chiral functionality,13
the zinc-Reformatsky reaction provides poor syn/anti diaster-
eoselectivity. This result stands in contrast to the tandem chain
extension−aldol (TCA) reaction, which provides good syn-
diastereocontrol (average dr 10:1), even though the only
apparent difference between the two reactions is the presence
of a γ-keto group in the proposed TCA intermediate (6). We
initially proposed that the TCA reaction was proceeding
through a monomeric form (via a metallo-Claisen rearrange-
ment) in which complexation between the keto-carbonyl and
Zn2+ biases the molecule toward the formation of a Z-enolate in
preparation for the aldol reaction;7 however, a recent DFT
study suggested that the structure of the TCA intermediate is
dimeric, similar to that proposed for a zinc-Reformatsky
reaction.7 Three possible dimeric structures (17a, 17b, and
17c) were calculated and are illustrated in Scheme 4.4
and has been extensively studied. X-ray crystallography and
ebulliometry in THF have shown that the intermediate is a
carbon-bound zinc organometallic structure, 14, and that it
exists as a dimer in both solution and its solid phase.11b
1
Observations made by H and 13C NMR spectroscopy in a
variety of coordinating solvents11a have also shown the
intermediate to possess a carbon-bound zinc. Using computa-
tional studies, Dewar et al. concluded that the carbon−zinc
a
Scheme 2. Tandem Processes for the Chain Extension Reaction
a
Path A: α-methylation. Path B: α-iodination. Path C: aldol reaction.
5943
dx.doi.org/10.1021/jo3004925 | J. Org. Chem. 2012, 77, 5942−5955

The Journal of Organic Chemistry
Article
Scheme 4. Proposed Structures for a Dimeric Form of the TCA Intermediate of β-Keto Esters
Scheme 5. Reformatsky Reactions of 20 and 21 and Tandem Chain Extension−Aldol Reaction of 25
We present in this report: (1) a more detailed DFT study of
the dimers in terms of stability and reactivity toward aldehydes;
(2) experimentally determined reactivity data of the TCA
intermediate as compared to the Reformatsky reaction; (3)
spectroscopic evidence of a zinc intermediate; as well as (4)
experimental and spectroscopic evidence that the interaction of
the ketone with zinc is essential for syn-selectivity.
TCA intermediate. The ligand on intermediate 6 in the TCA
reaction is unknown, since an ethyl group, an iodomethyl
group, or other species resulting from decomposition of the
zinc carbenoid could act as the ligand. Computational studies
(vida infra) were performed with the ethyl group acting as a
ligand due to the clear presence of ethylzinc moieties in NMR
spectra. Structures have been drawn accordingly.
Control Study. In order to probe the role of the γ-keto
group in the TCA reaction, control Reformatsky reactions were
performed with two α-iodoesters, one with a γ-keto-group 20
and the other without a γ-keto group 21 (Scheme 5).
Benzaldehyde was chosen as the aldehydic reaction partner
for the reactions carried out at 0 °C in methylene chloride,
which is the preferred solvent used for the zinc carbenoid-
mediated chain extension reaction. Zinc insertion into the
carbon−iodine bond of 20 and 21 was performed with
diethylzinc, while diiodomethane was employed as an additive.
The addition of diiodomethane in the control Reformatsky
studies was intended to ensure that the reaction environment in
the control study closely mimicked the ligand found on the
A Reformatsky reaction performed with α-iodo-γ-keto ester
20 resulted in the formation of products 22 and 23, with a
syn:anti ratio of 7:1. The importance of the added diiodo-
methane in this Reformatsky reaction was revealed through the
observation of poor diastereocontrol when the reaction was
performed in the absence of diiodomethane. Reaction of the
substrate (21) that lacked the keto-group provided products 24
and 25 with a syn:anti ratio of 1:1, even in the presence of
diiodomethane. These results were compared to the TCA
reaction of 26 and benzaldehyde, which was reported
previously and exhibited a syn:anti ratio of 12:1.7 It was
postulated that both the TCA reaction of 26 and the
Reformatsky reaction of 20 were proceeding through similar
5944
dx.doi.org/10.1021/jo3004925 | J. Org. Chem. 2012, 77, 5942−5955

The Journal of Organic Chemistry
Article
intermediates, a γ-keto α-carbon-bound zinc organometallic
ester 17, while the reaction with 21 proceeded through a
traditional Reformatsky intermediate 14. The results of this
control study strongly indicated that the γ-keto group of 20 was
essential for the syn-selectivity of the aldol reaction. In turn,
these results also suggested that the γ-keto group of the TCA
intermediate 17 plays a large role in the syn-selectivity observed
in the TCA reaction of β-keto esters like 26.
Scheme 6. Chain Extension Reactions of 26
NMR Study. Having verified a strong correlation between
the γ-keto group of the TCA intermediate and the syn-
1
selectivity of the aldol reaction, H and 13C NMR experiments
were performed in order to determine whether the TCA
intermediate (i.e., 6 or 17) was indeed a carbon−zinc bound
structure with a γ-keto-Zn2+ complexation. Initial attempts were
made to generate a TCA-like intermediate from 20 and
diethylzinc for NMR experiments (Figure 1). The intermediate
albeit not completely identical, to the intermediate of the TCA
reaction.
The methyl ester and t-butyl protons of the TCA
intermediate 28 could be easily assigned at 3.70 and 1.24
ppm, respectively (Figure 2a,b). The assignments for the α-
methine and β-methylene protons proved to be more of a
challenge; therefore, an experiment with α-deuterated methyl
pivaloylacetate 27 was employed (Figure 2b). The proposed
mechanism (Scheme 1) suggests chain extension of 27 would
provide an intermediate (29) in which deuterium is
incorporated α to the ester. The absence of a 1H NMR
resonance at 2.66 ppm when using the deuterated starting
material indicated this resonance corresponded to the methine
of the TCA intermediate 28. The methylene protons were
assigned as a broad resonance between 2.80 and 4.00 ppm
through the use of 2D NMR experiments. These broadened
resonances became more discernible in variable temperature
NMR investigations of the intermediate (Figure 3).
The broadness of the methine and methylene resonances was
proposed to result from a dynamic process in which
intermediate exchange was being observed on the NMR
time-scale at room temperature. Whether this was an
intermolecular or intramolecular dynamic process could not
be verified. A variable temperature study was employed in an
effort to approach slow exchange and determine the number of
the various intermediate species that existed. The lowest
temperature at which useful data could be obtained was −15
°C, a temperature at which the intermediate approached slow
exchange as the broad methylene resonance between 2.80−4.00
ppm resolved into two resonances at 3.71 and 2.97 ppm. The
methine resonance at 2.66 ppm also resolved into two
resonances at 2.71 and 2.58 ppm with an approximate integral
ratio of 4:1, respectively.
