Organocatalytic enantioselective α-hydroxymethylation of aldehydes: Mechanistic aspects and optimization

Article abstract of DOI:10.1021/acs.joc.5b00380

Further studies of the direct enantioselective α-hydroxymethylation of aldehydes employing the α,α-diarylprolinol trimethylsilyl ether class of organocatalysts are described. This process has proven efficient for access to β-hydroxycarboxylic acids and δ-hydroxy-α,β-unsaturated esters from aldehydes in generally good yields, excellent enantioselectivity, and compatibility with a broad range of functional groups in the aldehyde. The goal of these studies was to identify the critical reaction variables that influence the yield and enantioselectivity of the α-hydroxymethylation process such as catalyst structure, pH of the medium, purity of the reactants and reagents particularly with respect to the presence of acidic impurities, and the nature of the buffer, along with the standard variables including solvent, time, temperature and mixing efficiency. The previously identified intermediate lactol has been further characterized and its reactivity examined. These studies have led to identification of the most critical variables translating directly into improved substrate scope, reproducibility, enantioselectivity, and yields.

Full text of DOI:10.1021/acs.joc.5b00380

Article
pubs.acs.org/joc
Organocatalytic Enantioselective α‑Hydroxymethylation of
Aldehydes: Mechanistic Aspects and Optimization
Robert K. Boeckman, Jr.,* Kyle F. Biegasiewicz, Douglas J. Tusch, and John R. Miller
Department of Chemistry, University of Rochester, Rochester, New York 14627-0216, United States
S
* Supporting Information
ABSTRACT: Further studies of the direct enantioselective α-hydroxymethylation of aldehydes employing the α,α-diarylprolinol
trimethylsilyl ether class of organocatalysts are described. This process has proven efficient for access to β-hydroxycarboxylic
acids and δ-hydroxy-α,β-unsaturated esters from aldehydes in generally good yields, excellent enantioselectivity, and compatibility
with a broad range of functional groups in the aldehyde. The goal of these studies was to identify the critical reaction variables
that influence the yield and enantioselectivity of the α-hydroxymethylation process such as catalyst structure, pH of the medium,
purity of the reactants and reagents particularly with respect to the presence of acidic impurities, and the nature of the buffer,
along with the standard variables including solvent, time, temperature and mixing efficiency. The previously identified
intermediate lactol has been further characterized and its reactivity examined. These studies have led to identification of the most
critical variables translating directly into improved substrate scope, reproducibility, enantioselectivity, and yields.
up. As a result, a single step, readily scalable method for the
generation of these synthons would be highly desirable.
INTRODUCTION
■
The efficient construction of complex molecules from simple
and readily available starting materials remains a focal point in
organic synthesis. As a result, the utilization of chiral building
blocks at early stages in a synthesis is often desirable in an effort
to develop a highly branched and thus more efficient approach
to the desired target. Among this array of building blocks, α-
substituted β-hydroxyaldehydes and carboxylic acids have
demonstrated their utility in the synthesis of a variety of
complex molecules.1−4
The most common routes for the preparation of α-
substituted β-hydroxyaldehydes and carboxylic acids have
utilized chiral auxiliary based enolate chemistry such as those
originally developed by Evans.5,6 While these methods are
effective in furnishing the desired molecules in high yield and
stereoselectivity, the process usually requires multiple synthetic
operations including the introduction and removal of the
auxiliary utilized as well as the employment of an equivalent of
formaldehyde in gaseous form or the formaldehyde equivalent,
benzyl chloromethyl ether, as the alkylating reagent. In
addition, these methods require recycling of stoichiometric
amounts of chiral auxiliary limiting their practicality upon scale-
Over the course of the past decade, synthetic organic
chemists have begun to address the aforementioned limitations
of auxiliary based methods through applications of catalysis,
including organocatalysis.7−11 These methods have become
particularly useful for α-functionalization reactions of aldehydes
catalyzed by chiral secondary amines, and have been shown to
be effective for a broad range of electrophiles and nucleophiles.
The advantage of this approach is that one can enantiose-
lectively generate a new stereocenter at the α-position of an
aldehyde in a single step using only a catalytic quantity of the
chiral catalyst under mild, often near neutral conditions.
Furthermore, unlike some auxiliary based methods, the chiral
catalyst can often be recovered during the workup and reused.
One of the most prominent applications of organocatalysis,
beginning with the pioneering work by List, Lerner, and
Barbas,12 has been to intermolecular aldol reactions.13−15
However, in the case of intermolecular crossed aldol
reactions employing the powerful C1 synthon formaldehyde,
Received: February 17, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
only a limited number of such transformations have been
reported, primarily employing transition metal based Lewis acid
catalysts. Ito and co-workers performed an α-hydroxymethyla-
tion on 2-cyanopropionate derivatives with formaldehyde
catalyzed by a Rh(I)-TRAP complex.16,17 Sodeoka and co-
workers described an asymmetric α-hydroxymethylation of β-
keto esters with formaldehyde catalyzed by a Pd(II)-BINAP
complex,18 while similarly Shibasaki and co-workers developed
a related reaction employing a Ni(II)-Schiff base complex as
catalyst.19 In addition, α-hydroxymethylation has been
performed on silyl enol ethers by Yamamoto and co-workers
mediated by a AgOTf-BINAP complex,20 and by Kobayashi and
co-workers employing a chiral scandium complex.21 Although
these methods provided useful levels of enantioselectivity, we
sought to pursue a simpler, more practical and scalable method
for the enantioselective α-hydroxymethylation of aldehydes.
Others workers also realized the potential utility of such an
enantioselective organocatalytic crossed aldol reaction with
RESULTS AND DISCUSSION
■
Establishing A Feasible Catalytic Cycle. The originally
postulated mechanism of the α-hydroxymethylation is common
to other organocatalytic aldol reactions.12−15 Condensation of
the diaryltrialkylsilyloxy prolinol catalyst 1 with an aldehyde
generates an equilibrium concentration of enamine 2.
Subsequent attack of enamine 2 on formaldehyde occurs
from the less hindered si face of the enamine, as the re face is
blocked by nonbonding steric interactions from the sterically
demanding α-substituent of the catalyst 1. Hydrolysis of the
resulting iminium ion (3) results in α-hydroxymethylated
product 4 (Scheme 1).
Scheme 1. Postulated Catalytic Cycle for Enantioselective α-
Hydroxymethylation Catalyzed by 1
́
formaldehyde. In 2004, Cordova and co-workers described the
use of L-proline to catalyze the enantioselective α-hydrox-
ymethylation of a limited number of aliphatic aldehydes in
moderate yields and reportedly exceptionally high enantiose-
lectivity (>99%).22 However, we and others have been unable
to reproduce the reported levels of enantioselectivity. In our
hands, isovaleraldehyde reproducibly afforded only 70% ee
under the reported reaction conditions. In 2006, Pihko and co-
workers reported an organocatalytic α-methylenation of
aldehydes under catalysis by amine salts including chiral
amine salts of acids.23 This study may have been initially
targeting α-hydroxymethylation, but that result could not be
obtained without concomitant elimination of water under the
acidic conditions employed.
In 2009, our laboratory reported an alternate protocol
catalyzed by α,α-diphenylprolinol trimethylsilyl ether.24 This
class of catalysts, first reported by Jørgensen25 and Hiyashi,26
has been utilized for a variety of α-functionalization reactions of
aldehydes.27−30 Our studies defined an experimental protocol
for enantioselective α-hydroxymethylation which was demon-
strated to be readily scalable and general across a number of
functionalized aldehydes. More recently, Hayashi and co-
workers have reported a related enantioselective α-hydrox-
ymethylation of aldehydes catalyzed by the related 3,5-
bistrifluoromethylaryl prolinol in comparable yields and
enantioselectivity, but with the opposite sense of asymmetric
induction.31 In contrast to our protocol, where the
enantioselectivtiy is apparently controlled by the steric bulk
of the diaryl- and trialkylsilyloxy- substituent of the prolinol, the
To investigate the feasibility of the foregoing catalytic cycle,
we began with the reaction of the preformed enamine 5,
derived from pyrrolidine and isovaleraldehyde with varying
amounts of 37% aq formaldehyde (formalin).32 Our prelimi-
nary experiments demonstrated that use of ∼3.0 equiv of
formalin in a variety of solvents followed by the reduction of
crude products with sodium borohydride (NaBH4) provided
substantial amounts of allylic alcohol 633 rather than the desired
α-hydroxymethylation product 722 accompanied by trace
amounts of alcohol 8 derived from the self-condensation of
isovaleraldehyde followed by dehydration (1.5:1:trace of
6:7:8).34 However, we noted variability in the product
distribution depending on the sample of commercial formalin
used. Given the known propensity of formaldehyde to undergo
oxidation to formic acid over time, we measured the pH of the
samples of commercial formalin used and found that they
ranged from pH 3−5. We thus postulated that the formation of
the undesired elimination product 6 was due to the presence of
formic acid. Indeed, Pikho and co-workers reported a similar
observation during their studies of the α-methylenation of
aldehydes.23 Upon the basis of this observation, an attempt was
made to buffer the reaction medium. Gratifyingly, we
discovered that buffering the reaction medium with commer-
cially available solid pH 7 phosphate buffer almost completely
suppressed the formation of elimination product 6 and gave
́
catalyst systems employed by Cordova and Hayashi apparently
impart enantioselectivity through hydrogen bonding with the
carboxylic acid or alcohol functionality of the catalyst serving to
orient the formaldehyde electrophile syn to the α substituent of
the proline in the transition state.32,33
To better understand the controlling reaction variables of our
protocol, we have undertaken a more in-depth study of the
these variables, such as reactant and catalyst purity, catalyst
structure, pH of the reaction medium, source of formaldehyde,
methods of buffering, solvent polarity, and stirring rates. We
have also further characterized the intermediate lactol, and
studied the chemical and configurational stability and reactivity
of this intermediate.24
B
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
predominantly the desired diol 7 (1:12 6:7) as shown in
Scheme 2.
It should be noted that this route is readily scalable and
produces material of equal spectroscopic purity (1H NMR, 13
C
NMR) and physical appearance in comparison to the
commercial catalyst (see Figure 1) and has been broadly
Scheme 2. Preliminary Experiments with Enamine 5
With a workable source of formaldehyde identified, we then
focused on optimizing the concentration of the reaction. We
determined that running the reaction at a concentration of 0.5
M completely suppressed the formation of dimer 8. While the
desired α-hydroxymethylation occurred in a variety of different
solvents, we observed significantly better yields with nonpolar
solvents that were immiscible with formalin including toluene,
hexane, DCM, and chloroform. We were pleased to find that
using a catalytic amount of pyrrolidine (20−30%) the desired
α-hydroxymethylation occurred in the biphasic medium in ∼12
h affording the expected diol in good yield (after reduction with
NaBH4). Attempts to decrease the catalyst loading, while
successful, led to inconveniently long reaction times in these
initial experiments.
Once a catalytic cycle for α-hydroxymethylation was
demonstrated with pyrrolidine, we then set out to explore the
asymmetric variant. Our chosen catalyst class was the
Jørgensen−Hayashi diarylprolinol derived catalysts. Although
commercial sources of some of these catalysts are available, we
found that the commercial samples afforded inconsistent results
as the result of the presence of unknown impurities, which were
difficult to remove in our hands (vide infra). Thus, we chose to
prepare these materials to ensure reproducibility.
Our catalyst preparation, shown in the context of the
synthesis of (S)-2-(diphenyl((trimethylsilyl)oxy)methyl)-
pyrrolidine (13), began with esterification of commercially
available N-Boc-L-proline (9) producing the corresponding
methyl ester 10. The addition of two equivalents of
phenylmagnesium bromide (PhMgBr) reagent in THF to
ester 10 provided N-Boc-L-diphenylprolinol 11. The Boc group
was then removed by treatment with sodium hydroxide
(NaOH) in ethanol (EtOH) at reflux to give diphenylprolinol
12. Finally, protection of the hydroxyl function with
trimethylsilyl triflate afforded target catalyst 13 (Scheme 3).
Figure 1. Commercial vs synthetic (S)-2-(diphenyl((trimethylsilyl)-
oxy)methyl)pyrrolidine (25 g scale batch).
applied to synthesize all derivatives used in our study. While all
steps are high yielding and the route can be performed without
purification until after silylation, we found that it is critical to
recrystallize prolinol intermediate 12 prior to silylation in order
to ensure high yields and enantioselectivities in the ensuing α-
hydroxymethylation reactions.
Establishing An Enantioselective Catalytic Cycle. We
began our investigation of the asymmetric variant of the α-
hydroxymethylation process by initially investigating the
reaction of isovaleraldehyde 14 with diarylprolinol silyl ether
13 in a biphasic mixture of toluene and ∼3.0 equiv of formalin
containing a commercial solid phosphate buffer (Scheme 4).