1
Figure 1. H NMR spectra of the intermediate from 20.
formed in this fashion would be formed in the absence of the
carbenoid and its decomposition products, thus making its
analysis easier. However, the intermediate generated by this
method was not amenable to extensive spectroscopic experi-
ments, as it converted to the γ-keto ester 3014 within a few
minutes of being formed, possibly through hydride delivery
from the ethyl group. In Figure 1, the solid arrows are used to
indicate the resonances of the two methylene units in 30.
Eventually, the TCA intermediate was generated (Scheme 6)
in CDCl3 by sequential treatment of methyl pivaloylacetate 26
or its deuterated analogue 27 with 2 equiv of diethylzinc and 2
equiv of diiodomethane. The intermediate formed in this
fashion was stable for up to 24 h, lending further support to the
importance of the iodomethyl group in the intermediate. All
1
spectra acquired from this intermediate were compared to H
and 13C NMR spectra of the corresponding γ-keto ester 30 and
to those reported in the literature for zinc-Reformatsky
intermediates.11
1
The H NMR spectrum of the intermediate generated by
treatment of methyl pivaloylacetate 26 with diethylzinc and
diiodomethane was similar to that of the intermediate
generated from 20, although the resonances occurred at
slightly different chemical shifts (Figure 1 vs Figure 2a). Most
notable was the presence of a broad resonance in both spectra
at 2.05 ppm in Figure 1 and at 2.66 ppm in Figure 2a. The
occurrence of the broad resonance, in addition to singlets for
the methyl ester, suggested that the intermediate generated in
the study of the Reformatsky reaction with 20 was similar,
At 0 and −15 °C, the signals at 3.71 and 2.91 ppm each
integrated for one proton relative to the sum of the integrals for
the methine protons at 2.71 and 2.58 ppm. In view of this
integral ratio, the decoalescence of the resonance for the
5945
dx.doi.org/10.1021/jo3004925 | J. Org. Chem. 2012, 77, 5942−5955

The Journal of Organic Chemistry
Article
1
Figure 2. Identification of the organometallic’s methine resonance. H NMR spectra of (a) 28; (b) 29; (c) 30.
methylene protons at 3.25 ppm (19 °C) into resonances at 3.71
and 2.91 ppm (0 °C), appeared to be a separation of signals for
diastereotopic methylene protons, as opposed to signals
representing different zinc-organometallic species. The decoa-
lescence of the methine resonances into two methine
resonances at low temperature suggests that at least two
distinct organometallic intermediates exist in solution. This
dynamic process was not reported in the NMR studies of the
Reformatsky intermediate,11 lending further support to the
importance of the ketone in the dynamic process. The
monomeric, dimeric, or oligomeric nature of these species is
unknown. No differences were observed in the NMR spectra of
the intermediate generated at different concentrations.
5946
dx.doi.org/10.1021/jo3004925 | J. Org. Chem. 2012, 77, 5942−5955

The Journal of Organic Chemistry
Article
Figure 3. Variable temperature NMR study of intermediate 28.
More insight into the structure of the TCA intermediate was
obtained with 13C NMR experiments (Figure 4). The 13C
NMR spectrum of this intermediate (28) exhibited chemical
shifts for the ketone and the ester carbonyls at 230.0 and 184.9
ppm, respectively. Importantly, both chemical shifts were
downfield of the corresponding functional groups in the γ-keto
ester 30, which resonated at 214.3 and at 173.7 ppm,
respectively. The downfield shift of the carbonyls’ resonances
suggests complexation of Zn2+ species in solution. Other
resonances corresponding to the TCA intermediate were
assigned through the use of DEPT and 2D NMR experiments
(Table 1).
The use of THF as an additive provided further evidence that
linked Zn2+ complexation of the TCA intermediate’s ketone
carbonyl to the 10:1 syn-selectivity of the TCA reaction. When
5.4 equiv of THF was added to the intermediate, the 13C NMR
5947
dx.doi.org/10.1021/jo3004925 | J. Org. Chem. 2012, 77, 5942−5955

The Journal of Organic Chemistry
Article
Figure 4. Effect of the zinc organometallic on the chemical shift of the ketone’s carbonyl. 13C NMR spectra of (a) 28; (b) 29; (c) 30.
spectrum exhibited a disappearance in the ketone’s carbonyl
resonance at 230 ppm (Figure 5). The resonance for the ester
carbonyl was unaffected by the presence of THF. This result
suggested that the ester functionality in the intermediate forms
a stable complex with Zn2+, which is unperturbed by the
presence of the THF donor. This stable complex involving the
ester functionality is reminiscent of the Reformatsky
intermediate, which is also unaffected by the presence of
donor solvents.10 Absence of the ketone’s resonance suggested
that THF may be competing with the ketone for the remaining
site on the tetracoordinate Zn2+. In theory, the uncomplexed
ketone should exhibit a resonance at approximately 214 ppm, a
chemical shift similar to that of the ketone resonance found in
30. The spectrum of the intermediate with added THF is
5948
dx.doi.org/10.1021/jo3004925 | J. Org. Chem. 2012, 77, 5942−5955

The Journal of Organic Chemistry
Article
Table 1. Nuclear Magnetic Resonance Spectroscopy of 28: Assignments for Chemical Shifts
1H NMR
13C NMR
28
30
28
30
nucleus
H1
δ
mult
δ
mult
s
nucleus
δ
δ
1.24
s
1.15
C1
C2
C3
C4
C5
C6
C7
26.7
44.1
26.7
44.1
230.0
214.3
28.2
H4
H5
2.80−4 0.00
broad
broad
2.80
2.55
t
t
38.5 (broad)
34.9 (broad)
184.9
2.66
31.7
173.7
51.9
H7
3.70
s
3.65
s
53.8
Table 2. Effects of THF Addition on the Stereoselectivity of
the TCA Reactions
[26]
additive
time/min
22:23 syn:anti
0.34
0.34
none
5.4 equiv of THF
60
60
7:1
3.3:1
reactions with benzaldehyde (Table 3). The syn:anti selectivity
of these quenches is dependent on the concentration of the
Table 3. Effect of Substrate Concentration on the
Stereoselectivity of the TCA Reactions
[26]
time/min
22:23 syn:anti
0.032
0.078
0.34
60
60
60
11:1
11:1
7:1
Figure 5. 13C NMR spectra of the carbonyl region of 28 (a) with 5.4
equiv of THF and (b) without THF.
substrate, in that stereoselectivity decreases as the concen-
tration of the substrate increases.
consistent with a dynamic relationship between a complexed
and uncomplexed ketone. Once again, this proposed Zn2+−
ketone complexation could be inter- or intramolecular in
nature.
Quenching of the TCA intermediate was performed with
benzaldehyde in the absence and presence of THF (Table 2). A
syn:anti (22:23) selectivity of 7:1 was observed when
benzaldehyde was added to a solution that lacked THF.