Quite unexpectedly, we obtained what we later determined to
be the cyclic lactol 15 derived from the expected β-
hydroxyaldehyde by incorporation of an additional equivalent
of formaldehyde. As we reported earlier, we attributed the
excellent enantioselectivity observed (vide infra) to the rapid
formation of 15. We surmised that closure to the cyclic acetal
may serve to prevent potential racemization of the newly
generated stereocenter via one or more of several conceivable
pathways.24
Initial attempts to handle and purify lactol 15 did not prove
fruitful, thus we concluded that either protection of the lactol or
transformation to a more stable derivative by, for example,
oxidation would readily provide configurationally stable,
isolable materials suitable for further transformations. The
structure of the intermediate lactol 15 was originally inferred
from the formation of the silyl protected aldehyde 16 upon
treatment of crude lactol with TBSCl/imidazole. A subsequent,
Scheme 3. Preparation of the Jørgensen−Hayashi Prolinol Catalysts
C
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
substrate, we observed that original catalyst 13 (Ar = Ph, R′ =
CH3) was superior in providing the desired β-hydroxy acid 18
in excellent yield and enantioselectivity (94% yield, 94% ee).
The trends discernible from the data in Table 1 demonstrate
that as the size of the aryl group and/or the silyl protecting
group increases, it generally leads to a more sluggish reaction.
Furthermore, we found that increasing the electron density in
the aryl rings was tolerated, whereas utilizing catalyst 22 (Ar =
3,5-(CF3)2Ph, R′ = CH3) bearing electron deficient aryl groups
(Table 1, entry 4), resulted in no conversion. We postulate that,
in this case, the catalyst is sufficiently nonbasic and/or non-
nucleophilic that either the rate of enamine formation is slowed
or the equilibrium concentration of the intermediate enamine is
too low to sustain a reaction rate observable over the time scale
employed. Alternatively, it is also possible that the enamine
intermediate generated from condensation with 14 is
insufficiently nucleophilic to perform α-hydroxymethylation at
an observable rate on the time scale employed. However, the
successful application of the free alcohol analogue of this
catalyst to enantioselective α-hydroxymethylation by Hayashi
suggests that the former rather than the latter rationale may be
more plausible.31 It should also be noted that the reaction was
run with a commercial sample of catalyst 13 (Table 1, entry 12)
and the results were comparable in this case (89% yield, 91%
ee). However, for this case, in our hands there was significant
variability in yield and enantioselectivity, sample to sample, for
different commercial samples of catalyst 13.
We also examined the effects of catalyst loading. We were
pleased to see the reaction perform with consistent yield and
enantioselectivity at as low as 5 mol % catalyst loading at
constant reactant concentration (0.5 M of substrate in toluene)
although at the expense of ∼4−5-fold increase in the time
required for full conversion.
Optimization Studies: The Solvent. Having established a
reliable set of partially optimized biphasic reaction conditions,
we next examined whether the nature of organic solvent
affected the observed enantioselectivities and yield. The results
of these solvent screening studies are outlined in Table 2.
Scheme 4. Enantioselective Hydroxymethylation of
Isovaleraldehyde (14)
more detailed examination of the silylation reaction revealed
that the product is actually a mixture of silyl protected aldehyde
16 and the corresponding TBS protected lactol 17, which have
proven challenging to separate via a variety of purification
methods. As was previously reported, the resulting silyl
protected aldehyde 16 proved too sensitive to racemization
to allow purification by flash chromatography.24 However, the
mixture of 16 and 17 was obtained pure enough for further use
of the protected aldehyde as a synthetic equivalent of chiral α-
hydroxymethyl isovaleraldehyde.24 Alternatively, Pinnick oxida-
tion ((NaClO2, NaH2PO4, 2-methyl-2-butene) of the inter-
mediate lactol 15 afforded desired (R)-2-(hydroxymethyl)-3-
methylbutanoic acid (18) in excellent yield (94%) and
enantioselectivity (94% ee) as depicted in Scheme 4.
Optimization Studies: Catalyst Structure and Load-
ing. We initiated our optimization efforts by examining the
effects of modifying the steric and electronic properties of
diphenylprolinol catalyst 13 that was employed in our initial
studies. We screened a series of derivatives of 13 with varying
aryl groups in the catalytic α-hydroxymethylation of isovaler-
aldehyde (14) under identical reaction conditions. For this
Table 1. Screening Catalysts for α-Hydroxymethylation
entry
cat #
-Ar
-SiR3
mol % cat
yield (%)
ee (%)
1
19
20
21
22
23
13
24
25
13
13
13
4-(OMe)Ph
4-(F)Ph
3,5-(Me)2Ph
3,5-(CF3)2Ph
2-naphthyl
Ph
-SiMe3
-SiMe3
-SiMe3
-SiMe3
-SiMe3
-SiMe3
-SiEt3
30
30
30
30
30
30
30
30
20
10
5
75
81
72
n.r.
61
94
86
69
89
91
95
89
89
95
93
n.r.
93
94
94
96
93
95
92
91
2
3
4
5
6
7
Ph
8
Ph
-SiMePh2
-SiMe3
-SiMe3
-SiMe3
-SiMe3
9
-Ph
10
11
12
-Ph
-Ph
a
13
-Ph
30
a
Commercial sample of catalyst.
D
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 2. Solvent Effects on α-Hydroxymethylation Yield and Enantioselectivity
entry
solvent
yield (%)
ee (%)
1
2
3
4
5
6
7
8
toluene
hexanes
EtOAc
Et2O
91
69
59
63
44
38
38
n.r.
94
97
86
93
70
87
87
n.r.
THF
CH2Cl2
CHCl3
DMF
Scheme 5. Evaluation of Formalin vs Paraformaldehyde Using the Optimal pH 7 Buffer
As can be seen from these results, we observed that the
reaction generally performed well in nonpolar solvents
(hexanes, toluene) as well as in ethyl acetate (EtOAc). It
should be noted that the lower yields obtained in hexanes and
EtOAc can be primarily attributable to slower conversion
(∼60−70% after 15 h) and that, for all these solvents, only
traces of elimination products were observed. However, we
found that chlorocarbon solvents (DCM, and chloroform), still
afforded high enantioselectivity (87% ee); however, the yield of
α-hydroxymethylation product suffered owing to byproduct
formation via competing acid catalyzed elimination.
When THF was employed as solvent, large quantities of
elimination product along with a major decrease in
enantioselectivity (70% ee) were observed. The polarity of
THF is not markedly different from the solvents that afforded
high enantioselectivity, thus we tentatively attribute these
effects to the increased miscibility of THF and formalin.
Optimization Studies: The Source of Formalin and the
Nature of the Buffer. Thus far in our studies, the primary
competing process observed has been elimination of water from
the primary product, the β-hydroxy aldehyde, or a related
intermediate. This issue was identified during our original
study.24 At that time, we had postulated that the elimination
was occurring as a result of the presence of acid in the reaction
medium, likely from variable amounts of formic acid present in
the commercial samples of formalin employed. To overcome
this problem, the α-hydroxymethylation reaction was per-
formed in the presence of commercial pH 7 buffer salts, a
mixture of dibasic sodium phosphate and monobasic potassium
phosphate (Certified pH 7.00
0.02 at 25 °C). This
modification immediately afforded dramatically improved
results for isovaleraldehyde (14), our test case, affording 94%
yield, and 94% enantioselectivity, with no formation of the
byproduct from competing elimination (Scheme 5). While we
were delighted with this result, during our further studies it was
found that this buffering procedure was not entirely reliable or
broadly applicable across a broad range of structurally varied
aldehyde substrates.
Thus, we felt it was necessary to examine the effects of the
choice of the buffer salts on the yield and enantioselectivity of
the α-hydroxymethylation process. We performed the same
series of experiments with different freshly prepared pH 7
buffer salts. These included buffers derived only from sodium
salts (monobasic sodium phosphate and dibasic sodium
phosphate), the commerical mixtures of sodium and potassium
salts, and only potassium salts (monobasic potassium
phosphate and dibasic potassium phosphate). The results
observed for the buffer containing only sodium salts mirrored
those of the mixtures of sodium and potassium salts present in
commercial as well as freshly prepared buffers comprising
sodium/potassium salt mixtures. The results were not entirely
reproducible or general across a range of aldehyde substrates.
However, when a freshly prepared mixture of pH 7 buffer salts
containing only potassium salts was employed, it was found that
the reaction not only performed well with the isovaleraldehyde
(14) test case, but across a wide variety of substrates (Scheme
5) as will be discussed in more detail below. We attribute these
observations to the observed increase in solubility of the pH 7
E
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 3. Additive Effects in the Production of Lactol 15
entry
additive (mol %)
buffer salts
ratio of products (15:26)
1
2
3
4
5
6
NaOAc (10)
Na/K
Na/K
Na/K
Na/K
K/K
99:1
50:50
1:99
95:5
1:99
99:1
AcOH (10)
isovaleric acid (10)
isovaleric acid (0.1)
isovaleric acid (10)
isovaleric acid (0.1)
K/K
Scheme 6. Racemization, Elimination, or Loss of Formaldeyde from Lactol 15
buffer salt mixture derived from only potassium salts in the
reaction mixture, which resulted in better apparent buffering
capacity. Furthermore, we determined that addition of the
buffer salts to the reaction mixture containing the catalyst,
formalin, and toluene and stirring for 15 min prior to addition
of the aldehyde substrate resulted in the most reliable, efficient
formation of the intermediate lactol 15.
Na/K salt buffer. The addition of 10 mol % of AcOH led to a
significant amount of elimination product 26, while the
addition of NaOAc had no effect on the formation of the
intermediate lactol 15. The reaction was then conducted in the
presence of varying amounts of isovaleric acid, the organic acid
derived from the substrate isovaleraldehyde (14). We found
that the presence of as little as 0.1 mol % of acid present can
lead to formation of small amounts of elimination product 26.
Subsequently, we determined that use of the K salt buffer
suppresses elimination in the presence of 0.1 mol % of
isovaleric acid, but elimination product 26 is still observed
when 10 mol % of isovaleric acid is present. These results
indicate that use of purified aldehydes in the presence of the K
salt buffer provides optimal results in the α-hydroxymethylation
process (Table 3).
Optimization Studies: Structure and Reactivity of the
Intermediate Lactol 15. Considering our results in total so
far, we still found it puzzling that an apparently tiny amount of
acid impurity in the aldehyde substrate was sufficient to
promote observable amounts of elimination product 26,
particularly when the commercial pH 7 Na/K buffer was
utilized. Further consideration of the mechanistic pathway
leading to elimination led us to question what intermediate was
actually responsible for generation of the elimination product
26. Monitoring the reaction in toluene, we observed the
presence of equilibrium between the open and closed forms,
lactol 15 and hydroxy aldehyde 27, in favor of the closed form
15 (15:27 ratio 3:1) as shown in Scheme 6. While monitoring
the same reaction during the previously described solvent
screen, we observed that the equilibrium between 15 to 27
shifted toward the open form aldehyde 27 in solvents that had
not performed well in the α-hydroxymethylation reaction.
We also observed that the age of the formalin reagent was a
variable. On one occasion, when an old (unknown age,
potentially years old) sample of commercial formalin was
utilized, the yield suffered dramatically. We speculated that this
sample had largely been converted to paraformaldehyde and/or
formic acid. In light of this observation, we attempted use of
paraformaldehyde in aq buffered biphasic toluene/water
medium and observed no observable reaction. Apparently,
paraformaldehyde does not undergo appreciable dissociation to
monomeric formaldehyde (or its hydrate) under these reaction
conditions, and is thus not suitable for use in the catalytic α-
hydroxymethylation process (Scheme 5). Thus, use of fresh
samples of commercial formalin (stabilized by 10−15%
methanol) should be employed to obtain optimal results.
Optimization Studies: Purity of the Aldehyde Sub-
strates. During more detailed studies of the reaction variables,
another significant observation was made: no competing
elimination occurred when the substrate aldehydes were freshly
purified by distillation or chromatography. We began to
consider the possibility that the corresponding acid, presumably
resulting from air oxidation of the aldehyde was promoting
elimination. To test this hypothesis, reactions were run in
which substoichiometric quantities of acetic acid (AcOH) and
sodium acetate (NaOAc) were added to the reaction mixture
under the standard reaction conditions using the commercial
F
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 7. Detailed Catalytic Cycle for Enantioselective α-Hydroxymethylation of Aldehydes
Thus, we envisioned the following mechanistic scheme to
account for all of our observations. When significant amounts
of the open form hydroxy aldehyde are present, as is the case in
more polar solvents, the catalyst 13 may form the enamine 29
derived from 27 leading to partial or complete racemization to
a mixture enantiomeric hydroxy aldehydes 27 and 30 and/or
elimination to aldehyde 26 after hydrolysis (Scheme 6).
Another trend we observe is illustrated in Scheme 6. We also
observe formation of minor amounts of another aldehyde,
tentatively assigned as hydroxymethyl aldehyde 28 that we
attribute to the loss of one equivalent of formaldehyde from
aldehyde 27. Minimally, the presence of an additional
equivalent of formaldehyde completely suppresses the for-
mation of 28.