However, in the presence of 5.4 equiv of THF, an erosion of
the syn:anti ratio to 3.3:1 was observed, a ratio more consistent
with a typical Reformatsky reaction of a α-halo ester.13 This
study provides additional support to the hypothesis that Zn2+−
ketone complexation is involved in the syn-selectivity of the
TCA reaction.
The original study of the TCA reaction had been performed
with a substrate concentration of 0.08 M in methylene
chloride.7 The chain-extended aldol product derived from 26
and benzaldehyde at this concentration had a syn:anti ratio of
12:1. During the NMR investigation of the zinc-organometallic
intermediate (28), the concentration of methyl pivaloylacetate
was 0.34 M in CDCl3. While syn stereoselectivity on the order
of a 7:1 ratio was still observed, this selectivity was significantly
lower than the 12:1 ratio observed in the preparative-scale
study of Lai.7 We undertook an NMR study of the TCA
reaction at various substrate concentrations. No discernible
differences in the organometallic intermediate were observable
1
when comparing the H NMR and 13C NMR spectra acquired
Concentration and Stereoselectivity. An investigation of
the reaction concentration also provided insight into the TCA
at the concentrations of 0.34 and 0.08 M. However, upon
addition of benzaldehyde to the 0.08 M reaction mixture, a
5949
dx.doi.org/10.1021/jo3004925 | J. Org. Chem. 2012, 77, 5942−5955

The Journal of Organic Chemistry
Article
a
Scheme 7. Keto-Enol Tautomerization Step of the Dimeric Structure
a
Given numbers are Gibbs free energies relative to 17a and in kJ/mol.
syn:anti selectivity of 11:1 was observed. This diastereomeric
ratio was higher than that observed when the reaction was
performed at 0.34 M and was closer to the syn:anti ratio of 12:1
that was observed in the original preparative scale study. An
NMR tube experiment performed with a more dilute reaction
mixture (0.032 M) also provided the TCA products in a
syn:anti ratio of 11:1.
because of increased stabilization by the solvent. However, the
results show that the inclusion of solvent effects is mandatory
for understanding selectivity outcomes. For reduction in
calculation time, the iodomethyl residue resulting from the
Furukawa reagent as implemented in our earlier work has been
exchanged to an ethyl residue, as it is not expected to
participate in the aldol reaction mechanism. The reaction steps
leading up to the inserted carbene moiety in the TCA pathway
have been investigated thoroughly;4 therefore, the theoretical
study herein will focus specifically on the reaction sequence
starting from the chain-elongated intermediate.
First, the keto−enol tautomerization step, which theoretically
can yield either Z- or E-enolates (Scheme 7), was calculated.
The structure of the E-enolate (Scheme 7) could not be located
at all because of the small interatomic distance between the
carbon atom of the double bond and the zinc ion. As a result,
this structure immediately relaxes to a Z-enolate, and thus only
pathways following 17b have been calculated. In agreement
with the findings of the 1H NMR study, two stable
intermediates for the tautomerization step have been found:
structure 17a and the Z-enolate 17b, containing one enol and
one keto tautomer, are energetically similar. Thus, no
thermodynamic driving force is apparent for the reaction at
this point. However, it should be noted that gas phase
calculations predict 17b to be 11 kJ/mol more stable than 17a
(Supporting Information). The transition state (TS), which
proceeds along the carbon zinc bond, possesses an activation
barrier of 62 kJ/mol relative to 17a and thus is easily
surmountable. Furthermore, the predicted coordination
between the zinc and the ketone group is also supported by
these calculations. Additionally, a conceivable structure 17c
(Scheme 7) containing two enolate moieties could not be
located because of spontaneous relaxation to 17b. Thus only
reaction pathways starting from 17b have been taken into
account.
The diastereoselectivities observed at the various substrate
concentrations reveal a concentration dependence on the
selectivity of the TCA reaction, although the identical
diastereoselectivities of the reactions at 0.03 and 0.08 M
suggests a limit to substrate-concentration effects. Changing the
concentration may affect the extent to which the intermediate’s
monomers aggregate, although the similarities of the
intermediates’ NMR spectra for the organometallic intermedi-
ates generated at different concentrations suggests that the
fundamental structure of this aggregate is not greatly affected.
The more concentrated solution (0.34 M) would favor
increased intermolecular aggregation via the ketone carbonyl.
Since the ketone carbonyl has been implicated in the syn-
selectivity of the aldol reaction, enhanced aggregation involving
the ketone might be expected to negatively impact the
diastereoselectivity of the TCA reaction. Therefore, erosion in
the syn-selectivity when using more concentrated reaction
mixtures is consistent with the involvement of the ketone in the
diastereocontrol of the aldol reaction
DFT Study. NMR results described above reveal the
presence of an organometallic aggregate in the TCA reaction,
and previous calculations revealed that dimerization provides a
significant gain in free energy in comparison to the
correspondent monomeric structures.4 Therefore, the calcu-
lations presented in this body of work concentrate solely on
dimeric structures. To simulate the reaction conditions most
efficiently, the aldol reaction mechanism has been calculated
including C-PCM solvent effects resembling dichloromethane,
which is also the solvent used in the experimental
investigations. In comparison to the gas phase model, the C-
PCM calculations predict activation barriers to change
significantly and reaction energies to be on average 10 kJ/
mol lower (Table S1, Supporting Information). Major
structural changes could only be observed at certain Zn−C
or Zn−O coordination interactions, which became weaker
For conservation of computational time, but to still
encompass substitution effects, acetaldehyde was used to
evaluate the aldehyde addition reaction. Calculations demon-
strate that the reaction proceeds via a one-step six-membered
transition state, in which the oxygen of the acetaldeyde is
coordinated to one of the zinc ions of 17b. As illustrated in
Figure 6, these zinc ions have different coordination environ-
5950
dx.doi.org/10.1021/jo3004925 | J. Org. Chem. 2012, 77, 5942−5955

The Journal of Organic Chemistry
Article
energies. Only one pathway of the reaction, leading to the
favored syn product, is exergonic (Table 4, entry 5). The low
energy barriers indicate the potential for reversibility and
suggest that the reaction could be operating under thermody-
namic control; however, experimental results appear to support
kinetic control of the reaction.7
The reaction pathway with the lowest activation barrier
leading to syn intermediate (Table 4, entry 5) and the
equivalent pathway to anti (Table 4, entry 6) IM have been
chosen for detailed discussion (Figure 7). In the initial Z-
enolate, the enolate oxygen is still bound to ZnO but already
slightly complexed to ZnC. If acetaldehyde enters the dimer
complex and the reaction starts, the enolate oxygen changes its
coordination to ZnC in favor of substrate acetaldehyde
coordinating to ZnO (Figure 7, TS). In the transition state,
the C−C bond is formed simultaneously with the change in
coordination of oxygen from ZnC to ZnO. Thus in IM, the
oxygen of the former substrate acetaldehyde is bound to ZnO.