Formulation of the Complete Catalytic Cycle. Taking
into account all of the data obtained from the foregoing studies,
we are now able to formulate a more detailed description of the
catalytic cycle including the potential formation of elimination
products. As described in our original work, enamine catalysis
from starting aldehyde to α-hydroxymethylated aldehyde 4
follows the typical pathway.24 The formation of the observed
lactol 31 could occur via initial formation of aldehyde 4
followed by rapid consumption of an additional equivalent of
formaldehyde and cyclization to furnish 31 (Scheme 7, Pathway
A). Alternatively, reaction of intermediate 3 or the derived
aminol with a second equivalent of formaldehyde could also
occur leading to formation of 31. Another possibility could
arise from addition of a second equivalent of formaldehyde in
the form of its mono-, di-, or oligomeric oxonium ion to α-
hydroxymethylated iminium species 3. Subsequent ring closure
to aminal 32, departure of aminocatalyst 1 giving oxonium ion
33, and finally quenching with water generates lactol 31
(Scheme 7, Pathway B). On the basis of our experiments, we
hypothesize that further reactions of lactol 31 are under
thermodynamic control. Lactol 31 is in equilibrium with the
corresponding ring-opened aldehyde 34 (3:1 ratio in toluene).
Changes in this equilibrium ratio are primarily based on pH and
polarity of solvent. When the pH of the medium is acidic, this
encourages the equilibrium between 31 and 34 to shift toward
34, allowing for condensation of catalyst with aldehyde 34,
leading to either racemization (34 and its enantiomer 35) and/
or subsequent elimination to furnish unsaturated aldehyde 37.
Of course, additional possibilities also exist such as loss of a
proton from 3 (R2 = H to form the related enamine) followed
by nonstereoselective reprotonation to give 4 and 36 or by loss
of water to form a conjugated iminium ion and hydrolysis to
37.
When the pH is neutral and the medium is nonpolar, the
equilibrium favors lactol 31 which is sufficiently configuration-
ally stable to (1) be transformed into isolable synthons, and (2)
to prevent racemization by blocking further condensation of
catalyst, amine 1. In light of this analysis, one can conclude that
all the equilibrium constants must strongly favor lactol 31 when
excess formaldehyde is present. We also suggest that the
optimal conditions arise from the requirement for the presence
of at least one additional equivalent of formaldehyde, which
serves to (1) ensure that there is excess formaldehyde present
to drive the equilibrium to lactol 31 from 4, and (2) possibly
also to sequester the catalyst after the completion of the
reaction (Scheme 7).23
Once the optimal reaction conditions had been determined,
we began to further examine the scope of this process with
respect to the aldehyde. We sought to extend the scope even
more broadly than in our preliminary studies.24 We chose a
group of structurally diverse aldehydes as shown in Table 4. We
found that the reaction performed well across a variety of
branched aliphatic aldehydes such as 18, and 38−41 (Table 4,
Entries 1−5).
We generally found that as the β-branching of the aldehyde
increases, the reaction becomes increasingly sluggish, likely due
to steric hindrance by the β branches impeding either
formation of the required enamine, or perhaps more plausibly,
reaction with formaldehyde. This proposal was further
supported in the cases of substrates with α-branching, which
required refluxing conditions to promote α-hydroxymethyla-
tion. In our original study we had reported that α-
hydroxymethylation of 2-methylbutyraldehyde provided the
expected acid 40 (after oxidation) bearing a quaternary center
with very high enantioselectivity (Table 4, entry 4).24 Upon re-
examination of this reaction, it was determined that our
G
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 4. Scope of Aldehydes for Enantioselective α-Hydroxymethylation−Pinnick Oxidation
analytical method had failed to separate the two enantiomers
during the determination of the % ee. In fact, the reaction
proceeds with no observable enantioselectivity. However, as the
difference in the steric demand of the two groups at the
reacting center increases as is the case for 2,3-dimethylbutyr-
aldehyde, the resulting acid 41 (Table 4 Entry 5) exhibits a low
but not practically useful level of enantioselectivity (19% ee).
H
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 5. Substrate Scope for Tandem α-Hydroxymethylation−Wittig Olefination
The evaluation of functional groups including protecting
group compatibility with the conditions required for α-
hydroxymethylation were also studied. We were pleased to
find that a variety of functional groups including some common
protecting groups were tolerant of the reaction conditions. The
functional groups that tolerated the α-hydroxymethylation−
oxidation sequence and provided the desired chiral α-
hydroxymethyl acids 42−47 in generally very good to excellent
yields and % ee included a terminal alkyne, aryl group, methyl
ester, ketone, TBS protected alcohol, and Boc protected amine
(Table 4, entries 6−11). Particularly noteworthy is the
completely regioselective α-hydroxymethylation of an ω-keto-
aldehyde to afford keto acid 45 in 87% yield and 96% ee.
To extend the utility of the enantioselective α-hydroxyme-
thylation, we wanted to highlight the versatility of the
intermediate lactol. Attempts to selectively protect the open
form of the intermediate lactol 15 did not achieve optimal
results affording only moderate selectivity (3.5:1 16:17)
employing a variety of silylating agents and reaction conditions.
Therefore, we shifted our focus to employing the lactol 15
directly since we had demonstrated that the lactol 15 was in
equilibrium with the requisite hydroxy aldehyde intermediate.
Presumably by employing reactions that would selectively
transform the aldehyde isomer in situ, the resulting shift in the
equilibrium between lactol and aldehyde forms would allow
complete conversion of lactol 15 to the requisite product(s).
To this end, we discovered that treatment of the intermediate
lactols derived from five substrates, four of which had been
examined in Table 4, with ethyl 2-(triphenylphosphor-
anylidene)propanoate, afforded the corresponding δ-hydroxy-
α,β-unsaturated esters in high yield and stereoselectivity (Table
5). An example of this type of tandem process has also recently
been reported by Hayashi and co-workers, employing a
different stabilized ylide.31
While we were pleased with the substrate scope of the α-
hydroxymethylation, some limitations have been encountered
thus far. Substrates with the general structures 53−56 did not
undergo α-hydroxymethylation under the optimized reaction
conditions (Figure 2). For 53, we suspect that the low reactivity
Figure 2. Known limitations on the scope of enantioselective α-
hydroxymethylation.
may be a result of the formation of an unusually stable enamine
upon reaction with aminocatalyst 13. While this type of
aldehyde has successfully been employed in other organo-
catalytic reactions, they typically require higher reaction
temperatures and the presence of acid for efficient reaction to
occur.35 Under the optimized reaction conditions utilized above
at elevated temperatures, our expectation would be that
elimination and/or erosion of stereocontrol would occur. In
the case of aldehydes 54−56, the resulting enamines are
notoriously unstable and decompose under the reaction
conditions. In common with aldehydes of the structural
framework of 53, aldehydes 54, 55, and 56 have been
employed in other organocatalytic reactions using different
catalysts and usually under nonaqueous reaction conditions
(Figure 2).36
Studies of the Scalability of the Enantioselective α-
Hydroxymethylation Protocol. The scalability of the
I
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 8. Scaled Examples of Enantioselective Hydroxymethylation
solution (5 mL) and deionized H2O (300 mL) to give a purple
solution.
foregoing α-hydroxymethylation protocol was then examined.
Gratifyingly, we have shown that the reaction is quite readily
scalable affording the expected products in comparable yields
and enantioselectivity with minimal modifications to the
optimal experimental protocol. As shown in Scheme 8, the
reactions affording hydroxy acids 18 and 43, as well as
unsaturated ester 51 have been conducted on 33−440 mmol
scale in high yield and enantioselectivity. It should be noted
that all of these reactions have proven highly reproducible on a
number of different reaction scales.
In conclusion, we have developed a highly effective and
general direct organocatalytic enantioselective α-hydroxyme-
thylation of aldehydes. The procedure has been demonstrated
to be effective in obtaining β-hydroxycarboxylic acids and
-hydroxy-α,β-unsaturated esters in high yields and stereo-
selectivity. Future work will be directed to identifying additional
applications of this methodology and further exploration of the
chemistry of the intermediate lactol.
para-Anisaldehyde Stain. Prepared by the slow addition of
concentrated sulfuric acid (20.5 mL) to a mixture of anh ethanol (530
mL) and H2O (28 mL). The contents were then cooled to 0 °C and
treated with glacial acetic acid (6.2 mL) and p-anisaldehyde (15.0 mL).
The resulting solution should be a clear to pale yellow color.
Solvents and Reagents. All workup procedures involved reagent
grade solvents. Reaction solvents were obtained from a solvent
purification system except for the α-hydroxymethylation procedures
where reagent grade solvents were used without drying. Deionized
water was used anywhere that water is included in the procedure.
Commercially Available Reagents. Starting aldehydes were
either purchased and used without purification or synthesized through
the previous literature precedent indicated. All α,α-diarylprolinol
catalysts were synthesized through known procedures with the
indicated modifications discussed herein from commercially available
L-proline. Formaldehyde (37% solution) was purchased from J.T.
Baker Chemicals. The commercial pH 7 solid buffer consists of dibasic
sodium phosphate and monobasic potassium phosphate. Sodium
chlorite (NaClO2) and sodium phosphate, monobasic, monohydrate
(NaH2PO4·H2O) were purchased and used directly.
EXPERIMENTAL SECTION
■
Preparation of Phosphate Buffer. Prepared by mixing
potassium dibasic hydrogen phosphate (64.9 g, 0.477 mol, 1 equiv)
and potassium monobasic hydrogen phosphate (91.1 g, 0.523 mol,
1.10 equiv) in a grinder and processing until a free-flowing uniform
solid formed. (Note: This can also be performed using a mortar and
pestle.)
General Information. All nonaqueous reactions were carried out
using flame-dried glassware under an atmosphere of argon. All aqueous
reactions were carried out using glassware that was not flame-dried and
under an atmosphere of air capped with either a rubber septum or
plastic septum (specified in the procedures below). All reactions were
performed with magnetic stirring unless noted otherwise.
Physical Data. 1H NMR and 13C NMR spectra were obtained on a
500 MHz spectrometer. Chemical shifts are reported in ppm (δ)
downfield from tetramethylsilane and are internally referenced to the
Chromatography. Liquid chromatography was performed on
EMD silica gel 60 (230−400 mesh, particle size 0.040−0.063 mm)
using the specified solvent system as eluent. For analytical purposes,
thin-layer chromatography was performed using EMD silica 60 F254
precoated glass plates. Plates were analyzed by short wave UV
illumination and/or through employment of a specific stain.
Potassium Permanganate Stain (KMnO4). Prepared by
dissolving potassium permanganate (KMnO4) (3 g) and potassium
carbonate (K2CO3) (20 g) in a 5% sodium hydroxide (NaOH)
1
deuterated solvent. H NMR data are reported as follows: chemical
shift (multiplicity, coupling constant (Hz), number of hydrogens).
Multiplicities are denoted accordingly: s (singlet), b (broad signal), d
(doublet), dd (doublet of coublets), ddd (doublet of doublet of
doublets), dt (doublet of triplets), tt (triplet of triplets), dq (doublet of
quartets), t (triplet), q (quartet), p (pentet), m (multiplet). Infrared
spectra (IR) were acquired on an FT-IR (ATR) taken neat and are
J
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
reported in wavenumbers (cm−1). High resolution mass spectra were
obtained employing ionization techniques consisting of electron
impact (EI) or chemical ionization (CI) on a mass analyzer (MAT
95XL). Optical rotation values were measured on a polarimeter.
Samples were inserted into a cell with a path length of 1 dm. Melting
points were determined using a capillary melting point apparatus.
Elemental analyses were obtained using an autobalance and
determined by an analyzer. Enantiomeric excess was either determined
by chiral GC by analysis of the racemic material using a CP-Chirasil-
DEX CB column (25 m × 25 mm × 0.25 mm) utilizing a gas
chromatograph (GC) with an integrator or by supercritical fluid
chromatography (SFC) analysis with a SFC instrument equipped with
a high-pressure diode array UV−vis detector, a back-pressure
regulator, and a carbon dioxide pump, using a Diacel Chiralpak IA
column or a Diacel Chiralpak IC column for enantiomer separation.
(S)-Diphenyl(pyrrolidin-2-yl)methanol (12). A 1 L three-
necked round-bottomed flask equipped with an egg-shaped Teflon
coated magnetic stir bar, a Friedrich’s condenser capped with a rubber
septum in the middle neck, and 250 mL pressure equalizing addition
funnel capped with a rubber septum in the right neck was charged with
magnesium turnings (6.7 g, 275 mmol, 2.75 equiv) through the
remaining open neck. The flask was sealed by capping the open neck
with a rubber septum and stirring was initiated. The apparatus was
then put under an atmosphere of argon and flame-dried. After allowing
the apparatus to cool to ambient temperature, a crystal of iodine (50
mg, 0.2 mmol) was dissolved in 125 mL of anhydrous THF and added
via syringe resulting in a light brown colored transparent solution.