Obviously, the only difference between both pathways is the
configuration at the central carbon atom of the former substrate
acetaldehyde (green layer). Depending on this configuration,
the methyl group X points either to the outer (Table 4, entry 5,
syn) or inner sphere (Table 4, entry 6, anti) of the dimeric
system. Thus, steric repulsion increases activation barriers and
reaction energies of entry 6 (Table 4). Because of the sp3
hybridization of the former acetaldehyde carbon center and the
orientation of the methyl group X, the central carbon atom of
the former substrate acetaldehyde (green layer) can reach into
the inner sphere in entry 5 (Table 4). This conformation results
in a more relaxed CCO angle and thus in a more stable
structure of the dimeric system. In entry 6 (Table 4), the
former acetaldehyde forces the reacted monomer to point more
toward the inner sphere resulting in a significant increase of
steric tension. Furthermore, the methyl group Y (Figure 7) is
able to rotate about 180° in entry 5 (IM) (Table 4) because of
less steric repulsion.
Figure 6. Overview of available options for nucleophilic attack by 17b
on acetaldehyde.
ments, namely, carbon (right, ZnC, Figure 6) or oxygen (left,
ZnO) bound. In addition to these distinctions, the attack of the
aldehyde can occur from two different sides of the double
bound: (i) approach from the outer sphere (OUT, Figure 6), or
(ii) approach from the inner sphere (IN), which is sterically
encumbered by the other monomer. As shown below, a
transition state derived from the (IN) approach is not
automatically less stable. Furthermore, because of the unsym-
metrical substitution pattern of the two zinc atoms,
acetaldehyde can react through complexation with either zinc
atom via four reaction paths involving variations in the facial
selectivities of both the aldehydes and the enolate. (Figure 6).
The possible reaction products resulting from these pathway
combinations are syn (S,R; R,S) and anti (S,S; R,R) esters. The
configuration of the carbon derived from the aldehyde clearly
depends on the methyl position during the course of reaction,
whereas the configuration at the carbon of the former double
bond of the enolate 17b is related to the direction of aldehyde
approach (OUT or IN, Table 4).
Table 4. Reaction Free Energies Relative to the Product
Distribution in kJ/mol
Such impacts on the energetics discussed above can also be
observed in the case of OUT attack. However, the resulting
activation barriers and reaction energies are less favorable. In
the case of the OUT intermediate structures (Table 4, entries 7
and 8), the nonreactive keto monomer unit and the zinc-
coordinated sp2-hybridized ester group form a rigid ring
structure. In contrast, coordination of the OMe portion of
the ester group (Table 4, entries 1 and 2) in the IN
intermediates allows for conformational relaxation resulting in
a decrease in activation barriers and reaction energies of about
∼10 kJ/mol. Thus, energetic asymmetry is created, and syn
intermediate products are favored.
a
b
c
d
entry
attack possibilities
TS
IM
product
e
f
1
2
3
4
5
6
7
8
ZnC
IN
Si
86
78
88
85
62
77
84
69
5
3
R ,R
R,S
S,R
S,S
anti
Re
Si
syn
syn
ant
syn
anti
anti
syn
OUT
IN
17
Re
Si
17
ZnO
−14
33
S,R
S,S
Re
Si
OUT
43
R,R
R,S
Re
14
a
Reaction possibilities ordered by (i) coordinating zinc ion, (ii)
direction of aldehyde attack, (iii) attacking face of acetaldehyde.
Reaction free energies of the transition states (TS). Reaction free
energies of the intermediate states (IM), i.e., before quenching.
Configuration of quenched products. Configuration of the carbon
atom of the former double bond. Configuration of the carbon atom of
In the case of the ZnC derived intermediate structures,
entries 3 and 4 (Table 4) already form a pure enolate from the
nonreacting keto monomer unit by breaking the Zn−C bond.
In contrast, the still existent Zn−C interaction in entries 1 and
2 (Table 4) (illustrated in Figure 8) explains the significant
energy difference to the OUT intermediates (entries 3 and 4).
For steric reasons, arising mainly from lower angle of the sp3-
hybridized carbon atom of the former aldehyde (in contrast to
the flat sp2-hybridized carbon atom of the ester group), the
OUT products (Table 4, entries 3 and 4) are not able to form
this Zn−C bond. Facial selectivity of the acetaldehyde exerts no
significant influence, as both intermediate structures (Si, Re) of
each attack (OUT, IN) possess similar free energies. In contrast
to ZnO, the distinct positions of the methyl group are
energetically inert (Figure 8). Furthermore, exchange between
b
c
d
e
f
the former acetaldehyde.
Reflecting on the activation barriers (TS) given in Table 4,
aldehyde attack via the ZnO coordination mode is on average
slightly more favorable. That being said, all activation barriers
seem surmountable, although the lowest barriers are associated
with the syn products (Table 4, entries 5 and 8). Since the
products are not formed until after the reaction is quenched
following the aldehyde addition step, the energetics of the
reaction intermediates are key to understanding the stereo-
chemical course of the reaction, as opposed to product free
5951
dx.doi.org/10.1021/jo3004925 | J. Org. Chem. 2012, 77, 5942−5955

The Journal of Organic Chemistry
Article
Figure 7. Most probable reaction pathway yielding syn and its equivalent reaction to anti intermediate (entries 5 and 6, Table 4). For simplification,
hydrogens are not depicted. Gray, carbon; red, oxygen; blue, zinc. S, R: configuration at carbon centers; C, O: ZnC, ZnO. Green layer, acetaldehyde;
red layer, unreactive keto-configured monomer.
the asymmetric center. However, the calculated energies of
these enolate structures are 80−100 kJ/mol higher than the
keto form because of lost Zn−O coordination interactions.
Thus, the influence of such equilibrium effects on the syn:anti
ratio can be neglected.
Computational investigations of the reaction up to the point
of a protic quench reveal that the syn products would be
favored because of a sterically controlled kinetic preference
(Table 4, entry 5). The alkoxide intermediate derived from this
aldol pathway is also the most stable intermediate and is the
sole exergonic pathway as determined computationally. As the
experimental results show the formation of both syn and anti
aldol products, the intermediates reaction free energies of
entries 1 and 2 (Table 4) appear to be overestimated, but
exergonic values are still in the range of accuracy of the method
used. However, the calculations provide additional support for
the involvement of the ketone in biasing the reactivity of the
intermediate to favor formation of the syn isomer.
CONCLUSION
■
Results of the NMR study for the TCA reaction of methyl
pivaloylacetate 26 confirmed that the intermediate arising from
the chain extension reaction possesses a carbon-bound zinc
1
organometallic (28). The H NMR and 13C NMR chemical
Figure 8. Intermediate products (IM) of entries 1 and 2 (Table 4).