Bromobenzene (26.3 mL, 39.3 g, 250 mmol, 2.5 equiv) was then
added to the addition funnel, and one-quarter of the volume was
added to the flask. The flask was then heated by a 1 L electric mantle
until the iodine color dissipated. Heating was stopped, and 125 mL of
anhydrous THF was added to the addition funnel to dilute the
bromobenzene. The bromobenzene solution was then added dropwise
to the flask at a rate sufficient to maintain reflux in the flask (added
over 20 min). The clear solution became cloudy and brown during the
addition. When the addition was complete, heat was reapplied and the
flask was allowed to stir at reflux for 1 h. Heating was discontinued,
and the brown cloudy reaction mixture containing unused magnesium
was cooled to room temperature. N-Boc-L-proline methyl ester
(10)37,38 (22.9 g, 100 mmol, 1 equiv) was then dissolved in anhydrous
THF (200 mL), and added dropwise to the Grignard solution via the
same addition funnel over 45 min. The reaction was stirred at ambient
temperature for 1.5 h and subsequently cooled to 0 °C in an ice water
bath. The reaction was quenched by controlled addition of a saturated
aqueous solution of ammonium chloride (150 mL) via addition funnel
over 5 min. The biphasic mixture was transferred to a 2 L separatory
funnel and diluted with water (150 mL). The layers were separated,
and the aqueous layer was extracted with diethyl ether (3 × 150 mL).
The combined organic layers were dried over sodium sulfate and
gravity filtered through a large powder funnel equipped with a conical
medium porosity filter paper into a 1 L single-necked (24/40) round-
bottomed flask. The solvent was removed from the filtrate by rotary
evaporation (30 mmHg) at room temperature providing a clear
colorless residue. The flask containing the previously prepared residue
was equipped with an egg-shaped Teflon-coated magnetic stir bar (15
× 32 mm). Ethanol (500 mL) was subsequently added to the flask,
followed by sodium hydroxide (40 g, 1 mol, 10 equiv) and stirring was
initiated. The flask was then fitted with a Friedrich’s condenser open to
the atmosphere and heated to reflux using a 1 L electric mantle and
allowed to stir at reflux for 1 h. Heating was discontinued, the
Friedrich’s condenser removed, and the reaction was concentrated by
rotary evaporation using a 50 °C water bath (30 mmHg) providing a
light yellow amorphous solid. The residue was dissolved in water (200
mL) and diethyl ether (200 mL), and the resulting biphasic mixture
was transferred to a 1 L separatory funnel. The layers were separated
and the aqueous layer was extracted with diethyl ether (2 × 200 mL).
The combined organic layers were washed with water (200 mL) and
then a saturated aqueous solution of sodium chloride (200 mL). The
organic layers were then concentrated by rotary evaporation (50
mmHg) at room temperature to afford a light yellow solid. The solid
was dissolved in boiling hexanes (200 mL), decolorizing carbon (2.5
g) was added, and the solution was filtered hot through a large powder
funnel equipped with a conical medium porosity filter paper into a 500
mL Erlenmeyer flask containing refluxing hexanes (30 mL). Once the
filtration was complete, the flask was cooled to 0 °C and aged at 0 °C
for 1 h resulting in crystallization. The hexanes were then decanted off
and the white crystals were washed three times with 20 mL of cold (0
°C) hexanes washes (3 × 20 mL) and the washes successively
decanted. The resulting crystal solid was transferred to a 250 mL 24/
40 single-necked round-bottomed flask that was then fitted with a
vacuum adaptor and dried on a vacuum pump (0.15 mmHg) overnight
(14 h) to provide (S)-diphenyl(pyrrolidin-2-yl)methanol (12) (15.9 g,
63%) as white crystals having mp = 74−77 °C. A second crop of
crystals were obtained by concentrating the mother liquor and hexanes
washes by rotary evaporation (30 mmHg) to dryness, dissolving the
residue in boiling hexanes (50 mL), then cooling the solution to 0 °C
and aging the solution for 1 h at 0 °C. After isolation by decantation,
the resulting crystalline solids were washed three times with cold
hexanes (10 mL) and isolated by decantation and dried under vacuum
(0.15 mmHg) overnight (14 h) as above to give 3.0 g (12%) of white
1
crystals of comparable purity to the first crop: H NMR (500 MHz;
CDCl3) δ 7.57 (dd, J = 8.4, 1.1 Hz, 2H), 7.50 (dd, J = 8.4, 1.2 Hz, 2H),
7.31−7.26 (m, 5H), 7.18−7.14 (m, 2H), 4.59 (s, 1H), 4.26 (t, J = 7.7
Hz, 1H), 3.04 (ddd, J = 9.1, 6.6, 5.0 Hz, 1H), 2.97−2.93 (m, 1H),
1.78−1.54 (m, 6H); 13C NMR (126 MHz; CDCl3) δ 148.3, 145.5,
128.3, 128.1, 126.57, 126.46, 126.0, 125.7, 64.6, 46.9, 26.4, 25.6; IR
(ATR) (cm−1) 3352, 3059, 3024, 2966, 2947, 2870, 1597, 1493, 1447,
1369, 1173, 991, 748, 698, 660, 636; MS (APCI+) m/z (relative
intensity) 253.8 ([M + H+], 100%), 285.8 ([M + H+ + MeOH], 22%).
Anal. Calcd for C17H19NO: C, 80.60, H, 7.56, N, 5.53. Found: C,
80.64, H, 7.68, N, 5.49.
(S)-2-(Diphenyl((trimethylsilyl)oxy)methyl)pyrrolidine (13).
A 1 L 24/40 single-necked round-bottomed flask was equipped with
an egg-shaped Teflon-coated magnetic stir bar (15 × 32 mm) and a 60
mL pressure equalizing addition funnel fitted with a rubber septum.
The flask was placed under an atmosphere of argon and flame-dried.
After cooling to ambient temperature, a solution of (S)-diphenyl-
(pyrrolidin-2-yl)methanol (12) (17.7 g, 70 mmol, 1 equiv) in
anhydrous DCM (350 mL) was charged into the flask. The solution
was cooled to −78 °C in a dry ice and acetone bath, followed by
addition of triethylamine (12.7 mL, 9.2 g, 91 mmol, 1.3 equiv) in one
portion. Trimethylsilyl trifluoromethanesulfonate (16.5 mL, 20.2 g, 91
mmol, 1.3 equiv) was added dropwise via the addition funnel over 30
min. The reaction mixture was then stirred and allowed to warm to 0
°C over 2 h. The cooling bath was removed and the reaction mixture
was allowed to warm to ambient temperature over 1 h. The reaction
was quenched by addition of a solution of sat aq sodium bicarbonate
(100 mL) over 30 s. The mixture was diluted with water (100 mL) and
transferred to a 1 L separatory funnel. The phases were separated, and
the aqueous phase was extracted with DCM (3 × 100 mL). The
combined organic phases were dried over anhydrous sodium sulfate
and gravity filtered through a powder funnel equipped with medium
porosity conical filter paper. The filtrate was concentrated by rotary
evaporation (30 mmHg) to afford the crude product as an orange oil.
Purification by column chromatography with elution by 60% diethyl
ether/hexanes yielded (S)-2-(diphenyl((trimethylsilyl)oxy)methyl)-
1
pyrrolidine (13) (18.3 g, 80%) as a light yellow oil: H NMR (500
MHz; CDCl3) δ 7.46 (dd, J = 8.3, 1.1 Hz, 2H), 7.36−7.35 (m, 2H),
7.29−7.19 (m, 7H), 4.03 (t, J = 7.3 Hz, 1H), 2.88−2.77 (m, 2H),
1.64−1.52 (m, 5H), 1.40−1.35 (m, 1H), −0.09 (s, 9H); 13C NMR
(126 MHz; CDCl3) δ 128.5, 127.72, 127.68, 127.64, 127.01, 126.84,
83.3, 65.5, 47.3, 27.6, 25.2, 2.3; IR (ATR) (cm−1) 3060, 3024, 2955,
2897, 2874, 1493, 1447, 1404, 1246, 1068, 879, 833, 752, 698; MS
(APCI+) m/z (relative intensity) 325.9 ([M + H+], 100%), 357.9 ([M
+ H+ + MeOH], 4%). Anal. Calcd for C20H27NOSi: C, 73.79, H, 8.36,
N, 4.30. Found: C, 73.74, H, 8.41, N, 4.21.
Procedure A: General Procedure for the α-Hydroxymethy-
lation and Pinnick Oxidation of Aldehydes (Preparation of
Racemic Material). A 10 mL round-bottom flask equipped with
magnetic stir bar was charged with pyrrolidine (0.017 mL, 0.20 mmol,
K
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
0.10 equiv), pH 7 buffer (1.0 g), and toluene (4 mL) and stirring was
initiated. To the vigorously stirring solution was added aqueous
formaldehyde solution (37% aq, 0.5 mL, 6.0 mmol, 3.0 equiv). The
resulting biphasic mixture was allowed to stir for 15 min before freshly
distilled aldehyde (2.0 mmol, 1 equiv) was added and the vessel was
capped with a yellow hard plastic Caplug. The reaction was then
stirred for the indicated time period (Note: reaction monitored by 1H
NMR analysis of a direct aliquot of the toluene layer until absence of
starting material was observed). The toluene layer was then separated
and an additional extraction with toluene (1 mL) was performed. The
toluene extracts were then concentrated in vacuo (Note: the water
bath during concentration should remain at room temperature). The
resulting oil was then redissolved in t-butanol (10 mL) and 2-methyl-
2-butene (90%, 2.35 mL, 20 mmol, 10.0 equiv) and allowed to stir. To
the stirring solution was added a solution of NaClO2 (80%, 0.9 g, 8.0
mmol, 4.0 equiv) and NaH2PO4·H2O (1.1 g, 8.0 mmol, 4.0 equiv) in
H2O (5 mL) (Note: When performing this step on large scale, this
step should be performed by precooling the t-butanol/2-methyl-2-
butene/substrate mixture to 0 °C, as it produces a slight exotherm)
and the reaction was capped with a rubber septum with a needle as
outlet. The resulting green biphasic solution was stirred for 6 h (Note:
reaction progressively turns to a cloudy colorless solution) at which
point the solvent was removed in vacuo to afford a watery residue. The
residue was then diluted with EtOAc (10 mL), 10% HCl (2.5 mL),
and brine (2.5 mL). The aqueous layer was then extracted with EtOAc
(3 × 10 mL), the organic extracts were dried with Na2SO4, and the
solvent was removed in vacuo to afford a clear oil. The resulting oil
was then purified via flash chromatography to afford the target β-
hydroxycarboxylic acid. (Note: this reaction suffers from a significant
amount of α-methylenation, but generally provided enough material
for preparation of racemic samples. In the situations where this is not
the case, racemic 2-(diphenyl((trimethylsilyl)oxy)methyl)pyrrolidine
(rac-13) was employed as the organocatalyst).
Procedure B: General Procedure for the α-Hydroxymethy-
lation and Pinnick Oxidation of Aldehydes (Preparation of
Enantioenriched Material). A 10 mL round-bottom flask equipped
with magnetic stir bar was charged with (S)-2-(diphenyl-
((trimethylsilyl)oxy)methyl)pyrrolidine (13) (0.195 g, 0.6 mmol,
0.30 equiv), pH 7 buffer (0.5 g), and toluene (4 mL) and stirring was
initiated. To the vigorously stirring solution was added aqueous
formaldehyde solution (37% aq, 0.5 mL, 6.0 mmol, 3.0 equiv). This
was allowed to stir for 15 min before freshly distilled aldehyde (2.0
mmol, 1 equiv) was added and the vessel was capped with a yellow
hard plastic Caplug. The reaction was then stirred for the indicated
time period (Note: reaction monitored by 1H NMR analysis of a direct
aliquot of the toluene layer until absence of starting material was
observed). The toluene layer was then separated and an additional
extraction with toluene (1 mL) was performed. The toluene extracts
were concentrated in vacuo (Note: the water bath during
concentration should remain at room temperature). The residue was
then redissolved in t-butanol (10 mL) and 2-methyl-2-butene (90%,
2.35 mL, 20 mmol, 10.0 equiv) and allowed to stir. To the stirring
solution was added a solution of NaClO2 (80%, 0.9 g, 8.0 mmol, 4.0
equiv) and NaH2PO4·H2O (1.1 g, 8.0 mmol, 4.0 equiv) in H2O (5
mL) and the reaction was capped with a rubber septum with a needle
as outlet (Note: When performing this step on large scale, precooling
the t-butanol/2-methyl-2-butene/substrate mixture to 0 °C is required,
as it produces a slight exotherm). The resulting green biphasic solution
was stirred for 6 h (Note: reaction progressively turns to a cloudy
colorless solution) at which point the solvent was removed in vacuo to
afford a watery residue. The residue was then diluted with EtOAc (10
mL), 10% HCl (2.5 mL), and brine (2.5 mL). The aqueous layer was
then extracted with EtOAc (3 × 10 mL), the organic extracts were
dried with Na2SO4, and the solvent was removed in vacuo to afford a
clear oil. The resulting oil was purified via flash chromatography to
afford target β-hydroxycarboxylic acid (Note: Any procedures that
deviate from the stated procedure have been indicated below).