For simplification, hydrogens are not depicted. Gray, carbon; red,
shifts for this intermediate have been assigned with the use of
deuterium labeling experiments, DEPT and 2D experiments. A
control study involving traditional Reformatsky reactions
oxygen; blue, zinc. S, R: configuration at carbon centers; C, O: ZnC,
ZnO. Green layer: acetaldehyde.
confirmed that the syn-selectivity of the TCA reaction is
anti and syn configured species is formally conceivable by a
keto−enol tautomerization of the carboxyl group adjacent to
dependent upon the presence of a γ-keto group in the TCA
intermediate (6). Furthermore, studies performed in the
5952
dx.doi.org/10.1021/jo3004925 | J. Org. Chem. 2012, 77, 5942−5955

The Journal of Organic Chemistry
Article
methylene iodide (1.6 mL, 20.0 mmol) and then cooled in an ice−
water bath (0 °C). Diethylzinc (1 M) in hexanes (20 mL, 20.0 mmol)
was added, and the resulting white suspension was stirred for 10 min.
Methyl pivaloylacetate 26 (0.64 mL, 4.0 mmol) was added via syringe,
and the reaction was allowed to stir at 0 °C for 30 min, at which time
iodine (6.00 g, 23.7 mmol) was added to the mixture in a single
portion. After a pink color developed and persisted for more than 30 s,
the mixture was treated with a saturated aqueous solution of sodium
thiosulfate (40 mL). After the pink color disappeared, a concentrated
aqueous solution of ammonium chloride (40 mL) was added to the
stirring suspension. The resulting mixture was extracted twice with
diethyl ether (2 × 70 mL). The combined organic layers were dried
with sodium sulfate, filtered, and concentrated on a rotary evaporator.
The crude oil was purified by flash column chromatography (10%
ethyl acetate in hexanes) to yield methyl 2-iodo-5,5-dimethyl-4-oxo-
hexanoate 20 (1.028 g, 86%) as a yellow oil. Trace amounts of chain
extended material (methyl 5,5-dimethyl-4-oxo-hexanoate (30)) and
the α,β-unsaturated ester (methyl 5,5-dimethyl-4-oxo-hex-2-enoate)
were observed in the 1H NMR spectrum: 1H NMR (400 MHz,
CD2Cl2, referenced to TMS 0.00 ppm) δ 4.57 (dd, 1H, J = 9.9, 5.0
Hz), 3.63 (s, 3H), 3.43 (dd, 1H, J = 18.2, 9.9), 3.07 (dd, 1H, J = 18.2,
5.0), 1.05 (s, 9H); 13C NMR (100 MHz, referenced to CD2Cl2 at
53.61 ppm) δ 213.1, 171.8, 53.0, 44.6, 43.8, 26.2, 12.3; HRMS-ESI m/
z calcd for C9H16IO3 [M+ + H] 299.0144, found 299.0148.
Methyl 2-(Hydroxy-(phenylmethyl))-5,5-dimethyl-4-oxo-
hexanoate (22 and 23). An oven-dried round-bottomed flask,
equipped with a stir bar and N2 gas inlet, was charged with methyl 2-
iodo-5,5-dimethyl-4-oxo-hexanoate 20 (0.1489 g, 0.50 mmol) in
methylene chloride (2 mL) and cooled in an ice−water bath (0 °C).
The solution was treated with diethyl zinc (1 M) in hexanes (0.5 mL,
0.50 mmol), and after stirring for 10 min, the reaction mixture was
treated with methylene iodide (0.04 mL, 0.50 mmol). After stirring for
an additional 20 min, benzaldehyde (0.15 mL, 1.50 mmol) was added.
After 1 h, the reaction was quenched with saturated aqueous
ammonium chloride (5 mL) and extracted twice with ethyl acetate
(2 × 30 mL). The combined organic layers were dried with sodium
sulfate, decanted from sodium sulfate, and concentrated on rotary
evaporator. The oily residue was subjected to column chromatography
(5% ethyl acetate in hexanes) to yield a mixture of the aldol products,
2-(hydroxy-(phenylmethyl))-5,5-dimethyl-4-oxo-hexanoic acid methyl
ester 22 and 23 (0.030 g, 21%) as a yellow oil. The syn:anti ratio
resulting from the reaction, 7:1, was determined from integration of
the 1H NMR spectrum of the crude product mixture. The identities of
the products were confirmed by comparing NMR spectra to results
obtained by Lai et al.7 Methyl 5,5-dimethyl-4-oxo-hexanoate 30 was a
byproduct of this reaction.
Methyl 2-Iodo-5,5-dimethylhexanoate (21). An oven-dried
250-mL round-bottomed flask was equipped with a stir bar and
charged with methylene chloride (90 mL), 3,3-dimethylbutanol (1.36
g, 10.39 mmol), N-methylmorpholine N-oxide (1.404 g, 10.39 mmol),
and crushed 4 Å molecular sieves (5 g). The suspension was cooled to
0 °C and then treated with tetra-N-propylammonium peruthenate
(TPAP) (0.183 g, 0.52 mmol). The reaction was monitored by thin
layer chromatography and, upon completion, filtered under a vacuum
through a plug of silica and Celite. The filter cake was rinsed with THF
(200 mL), and the combined methylene chloride and THF-filtrates
containing 3,3-dimethylbutan-1-al was used in the next step.
An oven-dried round-bottomed flask, equipped with a stir bar, was
charged with a solution of trimethylphosphonoacetate (3.78 g, 20.78
mmol) in THF (100 mL) at 0 °C under N2 gas. This clear solution
was treated with n-butyl lithium (2.5 M) in hexanes (8.31 mL, 20.78
mmol). After 30 min, the THF/methylene chloride solution of 3,3-
dimethylbutan-1-al 32 was added to the reaction. The mixture was
allowed to warm to room temperature and monitored by TLC. After
completion of the reaction, approximately 1 h, the reaction was
quenched with saturated aqueous ammonium chloride (100 mL) and
extracted with twice with diethyl ether (2 × 100 mL). The organic
layer was dried over sodium sulfate, vacuum filtered through Celite
and concentrated on rotary evaporator. The crude oil was purified by
flash column chromatography (1% ethyl acetate in hexane, Rf = 0.14)
presence of tetrahydrofuran (THF), a donor molecule,
provided additional evidence that γ-keto-Zn2+ complexation is
associated with the observed syn-selectivity of the TCA
reaction. Finally, studies in which the concentration of the
reaction is varied suggest strongly that the syn-selectivity in the
aldol reaction can be enhanced by decreasing the intermo-
lecular complexation of the proposed organometallic dimer by
using dilute reaction mixtures.
DFT calculations revealed that a reaction pathway from a
dimeric species is probable and that the preference of the syn-
product is related to intrinsic steric repulsion effects resulting
from the different configurations involved in the aldehyde’s
reaction pathways. Additionally, the E-enolate was found not to
be stable because of the short interatomic distance between the
double bond and the zinc ion ZO. The selectivity is
predetermined by the asymmetric character of the Z-enolate
17b, which includes both an enol monomer unit and a keto
monomer. Two sides of the double bond (17b) are available for
approach by the electrophilic aldehyde. The inherent
asymmetry of the reacting aldehyde determines the lower
reaction free energy leading to a syn-product.
The above studies confirm the unique diastereocontrol in the
TCA reaction. A strategy for controlling zinc-aldol reactions
through the presence of an appropriately positioned Lewis base
can be inferred from these data. Additionally, this enhanced
understanding of the intermediate formed in the zinc
carbenoid-mediated chain extension reaction may facilitate
the development of additional stereocontrolled tandem
reactions.
EXPERIMENTAL SECTION
■
Preparative scale reactions were run in oven-dried glassware and
stirred with Teflon-coated stir-bars. NMR studies were performed
directly in NMR tubes. The terms concentrated in vacuo or under
reduced pressure refer to the use of a rotary evaporator or vacuum
pump.
Tetrahydrofuran (THF) was distilled from purple benzophenone
ketyl prior to use. Benzene was distilled from calcium hydride prior to
use. Methylene chloride (CH2Cl2) was distilled from P2O5 prior to
use. Ethyl acetate (EtOAc) and hexanes were distilled prior to use.
Pyridine was distilled from calcium hydride and stored over potassium
hydroxide. Triethylamine (Et3N) was distilled from potassium
hydroxide.
Diethylzinc was purchased both as a solution (1.0 M in hexanes)
and neat (as used for NMR studies). Methylene iodide (CH2I2) was
stabilized through the addition of nonoxidized copper wire. Iodine was
sublimed prior to use. Other reagents were used as received from
commercial sources.
Column chromatography was performed with flash silica gel (32−
63 μm). Mobile phases were used as noted. Thin layer
chromatography (TLC) was performed on glass plates and visualized
by UV and anisaldehyde stain.
Nuclear magnetic resonance (NMR) spectroscopy was performed
at 360.130 MHz for 1H nuclei and 90.55 MHz for 13C nuclei, at
1
399.130 MHz for H nuclei and 100.55 MHz for 13C nuclei, or at
1
499.766 MHz for H nuclei and 125.679 MHz for 13C nuclei. All 13C
were performed with 1H-decoupling. Unless otherwise noted, all NMR
experiments were carried out in deuterochloroform (CDCl3), which
was stored over 4 Å molecular sieves, or dideuteromethylene chloride
(CD2Cl2). All chemical shifts are reported in parts per million (ppm)
and referenced to the resonance for CDCl3 (1H NMR δ 7.26 ppm; 13
C
NMR δ 77.16 ( 0.20) ppm). Unless otherwise noted, DEPT spectra
are DEPT-135.
Methyl 2-Iodo-5,5-dimethyl-4-oxo-hexanoate (20). An oven-
dried 250-mL round-bottomed flask, equipped with a stir bar and a N2
gas inlet, was charged with methylene chloride (40 mL) and
5953
dx.doi.org/10.1021/jo3004925 | J. Org. Chem. 2012, 77, 5942−5955

The Journal of Organic Chemistry
Article
to yield a mixture of E and Z methyl 5,5-dimethylhex-2-enoate (0.962
g, 59%) as a colorless oil (E:Z; 8:1).
(d, 1H, J = 7.7 Hz), 3.68 (s, 3H), 3.57 (s, 3H), 2.96 (b, 1H), 2.80 (b,
1H), 2.73−2.63 (m, 2H), 1.85−1.48 (m, 4H), 1.39−1.22 (m, 2H),
1.02 (m, 2H), 0.83 (s, 9H), 0.78 (s, 9H); 13C NMR (100 MHz,
CDCl3) δ 175.9, 175.5, 142.1, 141.8, 128.6, 128.5, 128.1, 127.9, 126.5,
126.3, 75.4, 74.6, 53.9, 53.7, 51.8, 51.7, 41.8, 41.2, 30.4, 30.3, 29.3,
29.2, 24.8, 22.6; HRMS-ESI m/z calcd for C16H25O3 [M+ + H]
265.1804, found 265.1807
E-isomer: 1H NMR (400 MHz, CDCl3) δ 6.96 (dt, 1H, J = 15.6, 7.8
Hz), 5.77 (d, 1H, 15.6 Hz), 3.69 (s, 3H), 2.05 (dd, 2H, J = 7.8, 1.0
Hz), 0.89 (s, 9H); 13C NMR (400 MHz) δ 167.1, 147.3, 122.9, 51.5,
46.9, 29.5. Z-isomer: 1H NMR (400 MHz, CDCl3) δ 6.27 (dt, 1H, J =
10.4, 7.7 Hz), 5.81 (d, 1H, J = 10.4 Hz), 3.66 (s, 3H), 2.06 (dd, 2H, J =
7.7, 1.6 Hz), 0.89 (s, 9H); 13C NMR (400 MHz) δ 167.1, 148.1, 120.6,
51.5, 42.4, 31.5; HRMS-ESI (mixture of E and Z isomers) m/z calcd
for C9H17O2 [M+ + H] 157.1229, found 157.1226.
Zinc-Organometallic Intermediate (28). An NMR tube was
rinsed with acetone and hexane. After drying with a flow of N2 gas, the
NMR tube was fitted with a rubber septum and a N2 gas inlet. A
solution of methyl pivaloylacetate 26 (37 μL, 0.24 mmol) in CDCl3
(0.7 mL) was introduced into the NMR tube. After observing the clear
A round-bottomed flask, equipped with a stir bar, was charged with
a mixture of E and Z isomers of methyl 5,5-dimethyl hex-2-enoate
(0.962 g, 6.16 mmol) in THF (15 mL) and 10% Pd/C catalyst (0.096
g). The black suspension was stirred under a blanket of N2 gas for
approximately 10 min. With the N2 gas inlet still connected and the
flask under a a positive nitrogen pressure, the suspension was purged
with H2 gas via a balloon. After a positive flow of H2 gas through the
N2 gas bubbler was observed, the N2 gas inlet was removed, and the
reaction mixture was stirred with the balloon of H2 gas attached. After
18 h, the balloon was removed, and the suspension was filtered by
gravity filtration. The filtrate was concentrated on a rotary evaporator
to yield methyl 5,5-dimethylhexanoate (0.611 g, 63%) as a colorless
1
solution by H and 13C NMR spectroscopy, the mixture was again
placed under N2 gas and treated with neat diethyl zinc (50 μL, 0.48
mmol). With addition of diethyl zinc, the resulting mixture became
warm but soon cooled to room temperature. An additional seal of
parafilm treated with a drop of hexanes was wrapped around the
1
rubber septum. The zinc-enolate was observed by H and 13C NMR
spectroscopy.