General Procedure for Methyl Esterification of β-Hydrox-
yacids. A 10 mL round-bottom flask equipped with magnetic stir bar
was charged with starting β-hydroxyacid (0.4 mmol) and N,N-
dimethylformamide (1 mL). The mixture was treated with potassium
carbonate (0.11 g, 0.8 mmol, 2 equiv) and iodomethane (0.027 mL,
0.44 mmol, 1.1 equiv) and the reaction was stirred for 15 h at room
temperature, progressively turning cloudy. The reaction mixture was
quenched with water (1 mL) and diluted with diethyl ether (2 mL)
and stirred until all suspended solids were dissolved. The organic layer
was removed and additional extractions were performed with Et2O (2
× 2 mL). The combined organic extracts were dried with anhydrous
sodium sulfate and concentrated in vacuo to afford the crude methyl
ester that was used directly for analysis of enantioselectivity.
General Procedure for Benzyl Esterification of β-Hydrox-
yacids. A 10 mL round-bottom flask equipped with magnetic stir bar
was charged with starting β-hydroxyacid (0.4 mmol) and N,N-
dimethylformamide (1 mL). The mixture was treated with potassium
carbonate (0.11 g, 0.8 mmol, 2 equiv) and benzyl bromide (0.052 mL,
0.44 mmol, 1.1 equiv) and the reaction was stirred for 15 h at room
temperature, progressively turning cloudy. The reaction mixture was
quenched with water (1 mL) and diluted with diethyl ether (2 mL)
and stirred until all suspended solids were dissolved. The combined
organic layer was removed and additional extractions were performed
with Et2O (2 × 2 mL). The organic extracts were dried with
anhydrous sodium sulfate and concentrated in vacuo to afford the
crude benzyl ester that was used directly for analysis of
enantioselectivity.
Procedure C: General Procedure for the α-Hydroxymethy-
lation and Wittig Olefination of Aldehydes (Preparation of
Racemic Material). A 10 mL round-bottom flask equipped with
magnetic stir bar was charged with pyrrolidine (0.017 mL, 0.2 mmol,
0.10 equiv), pH 7 buffer (1.0 g), and toluene (4 mL) and stirring was
initiated. To the vigorously stirring solution was added aqueous
formaldehyde solution (37% aq, 0.5 mL, 6.0 mmol, 3.0 equiv). This
was allowed to stir for 15 min before freshly distilled aldehyde (2.0
mmol, 1 equiv) was added and the vessel was capped with a yellow
hard plastic Caplug. The reaction was then stirred for the indicated
time period (Note: reaction monitored by 1H NMR analysis of a direct
aliquot of the toluene layer until absence of starting material was
observed). The toluene layer was then separated and an additional
extraction with toluene (1 mL) was performed. The toluene extracts
were then concentrated in vacuo (Note: the water bath during
concentration should remain at room temperature). The residue was
redissolved in DCM (2 mL) and subsequently added to a solution of
ethyl 2-(triphenylphosphoranylidene)propanoate (2.17 g, 6.0 mmol,
3.0 equiv) in dichloromethane (6 mL) dropwise via glass pipet at
room temperature and the vessel was capped with a rubber septum
(Note: On large scale this addition can be done via a pressure
equalizing addition funnel). The resulting green mixture was stirred for
24 h at room temperature. Upon completion by TLC analysis, the
reaction mixture was poured through a plug of Celite and concentrated
in vacuo to afford a green residue. The resulting residue was purified
via flash chromatography to afford the target esters (Note: this
reaction suffers from a significant amount of α-methylenation, but
generally provided enough material for preparation of a racemate
samples. In samples when this is not the case, racemic 2-
(diphenyl((trimethylsilyl)oxy)methyl)pyrrolidine (rac-13) was em-
ployed as the organocatalyst).
Procedure D: General Procedure for the α-Hydroxymethy-
lation and Wittig Olefination of Aldehydes (Preparation of
Enantioenriched Material). A 10 mL round-bottom flask equipped
with magnetic stir bar was charged with (S)-2-(diphenyl-
((trimethylsilyl)oxy)methyl)pyrrolidine (13) (0.195 g, 0.6 mmol,
0.30 equiv), pH 7 buffer (1.0 g), and toluene (4 mL) and stirring was
initiated. To the vigorously stirring solution was added aqueous
formaldehyde solution (37% aq, 0.5 mL, 6.0 mmol, 3.0 equiv). This
was allowed to stir for 15 min before freshly distilled aldehyde (2.0
mmol, 1.0 equiv) was added and the vessel was capped with a yellow
hard plastic Caplug. The reaction was then stirred for the indicated
time period (Note: reaction monitored by 1H NMR analysis of a direct
aliquot of the toluene layer until absence of starting material was
observed). The toluene layer was then separated and an additional
extraction with toluene (1 mL) was performed. The combined toluene
L
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
t
extracts were then concentrated in vacuo (Note: the water bath during
concentration should remain at room temperature). The residue was
then redissolved in DCM (2 mL) and subsequently added to a
solution of ethyl 2-(triphenylphosphoranylidene)propanoate (2.17 g,
6.0 mmol, 3.0 equiv) in DCM (6 mL) dropwise via glass pipet at room
temperature and the vessel was capped with a rubber septum (Note:
On large scale this addition can be done via a pressure equalizing
addition funnel). The resulting green mixture was stirred 24 h at room
temperature. Upon completion by TLC analysis, the reaction mixture
was poured through a plug of Celite and concentrated in vacuo to
afford a green residue. The resulting residue was purified via flash
chromatography to afford the target esters (Note: Any procedures that
deviate from the stated procedure have been indicated below).
5-Isopropyl-1,3-dioxan-4-ol (rac-15). A 10 mL round-bottom
flask equipped with magnetic stir bar was charged with racemic 2-
(diphenyl((trimethylsilyl)oxy)methyl)pyrrolidine (rac-13) (0.195 g,
0.6 mmol, 0.30 equiv), pH 7 buffer (0.5 g), and toluene (4 mL) and
stirring was initiated. To the vigorously stirring solution was added
aqueous formaldehyde solution (37% aq, 0.5 mL, 6.0 mmol, 3.0
equiv). This was allowed to stir for 15 min before freshly distilled
isovaleraldehyde (0.172 g, 0.21 mL, 2.0 mmol, 1.0 equiv) was added
and the vessel was capped with a yellow hard plastic caplug. The
reaction was then stirred for the 15 h at room temperature at which
833, 780 cm−1; HRMS Calcd for C9H19O3Si (M+ − Bu) 203.1098,
found 203.1094 (EI); [α]D20 −10.5° (c 2.0, DCM).
Note: Although the above mixture of compounds was purified via
flash chromatography, resolution of enantiomers of the mixture proved
difficult. As a result, we find that the crude reaction product above is
pure enough for further elaboration. The H NMR and 13C NMR
1
spectra for the crude material are provided in the Supporting
Information as well as spectra of the above mixture with arrows
indicating which shifts correspond to the respective products.
(R)-2-(Hydroxymethyl)-3-methylbutanoic acid (18). The
reaction was performed following the general procedure for α-
hydroxymethylation and Pinnick oxidation (Procedure B) starting
from isovaleraldehyde (α-hydroxymethylation reaction time was 15 h).
The resulting crude clear oil was then purified via flash
chromatography (gradient: DCM → 5% methanol/DCM) to afford
(R)-2-(hydroxymethyl)-3-methylbutanoic acid as a white solid (0.248
g, 94%, 94% ee) having mp 76−80 °C: Rf = 0.2 (5% methanol/DCM);
1H NMR (500 MHz; CDCl3) δ 6.91 (s, 2H), 3.89−3.85 (m, 1H),
3.81−3.78 (m, 1H), 2.44−2.41 (m, 1H), 2.06−2.01 (m, 1H), 0.99 (t, J
= 5.9 Hz, 6H); 13C NMR (126 MHz; CDCl3) δ 180.3, 61.5, 54.3, 27.8,
20.7, 20.2; IR (neat) 3263, 2962, 2877, 2631, 1704, 1196, 1010 cm−1;
HRMS Calcd for C6H13O3 (M+ + H) 133.0859, found 133.0862 (CI);
[α]D20 −6,2° (c 4.8, CHCl3). GC Analysis: Tinj = 250 °C, Tdet = 275
°C, flow = 2 mL/min, ti = 70 °C (1 min), tf = 160 °C, rate = 3 °C/min,
retention times of methyl ester: tmaj = 15.656, tmin = 15.397.
1
point H NMR analysis of a direct aliquot of the toluene layer of the
reaction mixture showed full consumption of starting material. The
toluene layer was then separated and an additional extraction with
toluene (1 mL) was performed. The combined toluene extracts were
then concentrated in vacuo (Note: the water bath during
concentration should remain at room temperature) to afford a clear
oil. The resulting oil was subsequently purified via flash chromatog-
raphy (30% diethyl ether/hexanes) to afford target lactol as a clear oil
(R)-2-(Hydroxymethyl)pentanoic acid (38). The reaction was
performed following the general procedure for α-hydroxymethylation
and Pinnick oxidation (Procedure B) starting from valeraldehyde (α-
hydroxymethylation reaction time was 15 h). The resulting oil was
then purified via flash chromatography (DCM → 5% methanol/
DCM) to afford (R)-2-(hydroxymethyl)pentanoic acid as a clear oil
1
1
(0.248 g, 94%, 93% ee): Rf = 0.2 (5% methanol/DCM); H NMR
(0.076 g, 26%): H NMR (500 MHz; CDCl3) δ 9.83 (t, J = 0.8 Hz,
(500 MHz; CDCl3) δ 7.47 (s, 2H), 3.76 (sextet, J = 6.1 Hz, 2H), 2.61
(quintet, J = 6.1 Hz, 1H), 1.63 (dq, J = 14.1, 7.3 Hz, 1H), 1.48 (dq, J =
14.3, 7.1 Hz, 1H), 1.38 (sextet, J = 7.5 Hz, 2H), 0.92 (t, J = 7.3 Hz,
3H); 13C NMR (500 MHz; CDCl3) δ 180.6, 63.1, 47.4, 30.5, 20.5,
14.1; IR (neat) 3367, 2959, 2935, 2874, 2661, 1709, 1466, 1196, 1053
cm−1; HRMS Calcd for C6H13O3 (M+ + H) 133.0859, found 133.0856
1H), 5.34 (d, J = 2.7 Hz, 2H), 5.21 (d, J = 5.9 Hz, 2H), 5.12 (d, J = 6.2
Hz, 2H), 4.94 (d, J = 6.2 Hz, 2H), 4.77 (s, 1H), 4.72−4.68 (m, 3H),
4.07 (dd, J = 11.5, 4.1 Hz, 1H), 3.96−3.93 (m, 3H), 3.76 (q, J = 10.5
Hz, 3H), 3.62 (dd, J = 11.6, 8.3 Hz, 1H), 3.40 (s), 2.78 (s, 1H), 2.48
(s, 2H), 2.40 (s, 1H), 2.30 (s), 2.20−2.12 (m, 2H), 2.00 (dt, J = 12.8,
6.6 Hz, 1H), 1.80−1.74 (m, 2H), 1.59 (t, J = 18.4 Hz, 3H), 1.48−1.43
(m, 2H), 1.27 (s, 1H), 1.06−0.86 (m, 31H); 13C NMR (126 MHz;
CDCl3) δ 96.1, 91.9, 89.8, 85.5, 66.0, 65.7, 46.9, 45.7, 31.7, 26.25,
26.09, 22.8, 20.9, 20.16, 19.99, 19.2, 14.3.
(CI); [α]D20 −4.2° (c 10, CHCl3). GC Analysis: Tinj = 250 °C, Tdet
=
275 °C, flow = 2 mL/min, ti = 70 °C (1 min), tf = 200 °C, rate = 1 °C/
min, retention times of methyl ester: tmaj = 29.905, tmin = 29.585.
(R)-2-(Hydroxymethyl)-3,3-dimethylbutanoic acid (39). The
reaction was performed following the general procedure for α-
hydroxymethylation and Pinnick oxidation (Procedure B) starting
from 3,3-dimethylbutyraldehyde (α-hydroxymethylation reaction time
was 120 h). The resulting oil was then purified via flash
chromatography (1% methanol/DCM→ 5% methanol/DCM) to
afford (R)-2-(hydroxymethyl)-3,3-dimethylbutanoic acid as a white
solid (0.202 g, 69%, 90% ee) having mp 129−133 °C: Rf = 0.2 (5%
Note 1: Decomposition of this compound was observed both neat
and in solution in a variety of solvents over the course of 5 h, not
allowing for full characterization.