Zinc-enolate of methyl pivaloylacetate: 1H NMR (400 MHz,
CDCl3) δ 5.13 (s, 1H), 3.69 (s, 3H), 1.17 (s, (9H), 1.14 (t, 9H, J = 8.1
Hz), 0.26 (q, 6H, J = 8.1 Hz); 13C NMR (100 MHz, CDCl3) δ 196.0,
1
oil: H NMR (400 MHz, CDCl3) δ 3.67 (s, 3H), 2.28 (t, 2H, J = 7.5
Hz), 1.59 (m, 2H), 1.19 (m, 2H), 0.89 (s, 9H); 13C NMR (400 MHz)
δ 174.4, 51.6, 43.7, 35.0, 30.4, 29.4, 20.4; HRMS-ESI m/z calcd for
C9H19O2 [M+ + H] 159.1385, found 159.1387.
1
175.4, 85.9, 51.7, 40.1, 28.4, 10.8, 3.9 [ethane: H NMR (400 MHz,
CDCl3) δ 0.85 (s); 13C NMR (100 MHz, CDCl3) δ 6.8]. The zinc-
enolate mixture was placed under N2 gas before treatment with
methylene iodide (39 μL, 0.48 mmol), the addition of which resulted
in slight effervescence and warming of the resulting pale yellow
mixture. The parafilm used to wrap the tube was sealed using a drop of
hexanes, and the pale yellow mixture was studied with various NMR-
experiments (1H NMR, 13C NMR, 2D-NMR, and D-NMR experi-
ments).
An oven-dried round-bottomed flask equipped with a stir bar was
charged with diisopropylamine (1.01 mL, 7.20 mmol) in THF (5 mL).
Under N2 gas, the solution was cooled in an ice−water bath (0 °C)
and treated with n-butyllithium (2.5 M) in hexanes (2.6 mL, 6.50
mmol). The resulting clear, pale yellow solution was allowed to stir for
30 min and then cooled to −78 °C. Methyl 5,5-dimethylhexanoate
(0.3845 g, 2.40 mmol) in THF (2 mL) was slowly added to the cooled
solution, and after 1 h, a single portion of iodine (1.65 g, 6.50 mmol)
was added to the enolate. After 18 h, the reaction was quenched with
concentrated sodium thiosulfate (15 mL) and allowed to stir until the
pink color disappeared. Concentrated aqueous ammonium chloride
(10 mL) was added to the resulting mixture, which was then extracted
twice with ethyl ether (2 × 50 mL). The combined organic layers were
dried over sodium sulfate, filtered, and concentrated on a rotary
evaporator to yield the crude product. The crude material was purified
by column chromatography (100% hexane) to yield methyl 2-iodo-5,5-
Zinc-organometallic intermediate (28): 1H NMR (400 MHz,
CDCl3) δ 3.70 (s, 3H), 3.25 (flat b, 2H), 2.66 (b, 1H), 1.24 (s,
9H); 13C NMR (100 MHz, CDCl3) δ 230.0, 184.9, 53.8, 44.1, 38.5,
34.9, 26.7; [ethyl iodide: 1H NMR (400 MHz, CDCl3) δ 3.17 (q, 2H, J
= 7.5 Hz), 1.82 (t, 3H, J = 7.5 Hz); 13C NMR (100 MHz, CDCl3) δ
20.7, −0.7]; [XZnCH2I: 1H NMR (400 MHz, CDCl3) δ 1.69 (b); 13
C
NMR (100 MHz, CDCl3) δ −12.6 (b)].16
Computational Methods. All calculations reported in this body
of work were performed using the program package Gaussian09,
Revision A.01.17 As the literature demonstrated good results,18 the
method consisting of the hybrid meta-GGA DFT functional M05−
2X19 combined with the 6-311+G(2d,2p) basis set as implemented in
Gaussian03 was used throughout. This basis set describes the second
row atoms by the McLean−Chandler basis,20 the basis set of McGrath
and Curtiss21 for third row atoms, and the Watchers−Hay22 basis set
for the first row of transition metals using the scaling factors of
Raghavachari and Trucks.23 As recommended when calculating
transition metals, diffuse and polarized functions were used. All
stationary points were characterized with a frequency analysis, where
minima must have no imaginary frequencies and the saddle points
have exactly one imaginary frequency. All given energies are free Gibbs
energies (in kJ/mol) relative to the free and separated educts, which
are zero point and thermally corrected. Therefore, all reported
energetic values refer to standard conditions such as 298 K and 1 atm
pressure.
Solvent corrected geometries and free energies were calculated with
C-PCM as implemented in Gaussian03.24 In this model, the species of
interest are embedded in a cavity of molecular shape surrounded by a
polarizable continuum, whose field modifies the energy and physical
properties of the solute. The solvent reaction field is described by
polarization charges distributed on the cavity surface. This procedure is
known to reproduce experimental solvation energies reasonably well.
Parameters for dichoromethane were chosen since this was the solvent
used for the experimental investigations.
1
dimethylhexanoate 21 (0.326 g, 48%) as a pale yellow oil: H NMR
(400 MHz, CDCl3) δ 4.24 (t, 1H, J = 7.6), 3.76 (s, 3H), 1.96 (m, 2H),
1.31 (m, 1H), 1.13 (m, 1H), 0.89 (s, 9H); 13C NMR (100 MHz) δ
172.2, 53.0, 43.7, 32.1, 30.5, 29.8, 21.6; HRMS-ESI m/z calcd for
C9H17INaO2 [M+ + Na] 307.0171, found 307.0167.
Methyl 2-(Hydroxy-phenyl-methyl)-5,5-dimethylhexanoate
(24 and 25). An oven-dried round-bottomed flask, equipped with a
stir bar, was charged with methyl 2-iodo-5,5-dimethyl-hexanoate 21
(0.1425 g, 0.50 mmol) in methylene chloride (5 mL) at 0 °C under N2
gas. The solution was treated with diethyl zinc (1 M) in hexanes (0.5
mL, 0.50 mmol), and after stirring for 10 min, the reaction mixture was
treated with methylene iodide (0.04 mL, 0.50 mmol). After stirring for
an additional 20 min, benzaldehyde (0.15 mL, 1.50 mmol) was added.
After 1 h, the reaction was quenched with saturated aqueous
ammonium chloride (5 mL) and extracted twice with ethyl acetate
(2 × 30 mL). The combined organic layer was dried with sodium
sulfate, decanted from sodium sulfate, and concentrated on a rotary
evaporator. The oily residue was subjected to column chromatography
(7.5% ethyl acetate in hexanes) to yield a mixture of the aldol
products, methyl 2-(hydroxy-phenyl-methyl)-5,5-dimethyl-hexanoate
24 and 25 (0.0524 g, 39%). Byproducts of this reaction included
methyl 2-iodomethyl-5,5-dimethyl-hexanoate and methyl 5,5-dime-
thylhexanoate 35. The syn:anti ratio of the aldol products, 1.6:1, was
determined from integration of the resonance for the benzylic methine
1
in the H NMR spectrum of the crude product. The assignments of
the syn and anti isomers were based upon vicinal coupling constants of
the resonances at 4.92 and 4.79 ppm.7,15 24 and 25: H NMR (400
1
MHz, CDCl3) δ 7.37−7.28 (m, 10H), 4.92 (d, 1H, J = 6.1 Hz), 4.79
5954
dx.doi.org/10.1021/jo3004925 | J. Org. Chem. 2012, 77, 5942−5955

The Journal of Organic Chemistry
Article
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Vreven, T.; Jr., J. A. M.; Peralta, J. E.; Ogliaro, F.;
Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.;
Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J.