Note 2: The 1H and 13C NMR contains a mixture of diastereomers
and oligomers and is incorporated into the Supporting Information
only for reference.
(R)-2-((((tert-Butyldimethylsilyl)oxy)methoxy)methyl)-3-
methylbutanal and tert-butyl(((5R)-5-Isopropyl-1,3-dioxan-4-
yl)oxy)dimethylsilane (16 and 17). To a stirred solution of (S)-2-
(diphenyl((trimethylsilyl)oxy)methyl)pyrrolidine (13) (0.39 g, 1.2
mmol, 0.30 equiv) in toluene (8.0 mL) was added solid pH 7 buffer
(0.20 g) followed by aqueous formaldehyde solution (37% aq, 1.0 mL,
12.0 mmol, 3.0 equiv) at rt. To the vigorously stirred suspension was
added aldehyde (0.43 mL, 0.34 g, 4.0 mmol) in one portion and the
resulting suspension was stirred for 12 h. The two layers were then
separated and the toluene concentrated in vacuo (Caution: keep the
water bath temperature <40 °C while evaporating). The residue was
then redissolved in DCM (5.0 mL) and added to a premixed solution
of imidazole (0.35 g, 5.2 mmol, 1.3 equiv) and TBSCl (0.60 g, 5.2
mmol, 1.3 equiv) stirred for 20 min at 0 °C under an atmosphere of
argon. The resulting mixture was stirred for 7 h at 0 °C before being
concentrated in vacuo. Purification of this material via column
chromatography on silica gel (with elution by 100% hexanes−2% ethyl
acetate/hexanes) provided an inseperable 3:1 mixture of aldehyde 16
and lactol 17 as a clear oil (0.55 g, 60%): 1H NMR (500 MHz, CDCl3)
δ 9.76 (d, J = 3 Hz, 1H), 4.83 (s, 2H), 3.95−3.75 (m, 2H), 2.40 (m,
1H), 2.15 (m, 1H), 1.04 (dd, J1 = 7 Hz, J2 = 5 Hz, 6H), 0.95 (s, 9H),
0.16 (s, 6H); 13C NMR (125 MHz, CDCl3) δ 204.2, 93.0, 64.8, 58.0,
26.0, 25.7, 20.4, 19.9, −5.0; IR (neat) 2959, 2858, 1724, 1253, 1041,
1
methanol/DCM); H NMR (500 MHz; CDCl3) δ 5.06 (s, 2H), 3.97
(t, J = 10.5 Hz, 1H), 3.83 (dd, J = 10.8, 3.9 Hz, 1H), 2.51 (dd, J = 10.2,
3.9 Hz, 1H), 1.03 (s, 9H); 13C NMR (126 MHz; CDCl3) δ 179.3,
61.3, 58.3, 32.2, 28.5; IR (neat) 3379, 3290, 2692, 2878, 1701, 1659,
1339, 1041 cm−1; HRMS Calcd for C7H15O3 (M+ + H) 147.1016,
found 147.1019 (CI); [α]D20 −13.3° (c 2.9, CHCl3). GC Analysis: Tinj
= 250 °C, Tdet = 275 °C, flow = 2 mL/min, ti = 40 °C (1 min), tf = 200
°C, rate = 4 °C/min, retention times of methyl ester: tmaj = 28.460, tmin
= 28.033.
2-(Hydroxymethyl)-2-methylbutanoic acid (40). A 10 mL
round-bottom flask equipped with magnetic stir bar and reflux
condenser was charged with pyrrolidine (0.05 mL, 0.60 mmol, 0.30
equiv), pH 7 buffer (0.5 g), and toluene (2 mL) and stirring was
initiated. To the vigorously stirring solution was added aqueous
formaldehyde solution (37% aq., 0.5 mL, 6.0 mmol, 3.0 equiv). This
was allowed to stir for 15 min before freshly distilled 2-
methylbutyraldehyde (0.172 g, 0.21 mL 2.0 mmol, 1 equiv) was
added and the vessel was heated to reflux with attached condenser.
The reaction was then stirred for 24 h at reflux. Heating was stopped,
the reaction was allowed to cool to ambient temperature, and the
toluene layer was then separated and an additional extraction with
M
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
toluene (1 mL) was performed. The toluene extracts were then
concentrated in vacuo (Note: the water bath during concentration
should remain at room temperature). The residue was then redissolved
in t-butanol (10 mL) and 2-methyl-2-butene (90%, 2.35 mL, 20 mmol,
10.0 equiv) and allowed to stir via magnetic stirring for 6 h. To the
stirring solution was added a solution of NaClO2 (80%, 0.9 g, 8.0
mmol, 4.0 equiv) and NaH2PO4·H2O (1.1 g, 8.0 mmol, 4.0 equiv) in
H2O (5 mL) and the reaction was capped with a rubber septum with a
needle as outlet. The resulting green biphasic solution was stirred for 6
h (Note: reaction progressively turns to a cloudy colorless solution) at
which point the solvent was removed in vacuo to afford a watery
residue. The residue was then diluted with EtOAc (10 mL), 10% HCl
(2.5 mL), and brine (2.5 mL). The aqueous layer was then extracted
with EtOAc (3 × 10 mL), the organics were dried with Na2SO4, and
the solvent was removed in vacuo to afford a clear oil. The resulting oil
was then purified via flash chromatography (DCM→ 5% methanol/
DCM) to afford 2-(hydroxymethyl)-2-methylbutanoic acid (0.206 g,
78%) as a white solid having mp =51−54 °C. (Note: This reaction was
attempted with (S)-2-(diphenyl((trimethylsilyl)oxy)methyl)-
pyrrolidine (13), but still furnished racemic product. SFC chromato-
grams are provided in the Supporting Information for both of these
runs: Rf = 0.2 (5% methanol/DCM); 1H NMR (500 MHz; CDCl3) δ
7.11 (s, 2H), 3.75 (d, J = 11.3 Hz, 1H), 3.53 (d, J = 11.3 Hz, 1H),
1.72−1.56 (m, 2H), 1.19 (s, 3H), 0.92 (t, J = 7.5 Hz, 3H); 13C NMR
(126 MHz; CDCl3) δ 183.0, 67.8, 48.1, 28.5, 18.9, 8.7; IR (neat) 3302,
2970, 2882, 1701, 1462, 1258, 1045 cm−1; HRMS Calcd for C6H13O3
(M+ + H) 133.0859, found 133.0855. SFC Analysis: Diacel Chiralpak
IA (0.46 cm ID × 25 cm L), 5% i-PrOH in scCO2, v = 4 mL/min, λ =
220 nm, 35 °C, 12 MPa; tR [min] of benzyl ester: 5.20, 5.49.
(R)-2-(Hydroxymethyl)hept-6-ynoic acid (42). The reaction
was performed following the general procedure for α-hydroxymethy-
lation and Pinnick oxidation (Procedure B) starting from hept-6-
ynal40 (α-hydroxymethylation reaction time was 12 h). The resulting
oil was then purified via flash chromatography (1% methanol/DCM→
5% methanol/DCM) to afford (R)-2-(hydroxymethyl)hept-6-ynoic
acid as a yellow oil (0.211 g, 68%, 91% ee): Rf = 0.2 (5% methanol/
DCM); 1H NMR (500 MHz; CDCl3) δ 3.82 (d, J = 5.7 Hz, 2H), 2.65
(quintet, J = 6.3 Hz, 1H), 2.24 (td, J = 6.6, 2.1 Hz, 2H), 1.97 (t, J = 2.2
Hz, 1H), 1.82 (dq, J = 14.4, 7.1 Hz, 1H), 1.75−1.61 (m, 3H); 13C
NMR (126 MHz; CDCl3) δ 180.2, 83.8, 69.1, 63.1, 47.2, 27.4, 26.1,
18.5; IR (neat) 3291, 3163, 2943, 2631, 1709, 1196, 1026 cm−1;
HRMS Calcd for C8H13O3 (M+ + H) 157.0859, found 157.0861 (CI);
[α]D20 −1.3° (c 2.9, CHCl3). SFC Analysis: Diacel Chiralpak IA (0.46
cm ID × 25 cm L), 5% i-PrOH in scCO2, v = 4 mL/min, λ = 220 nm,
35 °C, 12 MPa; tR [min] of benzyl ester: 8.08 (95.70%), 9.31 (4.31%).
(R)-2-Benzyl-3-hydroxypropanoic acid (43). The reaction was
performed following the general procedure for α-hydroxymethylation
and Pinnick oxidation (Procedure B) starting from hydrocinnamalde-
hyde (α-hydroxymethylation reaction time was 15 h). The resulting oil
was then purified via flash chromatography (DCM→ 5% methanol/
DCM) to afford (R)-2-benzyl-3-hydroxypropanoic acid as a white solid
(0.342 g, 95%, 92% ee) having mp = 60−64 °C: Rf = 0.3 (5%
methanol/DCM); 1H NMR (500 MHz; CDCl3) δ 7.35−7.25 (m,
5H), 6.65 (s, 1H), 3.83 (dd, J = 11.2, 2.8 Hz, 1H), 3.76 (dd, J = 11.2,
6.2 Hz, 1H), 3.11 (q, J = 9.2 Hz, 1H), 2.92−2.88 (m, 2H); 13C NMR
(126 MHz; CDCl3) δ 179.8, 138.3, 129.1, 128.8, 126.8, 62.0, 49.0,
34.2; IR (neat) 3283, 2947, 1705, 1196, 1030 cm−1; HRMS Calcd for
C10H12O3 (M+ + H) 180.0781, found 180.0788 (CI); [α]D20 +12.8° (c
2.0, CHCl3). SFC Analysis: Diacel Chiralpak IA (0.46 cm ID × 25 cm
L), 10% i-PrOH in scCO2, v = 4 mL/min, λ = 220 nm, 35 °C, 12 MPa;
tR [min] of methyl ester: 2.20 (4.22%), 2.37 (95.78%).
(R)-2-(Hydroxymethyl)-2,3-dimethylbutanoic acid (41). A 10
mL round-bottom flask equipped with magnetic stir bar and reflux
condenser was charged with (S)-2-(diphenyl((trimethylsilyl)oxy)-
methyl)pyrrolidine (13) (0.195 g, 0.6 mmol, 0.30 equiv), pH 7 buffer
(1.0 g), and toluene (2 mL) and stirring was initiated. To the
vigorously stirring solution was added aqueous formaldehyde solution
(37% aq, 0.5 mL, 6.0 mmol, 3.0 equiv). This was allowed to stir for 15
min before freshly distilled 2,3-dimethylbutyraldehyde39 (0.200 g, 2.0
mmol, 1 equiv) was added and the vessel was heated to reflux with
attached condenser. The reaction was stirred for 68 h at reflux.
Heating was stopped, the reaction was allowed to cool to ambient
temperature, and the toluene layer was then separated and an
additional extraction with toluene (1 mL) was performed. The toluene
extracts were then concentrated in vacuo (Note: the water bath during
concentration should remain at room temperature). The residue was
then redissolved in t-butanol (10 mL) and 2-methyl-2-butene (90%,
2.35 mL, 20 mmol, 10.0 equiv) and allowed to stir via magnetic stirring
for 6 h. To the stirring solution was added a solution of NaClO2 (80%,
0.9 g, 8.0 mmol, 4.0 equiv) and NaH2PO4·H2O (1.1 g, 8.0 mmol, 4.0
equiv) in H2O (5 mL) and the reaction was capped with a rubber
septum with a needle as outlet. The resulting green biphasic solution
was stirred for 6 h (Note: reaction progressively turns to a cloudy
colorless solution) at which point the solvent was removed in vacuo to
afford a watery residue. The residue was then diluted with EtOAc (10
mL), 10% HCl (2.5 mL), and brine (2.5 mL). The aqueous layer was
then extracted with EtOAc (3 × 10 mL), the organic extracts were
dried with Na2SO4, and the solvent was removed in vacuo to afford a
clear oil. The resulting oil was then purified via flash chromatography
(1% methanol/DCM→ 5% methanol/DCM) to afford (R)-2-
(hydroxymethyl)-2,3-dimethylbutanoic acid (0.105 g, 36%, 19% ee)
as a white solid having mp = 50−54 °C: Rf = 0.2 (5% methanol/
(R)-2-(Hydroxymethyl)-6-methoxy-6-oxohexanoic acid (44).