C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene,
M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.;
Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.;
Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.;
Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
ASSOCIATED CONTENT
■
S
* Supporting Information
Calculated energies and geometries, and H and 13C NMR
spectra recorded in the preparation 20, 21, 24, 25, and 28. This
material is available free of charge via the Internet at http://
pubs.acs.org.
1
AUTHOR INFORMATION
Corresponding Author
*E-mail: ckz@cisunix.unh.edu.
Notes
■
̈
Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.;
Cioslowski, J.; Fox, D. J., Eds.; Gaussian, Inc.: Wallingford CT, 2009.
(18) (a) Zhao, Y.; Truhlar, D. G. Org. Lett. 2006, 8, 5753. (b) Zhao,
Y.; Truhlar, D. G. J. Phys. Chem. A 2006, 110, 5121. (c) Zhao, Y.;
Truhlar, D. G. Acc. Chem. Res. 2008, 41, 157. (d) Sorkin, A.; Truhlar,
D. G.; Amin, E. A. J. Chem. Theory Comput. 2009, 5, 1254.
(19) Zhao, Y.; Schultz, N. E.; Truhlar, D. G. J. Chem. Theory Comput.
2006, 2, 364.
(20) (a) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem.
Phys. 1980, 72, 650. (b) McLean, A. D.; Chandler, G. S. J. Chem. Phys.
1980, 72, 5639.
(21) (a) Binning, R. C. J.; Curtiss, L. A. J. Comput. Chem. 1990, 11,
1206. (b) McGrath, M. P.; Radom, L. J. Chem. Phys. 1991, 94, 511.
(c) Curtiss, L. A.; McGrath, M. P.; Blaudeau, J.-P.; Davis, N. E.;
Robert, C.; Binning, J.; Radom, L. J. Chem. Phys. 1995, 103, 6104.
(22) (a) Wachters, A. J. H. J. Chem. Phys. 1970, 52, 1033. (b) Hay, P.
J. J. Chem. Phys. 1977, 66, 4377.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We acknowledge Professor Gary Weisman and Kathleen
Gallagher for advice and assistance with the NMR studies.
We also acknowledge NIH (R15 GM060967) and the Pfizer
Fellowship Supporting Diversity in Organic Chemistry for
support of this research. The authors thank The University of
Queensland for financial support.
REFERENCES
■
(1) Brogan, J. B.; Zercher, C. K. J. Org. Chem. 1997, 62, 6444.
(2) (a) Verbicky, C. A.; Zercher, C. K. J. Org. Chem. 2000, 65, 5615.
(b) Hilgenkamp, R.; Zercher, C. K. Tetrahedron 2001, 57, 8793.
(3) Reissig, H.-U.; Zimmer, R. Chem. Rev. 2003, 103, 1151.
(4) Eger, W. A.; Zercher, C. K.; Williams, C. M. J. Org. Chem. 2010,
75, 7322.
(23) Raghavachari, K.; Trucks, G. W. J. Chem. Phys. 1989, 91, 1062.
(24) (a) Klamt, A.; Schuurmann, G. J. Chem. Soc., Perkin Trans. 2
̈
̈
1993, 799. (b) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995.
(c) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. J. Comput. Chem.
2003, 24, 669.
(5) Hilgenkamp, R.; Zercher, C. K. Org. Lett. 2001, 3, 3037.
(6) Ronsheim, M. D.; Zercher, C. K. J. Org. Chem. 2003, 68, 1878.
(7) Lai, S.; Zercher, C. K.; Jasinski, J. P.; Reid, S. N.; Staples, R. J. Org.
Lett. 2001, 3, 4169.
(8) (a) Bruckner, C.; Lorey, H.; Reißig, H.-U. Angew. Chem. 1986,
̈
98, 559. (b) Reißig, H.-U.; Reichelt, I.; Lorey, H. Liebigs Ann. Chem.
1986, 1986, 1924.
(9) (a) Heim, R.; Wiedemann, S.; Williams, C. M.; Bernhardt, P. V.
Org. Lett. 2005, 7, 1327. (b) Tilly, D. P.; Williams, C. M.; Bernhardt, P.
V. Org. Lett. 2005, 7, 5155. (c) Schwartz, B. D.; Tilly, D. P.; Heim, R.;
Wiedemann, S.; Williams, C. M.; Bernhardt, P. V. Eur. J. Org. Chem.
2006, 2006, 3181.
(10) (a) Theberge, C. R.; Zercher, C. K. Tetrahedron 2003, 59, 1521.
(b) Tryder, N. M.S. Thesis, University of Hampshire, 2003; (c) Lin,
W.; Tryder, N.; Su, F.; Zercher, C. K.; Jasinski, J. P.; Butcher, R. J. J.
Org. Chem. 2006, 71, 8140.
(11) (a) Orsini, F.; Pelizzoni, F.; Ricca, G. Tetrahedron Lett. 1982, 23,
3945. (b) Dekker, J.; Boersma, J.; van der Kerk, G. J. M. J. Chem. Soc.,
Chem. Commun. 1983, 553.
(12) Dewar, M. J. S.; Metz, K. M., Jr. J. Am. Chem. Soc. 1987, 109,
6553.
(13) (a) Heathcock, C. H. In Asymmetric Synthesis; Morrison, J. D.,
Ed.; Academic Press: New York, 1984; Vol. 3, p 111; (b) Tagliavini,
E.; Trombini, C.; Umani-Ronchini, A. In Advances in Carbanion
Chemistry; Snieckus, V., Ed.; JAI Press Inc.: Greenwich, 1996; Vol. 2, p
111.
(14) Ronsheim, M. D.; Hilgenkamp, R.; Zercher, C. K. Org. Synth.
2002, 79, 146.
(15) (a) Stiles, M.; Winkler, R. R.; Chang, Y.; Trayner, L. J. Am.
Chem. Soc. 1964, 86, 3337. (b) House, H. O.; Crumrine, D. S.;
Teranishi, A. Y.; Olmstead, H. D. J. Am. Chem. Soc. 1973, 95, 3310.
(16) Charette, A. B.; Marcoux, J.-F. J. Am. Chem. Soc. 1996, 118,
4539.
(17) Gaussian 09, Revision A.01; Frisch, M. J.; Trucks, G. W.;
Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.;
Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji,
H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.;
Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.;
5955
dx.doi.org/10.1021/jo3004925 | J. Org. Chem. 2012, 77, 5942−5955