The reaction was performed following the general procedure for α-
hydroxymethylation and Pinnick oxidation (Procedure B) starting
from methyl-6-hexanoate41 (α-hydroxymethylation reaction time was
20 h). The resulting oil was then purified via flash chromatography
(DCM→ 5% methanol/DCM) to afford (R)-2-(hydroxymethyl)-6-
methoxy-6-oxohexanoic acid as a clear oil (0.346 g, 91%, 88% ee): Rf =
1
0.4 (5% methanol/DCM); H NMR (500 MHz, CDCl3) δ 6.10 (b,
2H), 3.68−3.64 (m, 2H), 3.62 (s, 3H), 2.50−2.47 (m, 1H), 2.31−2.29
(m, 2H), 1.63−1.56 (m, 3H), 1.45−1.42 (m, 1H); 13C NMR (125
MHz, CDCl3) δ 180.1, 174.3, 62.9, 51.6, 48.1, 33.7, 27.8, 22.5; IR
(neat) 3417, 2951, 2503, 1720, 1570, 1438, 1198, 1033, 910, 729
cm−1; HRMS Calcd for C8H15O5 (M+ + H) 191.0914, found 191.0922
(CI); [α]D20 +11.2° (c 4.0, methanol). SFC Analysis: Diacel Chiralpak
IC (0.46 cm ID × 25 cm L), 10% i-PrOH in scCO2, v = 4 mL/min, λ =
220 nm, 35 °C, 12 MPa; tR [min] of benzyl ester: 6.28 (94.24%), 6.71
(5.76%).
(R)-2-(Hydroxymethyl)-6-oxoheptanoic acid (45). The reac-
tion was performed following the general procedure for α-
hydroxymethylation and Pinnick oxidation (Procedure B) starting
from 6-oxoheptanal42 (α-hydroxymethylation reaction time was 15 h).
The resulting oil was then purified via flash chromatography (DCM →
5% methanol/DCM) to afford (R)-2-(hydroxymethyl)-6-oxoheptanoic
acid as a faint yellow oil (0.303 g, 87%, 96% ee): Rf = 0.4 (5%
1
methanol/DCM); H NMR (500 MHz; CDCl3) δ 6.66 (s, 2H), 3.79
(d, J = 5.8 Hz, 2H), 2.60 (quintet, J = 6.0 Hz, 1H), 2.48 (t, J = 6.7 Hz,
2H), 2.14 (s, 3H), 1.68−1.51 (m, 4H); 13C NMR (126 MHz; CDCl3)
δ 209.3, 179.5, 62.9, 47.3, 43.4, 30.1, 27.6, 21.3; IR (neat) 3287, 2947,
1701, 1362, 1180, 1045 cm−1; HRMS Calcd for C8H15O4 (M+ + H)
175.0965, found 175.0965 (CI); [α]D20 +12.8° (c 2.0, CHCl3). SFC
Analysis: Diacel Chiralpak IC (0.46 cm ID × 25 cm L), 10% i-PrOH in
scCO2, v = 4 mL/min, λ = 220 nm, 35 °C, 12 MPa; tR [min] of benzyl
ester: 5.96 (97.93%), 10.44 (2.07%).
(R)-5-((tert-Butyldimethylsilyl)oxy)-2-(hydroxymethyl)-
pentanoic acid (46). The reaction was performed following the
general procedure for α-hydroxymethylation and Pinnick oxidation
(Procedure B) starting from 5-((tert-butyldimethylsilyl)oxy)-
pentanal43 (α-hydroxymethylation reaction time was 15 h). The
1
DCM); H NMR (500 MHz; CDCl3) δ 3.81 (d, J = 11.2 Hz, 1H),
3.55 (d, J = 11.3 Hz, 1H), 2.13 (dt, J = 13.7, 6.9 Hz, 1H), 1.13 (s, 3H),
0.94 (dd, J = 6.9, 2.6 Hz, 6H); 13C NMR (126 MHz; CDCl3) δ
182.98, 66.9, 51.2, 31.8, 17.9, 17.4, 15.0; IR (neat) 3283, 2963, 2877,
1701, 1555, 1466, 1254, 1030 cm−1. Anal. calcd for C7H14O3: C, 57.51,
H, 9.65. Found: C, 57.60, H, 9.58; [α]D20 −7.2° (c 1.0, CHCl3). SFC
Analysis: Diacel Chiralpak IC (0.46 cm ID × 25 cm L), 5% i-PrOH in
scCO2, v = 4 mL/min, λ = 220 nm, 35 °C, 12 MPa; tR [min] of benzyl
ester: 5.49 (40.67%), 5.73 (59.33%).
N
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
resulting oil was then purified via flash chromatography (DCM→ 5%
methanol/DCM) to afford (R)-5-((tert-butyldimethylsilyl)oxy)-2-
(hydroxymethyl)pentanoic acid as a clear oil (0.493 g, 94%, 93%
scCO2, v = 4 mL/min, λ = 220 nm, 35 °C, 12 MPa; tR [min]: 4.89
(95.10%), 6.59 (4.9%).
Ethyl (S,E)-4-benzyl-5-hydroxy-2-methylpent-2-enoate (50).
The reaction was performed following the general procedure for α-
hydroxymethylation and Wittig olefination (Procedure D) starting
from hydrocinnamaldehyde (α-hydroxymethylation reaction time was
15 h). The resulting oil was then purified via flash chromatography
(50% hexanes/diethyl ether) to afford ethyl (S,E)-4-benzyl-5-hydroxy-
2-methylpent-2-enoate as a clear oil (0.442 g, 89%, 84% ee): Rf = 0.5
1
ee): Rf = 0.4 (5% methanol/DCM); H NMR (500 MHz; CDCl3) δ
6.59 (s, 2H), 3.80−3.78 (m, 2H), 3.65 (dd, J = 5.6, 3.1 Hz, 2H), 2.64
(quintet, J = 5.5 Hz, 1H), 1.76−1.61 (m, 4H), 0.89 (s, 9H), 0.05 (s,
6H); 13C NMR (126 MHz; CDCl3) δ 180.0, 63.10, 63.03, 47.2, 30.1,
26.1, 24.9, 18.5, −5.2; IR (neat) 3279, 2951, 2928, 2885, 1709, 1388,
1254, 1096, 833, 775 cm−1; HRMS Calcd for C12H27O4Si (M+ + H)
263.1673, found 263.1676 (CI); [α]D20 −6.8° (c 1.2, CHCl3). SFC
Analysis: Diacel Chiralpak IC (0.46 cm ID × 25 cm L), 10% i-PrOH in
scCO2, v = 4 mL/min, λ = 220 nm, 35 °C, 12 MPa; tR [min] of benzyl
ester and TBS removed diol: 13.77 (96.51%), 22.59 (3.49%).
(Note: The TBS moiety is removed by dissolving the crude reaction
mixture from benzyl protection in DCM (1 mL) and treating with
10% HCl (0.5 mL), stirring for 20 min, removing the DCM layer and
using it directly for SFC analysis.)
1
(50% hexanes/diethyl ether); H NMR (500 MHz; CDCl3) δ 7.36−
7.34 (m, 3H), 7.28 (d, J = 7.5 Hz, 1H), 7.24 (d, J = 7.4 Hz, 2H), 6.72
(dd, J = 10.1, 1.3 Hz, 1H), 4.30−4.24 (m, 2H), 3.76 (dt, J = 9.8, 4.6
Hz, 1H), 3.67 (t, J = 8.5 Hz, 1H), 3.05−2.97 (m, 1H), 2.92 (dd, J =
13.5, 6.3 Hz, 1H), 2.70 (dd, J = 13.5, 7.9 Hz, 1H), 1.76 (s, 3H), 1.53
(s, 1H), 1.38 (t, J = 7.1 Hz, 3H); 13C NMR (126 MHz; CDCl3) δ
168.0, 141.9, 139.3, 130.5, 129.2, 128.5, 126.4, 65.4, 60.8, 43.9, 37.5,
14.4, 12.9; IR (neat) 3437, 2932, 1705, 1451, 1273, 1219, 1103, 1030,
745, 698 cm−1. Anal. calcd for C15H20O3: C, 72.55, H, 8.12. Found: C,
72.478, H, 8.120. [α]D20 +23.2° (c 2.0, CHCl3). SFC Analysis: Diacel
Chiralpak IC (0.46 cm ID × 25 cm L), 10% i-PrOH in scCO2, v = 4
mL/min, λ = 220 nm, 35 °C, 12 MPa; tR [min]: 3.89 (92.08%), 5.35
(7.92%).
(R)-3-((tert-Butoxycarbonyl)amino)-2-(hydroxymethyl)-
propanoic acid (47). The reaction was performed following the
general procedure for α-hydroxymethylation and Pinnick oxidation
(Procedure B) starting from tert-butyl(3-oxopropyl)carbamate44 (α-
hydroxymethylation reaction time was 30 h). The resulting oil was
then purified via flash chromatography (DCM→ 10% methanol/
DCM) to afford (R)-3-((tert-butoxycarbonyl)amino)-2-
(hydroxymethyl)propanoic acid as a yellow oil (0.285 g, 65%, 91%
Ethyl (S,E)-7-((tert-butyldimethylsilyl)oxy)-4-(hydroxymeth-
yl)-2-methylhept-2-enoate (51). The reaction was performed
following the general procedure for α-hydroxymethylation and Wittig
olefination (Procedure D) starting from 5-((tert-butyldimethylsilyl)-
oxy)pentanal43 (α-hydroxymethylation reaction time was 15 h). The
resulting oil was then purified via flash chromatography (60% hexanes/
diethyl ether) to afford ethyl (S,E)-4-benzyl-5-hydroxy-2-methylpent-
2-enoate as a clear oil (0.621 g, 94%, 86% ee): Rf = 0.6 (50% hexanes/
diethyl ether); 1H NMR (500 MHz; CDCl3) δ 6.53 (dd, J = 10.3, 1.3
Hz, 1H), 4.19 (q, J = 7.1 Hz, 2H), 3.63 (dd, J = 10.7, 5.5 Hz, 1H), 3.58
(t, J = 6.2 Hz, 2H), 3.53 (t, J = 9.1 Hz, 1H), 2.70−2.63 (m, 1H), 1.88
(d, J = 1.4 Hz, 3H), 1.62−1.45 (m, 4H), 1.29 (t, J = 7.1 Hz, 4H), 0.88
(s, 9H), 0.03 (s, 6H); 13C NMR (126 MHz; CDCl3) δ 168.1, 143.1,
130.37, 130.36, 77.4, 66.0, 63.1, 60.8, 41.8, 30.4, 27.5, 26.1, 18.5, 14.4,
13.2, −5.2; IR (neat) 3453, 2932, 2859, 1709, 1254, 1211, 1096, 1049,
833, 775, 748 cm−1. Anal. calcd for C17H34O4Si: C, 61.77, H, 10.37.
Found: C, 61.58, H, 10.46. [α]D20 +10.8° (c 2.0, CHCl3). SFC
Analysis: Diacel Chiralpak IC (0.46 cm ID × 25 cm L), 10% i-PrOH in
scCO2, v = 4 mL/min, λ = 220 nm, 35 °C, 12 MPa; tR [min]: 2.13
(92.95%), 2.71 (7.05%).
1
ee): Rf = 0.1 (5% methanol/DCM); H NMR (500 MHz; CDCl3) δ
7.47 (s, 2H), 5.23 (s, 1H), 3.84 (s, 2H), 3.55−3.43 (m, 2H), 2.74 (s,
1H), 1.44 (s, 9H); 13C NMR (126 MHz; CDCl3) δ 177.1, 157.6, 80.5,
59.8, 47.8, 37.9, 28.4; IR (neat) 3340, 2978, 1686, 1520, 1366, 1250,
1165, 1076, 1034 cm−1; HRMS Calcd for C9H17NO5Na (M+ + Na)
242.0999, found 242.0996 (CI); [α]D20 −7.5° (c 2.4, CHCl3). SFC
Analysis: Diacel Chiralpak IC (0.46 cm ID × 25 cm L), 10% i-PrOH in
scCO2, v = 4 mL/min, λ = 220 nm, 35 °C, 12 MPa; tR [min] of benzyl
ester: 2.33 (95.42%), 2.96 (4.58%).
Ethyl (S,E)-4-(hydroxymethyl)-2-methylhept-2-enoate (48).
The reaction was performed following the general procedure for α-
hydroxymethylation and Wittig olefination (Procedure D) starting
from valeraldehyde (α-hydroxymethylation reaction time was 15 h).
The resulting oil was then purified via flash chromatography (50%
hexanes/diethyl ether) to afford ethyl (S,E)-4-(hydroxymethyl)-2-
methylhept-2-enoate as a clear oil (0.352 g, 88%, 90% ee): Rf = 0.5
1
(50% hexanes/diethyl ether); H NMR (500 MHz; CDCl3) δ 6.52
Ethyl (S,E)-5-hydroxy-2-methyl-4-(pyridin-3-ylmethyl)pent-
2-enoate (52). A 10 mL round-bottom flask equipped with magnetic
stir bar was charged with (S)-2-(diphenyl((trimethylsilyl)oxy)methyl)-
pyrrolidine (13) (0.195 g, 0.6 mmol, 0.30 equiv), pH 7 buffer (0.50 g),
and toluene (4 mL) and stirring was initiated. To the stirring solution
was added aqueous formaldehyde solution (37% aq., 0.5 mL, 6.0
mmol, 3.0 equiv) followed by freshly purified 3-(pyridin-3-yl)-
propanal45 (0.270 g, 2.0 mmol, 1 equiv) and capped with plastic
septum and the resulting solution was stirred for 48 h, at which point
1H NMR showed full consumption of starting material. The toluene
(dd, J = 10.4, 1.3 Hz, 1H), 4.19 (q, J = 7.1 Hz, 2H), 3.62 (dd, J = 10.6,
5.5 Hz, 1H), 3.51 (dd, J = 10.6, 7.7 Hz, 1H), 2.69−2.66 (m, 1H), 1.89
(d, J = 1.4 Hz, 3H), 1.51−1.47 (m, 2H), 1.33−1.25 (m, 6H), 0.89 (t, J
= 7.0 Hz, 3H); 13C NMR (126 MHz; CDCl3) δ 168.2, 143.4, 130.3,
66.1, 60.8, 41.9, 33.4, 20.5, 14.42, 14.30, 13.2; IR (neat) 3449, 2955,
2932, 2870, 1708, 1269, 1219, 1126, 1096, 1038, 752 cm−1. Anal. calcd
for C11H20O3: C, 65.97, H, 10.07. Found: C, 65.727, H, 10.128. [α]D20
+8.3° (c 2.0, CHCl3). SFC Analysis: Diacel Chiralpak IC (0.46 cm ID
× 25 cm L), 5% i-PrOH in scCO2, v = 4 mL/min, λ = 220 nm, 35 °C,
12 MPa; tR [min]: 5.59 (94.90%), 7.68 (5.10%).
layer was then separated and an additional extraction with toluene (1
mL) was performed. The combined toluene extracts were then
concentrated in vacuo (Note: the water bath during concentration
should remain at room temperature). The residue was then redissolved
in DCM (2 mL) and added to a solution of ethyl 2-
(triphenylphosphoranylidene)propanoate (2.17 g, 6.0 mmol, 3.0
equiv) in DCM (6 mL) dropwise via glass pipet at room temperature
and the vessel was capped with a rubber septum. The resulting green
mixture was stirred for 24 h at room temperature. Imidazole (0.177 g,
2.6 mmol, 1.3 equiv) and TBSCl (0.392 g, 2.6 mmol, 1.3 equiv) were
added sequentially and the reaction mixture was let stir 30 min. The
reaction mixture was poured through a plug of Celite and concentrated
in vacuo to afford a green residue. The resulting residue was purified
via flash chromatography (hexanes to 50% hexanes/Et2O) to afford
target compound as a clear oil (0.190 g, 52%, 83% ee): Rf = 0.5
Ethyl (S,E)-4-(hydroxymethyl)-2,5-dimethylhex-2-enoate
(49). The reaction was performed following the general procedure
for α-hydroxymethylation and Wittig olefination (Procedure D)
starting from isovaleraldehyde (α-hydroxymethylation reaction time
was 15 h). The resulting oil was then purified via flash chromatography
(50% hexanes/diethyl ether) to afford ethyl (S,E)-4-(hydroxymethyl)-
2,5-dimethylhex-2-enoate as a clear oil (0.348 g, 87%, 90% ee): Rf =
0.5 (50% hexanes/diethyl ether); 1H NMR (500 MHz; CDCl3) δ 6.62
(dd, J = 10.8, 1.3 Hz, 1H), 4.20 (q, J = 7.1 Hz, 2H), 3.75 (dd, J = 10.7,
5.0 Hz, 1H), 3.57 (dd, J = 10.7, 8.1 Hz, 1H), 2.47 (dtd, J = 11.2, 7.3,
4.3 Hz, 1H), 1.89 (d, J = 1.4 Hz, 3H), 1.79 (dq, J = 13.6, 6.8 Hz, 1H),
1.39 (s, 1H), 1.30 (t, J = 7.1 Hz, 3H), 0.94 (d, J = 6.8 Hz, 3H), 0.88 (d,
J = 6.8 Hz, 3H); 13C NMR (126 MHz; CDCl3) δ 168.1, 141.9, 131.0,
77.4, 77.2, 76.9, 64.4, 60.8, 48.2, 29.4, 20.9, 19.7, 14.4, 13.3; IR (neat)
3449, 2959, 2874, 1709, 1647, 1466, 1366, 1277, 1242, 1134, 1096,
1042, 748 cm−1. Anal. calcd for C11H20O3: C, 65.97, H, 10.07, O,
23.97. Found: C, 65.61, H, 10.16. [α]D20 +9.5° (c 2.0, CHCl3). SFC
Analysis: Diacel Chiralpak IC (0.46 cm ID × 25 cm L), 5% i-PrOH in
1
(diethyl ether) H NMR (500 MHz; CDCl3) δ 8.45−8.42 (m, 2H),
7.45 (d, J = 8.0 Hz, 1H), 7.18 (dd, J = 7.7, 4.9 Hz, 1H), 6.61 (dd, J =
10.2, 1.3 Hz, 1H), 4.19−4.14 (m, 2H), 3.57 (dd, J = 5.5, 4.0 Hz, 2H),
O
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
2.98 (dd, J = 13.6, 5.6 Hz, 1H), 2.83−2.78 (m, 1H), 2.55 (dd, J = 13.6,
8.4 Hz, 1H), 1.62 (d, J = 1.3 Hz, 3H), 1.28 (t, J = 7.1 Hz, 3H), 0.91 (s,
9H), 0.05 (d, J = 2.6 Hz, 6H); 13C NMR (126 MHz; CDCl3) δ 167.8,
150.6, 147.7, 141.3, 136.6, 135.1, 129.6, 123.2, 64.7, 60.6, 43.6, 34.4,
25.9, 18.4, 14.3, 12.7, −5.28, −5.39; IR (neat) 2951, 2928, 2855, 1709,
1470, 1254, 1231, 1099, 833, 775 cm−1. Anal. calcd for C20H33NO3Si:
(23) Erkkila, A.; Pihko, P. M. J. Org. Chem. 2006, 71, 2538−2541.
̈
(24) Boeckman, R. K., Jr.; Miller, J. R. Org. Lett. 2009, 11, 4544−
4547.
(25) Marigo, M.; Wabnitz, T. C.; Fielenbach, D.; Jørgensen, K. A.
Angew. Chem., Int. Ed. 2005, 44, 794−797.
(26) Hayashi, Y.; Gotoh, H.; Hayashi, T.; Shoji, M. Angew. Chem., Int.
Ed. 2005, 44, 4212−4215.
(27) Palomo, C.; Mielgo, A. Angew. Chem., Int. Ed. 2006, 45, 7876−
7880.
20
C, 66.07, H, 9.15, N, 3.85. Found: C, 65.99, H, 9.22, N, 4.074. [α]D
+30.1° (c 2.2, CHCl3). SFC Analysis: Diacel Chiralpak IC (0.46 cm ID
× 25 cm L), 3% i-PrOH in scCO2, v = 4 mL/min, λ = 220 nm, 35 °C,
12 MPa; tR [min]: 29.28 (8.52%), 32.59 (91.48%).
(28) Guillena, G.; Ramon
1465−1492.
(29) Marigo, M.; Jørgensen, K. A. Chem. Commun. 2006, 2001−
2011.
́
, D. J. Tetrahedron: Asymmetry 2006, 17,
ASSOCIATED CONTENT
■
S
* Supporting Information
(30) Jensen, K. L.; Dickmeiss, G.; Jiang, H.; Albrecht, Ł.; Jørgensen,
K. A. Acc. Chem. Res. 2012, 45, 248−264.
(31) Yasui, Y.; Benohoud, M.; Sato, I.; Hayashi, Y. Chem. Lett. 2014,
43, 556−558.
(32) (a) Igarashi, M. T.; Masaru. J. Heterocycl. Chem. 1995, 32, 807−
810. (b) Kimpe, N. D. V.; Roland, B. L. D.; Schamp, N. Chem. Ber.
1983, 116, 3846−3857.
(33) Mori, K.; Ohki, M.; Matsui, M. Tetrahedron 1970, 26, 2821−
2824.
1H NMR, 13C NMR spectra and GC, and SFC traces for all
products. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: rkb@rkbmac.chem.rochester.edu.
■
Notes
(34) Joshi, L. D.; Srivastava, A. C.; Dutta, B. K. Fert. Technol. 1976,
13, 128−130.
The authors declare no competing financial interest.
́
(35) Sulzer-Mossee, S.; Laars, M.; Kriis, K.; Kanger, T.; Alexakis, A.
ACKNOWLEDGMENTS
Synthesis 2007, 1729−1732.
■
(36) Ian Storer, R.; MacMillan, D. W. C. Tetrahedron 2004, 60,
7705−7714.
We are grateful for partial support of these studies by Novartis
Pharma AG and Novartis Pharmaceuticals Corp and the
University of Rochester.
(37) Huy, P.; Neudorfl, J.-M.; Schmalz, H. G. Org. Lett. 2011, 13,
̈
216−219.
(38) Arisawa, M.; Takahashi, M.; Takezawa, E.; Yamaguchi, T.;
Torisawa, Y.; Nishida, A.; Nakagawa, M. Chem. Pharm. Bull. 2000, 48,
1593−1596.
REFERENCES
■
(1) Evans, D. A.; Dow, R. L.; Shih, T. L.; Takacs, J. M.; Zahler, R. J.
Am. Chem. Soc. 1990, 112, 5290−5313.
(39) Meyers, A. I.; Walkup, R. D. Tetrahedron 1985, 41, 5089−5106.
(40) Sui, B.; Yeh, E.A.-H.; Curran, D. P. J. Org. Chem. 2010, 75,
2942−2954.
(2) Crimmins, M. T.; Carroll, C. A.; King, B. W. Org. Lett. 2000, 2,
597−599.
(3) Evans, D. A.; Connell, B. T. J. Am. Chem. Soc. 2003, 125, 10899−
(41) Claus, R. E.; Schreiber, S. L. Org. Synth. 1986, 64, 150.
(42) Hong, B.-C.; Chen, F.-L.; Chen, S.-H.; Liao, J.-H.; Lee, G.-H.
Org. Lett. 2005, 7, 557−560.
10905.
(4) Lister, T.; Perkins, M. V. Angew. Chem., Int. Ed. 2006, 45, 2560−
2564.
(5) Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem. Soc. 1982,
́
(43) Frankowski, K. J.; Golden, J. E.; Zeng, Y.; Lei, Y.; Aube, J. J. Am.
Chem. Soc. 2008, 130, 6018−6024.
104, 1737−1739.
(44) Yokoyama, H.; Ejiri, H.; Miyazawa, M.; Yamaguchi, S.; Hirai, Y.
Tetrahedron: Asymmetry 2007, 18, 852−856.
(6) Evans, D. A.; Urpi, F.; Somers, T. C.; Clark, J. S.; Bilodeau, M. T.
J. Am. Chem. Soc. 1990, 112, 8215−8216.
(45) Baldwin, J. E.; Claridge, T. D. W.; Culshaw, A. J.; Heupel, F. A.;
(7) List, B. Tetrahedron 2002, 58, 5573−5590.
(8) Berkessel, A.; Groeger, H. Asymmetric Organocatalysis; Wiley-
VCH: Weinheim, 2005.
Smrckova,
6922.
́
S.; Whitehead, R. G. Tetrahedron Lett. 1996, 37, 6919−
(9) List, B. Chem. Rev. 2007, 107, 5413−5883.
(10) Dalko, P. I.; Moisan, L. Angew. Chem., Int. Ed. 2004, 43, 5138−
5175.
(11) List, B. Angew. Chem., Int. Ed. 2010, 49, 1730−1734.
(12) List, B.; Lerner, R. A.; Barbas, C. F. J. Am. Chem. Soc. 2000, 122,
2395−2396.
(13) Hajos, Z. G.; Parrish, D. R. German Patent DE 2102623, 1971.
(14) Hajos, Z. G.; Parrish, D. R. J. Org. Chem. 1974, 39, 1615−1621.
(15) Eder, U.; Sauer, G.; Wiechert, R. Angew. Chem., Int. Ed. 1971,
10, 496−497.
(16) Kuwano, R.; Miyazaki, H.; Ito, Y. Chem. Commun. 1998, 71−72.
(17) Kuwano, R.; Miyazaki, H.; Ito, Y. J. Organomet. Chem. 2000,
603, 18−29.
(18) Fukuchi, I.; Hamashima, Y.; Sodeoka, M. Adv. Synth. Catal.
2007, 349, 509−512.
(19) Mouri, S.; Chen, Z.; Matsunaga, S.; Shibasaki, M. Chem.
Commun. 2009, 5138−5140.
(20) Ozasa, N.; Wadamoto, M.; Ishihara, K.; Yamamoto, H. Synlett
2003, 2219−2221.
(21) Ishikawa, S.; Hamada, T.; Manabe, K.; Kobayashi, S. J. Am.
Chem. Soc. 2004, 126, 12236−12237.
(22) Casas, J.; Sunden
6117−6119.
́ ́
, H.; Cordova, A. Tetrahedron Lett. 2004, 45,
P
DOI: 10.1021/acs.joc.5b00380
J. Org. Chem. XXXX, XXX, XXX−XXX