Organocatalytic asymmetric α-benzoyloxylation of α-branched aldehydes and enals: A useful approach to oxygenated quaternary stereocenters

Article abstract of DOI:10.1016/j.tet.2012.06.043

Direct asymmetric α-benzoyloxylation of α-branched aldehydes and α-branched enals via enamine and dienamine catalysis was used to construct quaternary oxygenated stereocenters with good yields and moderate to good enantioselectivity. This method uses an inexpensive and readily available cinchona alkaloid-derived primary amine as the catalyst, benzoyl peroxide as the oxygen source, and stoichiometric amounts of the aldehyde substrates, providing simple metal-free access to valuable protected 2-hydroxyaldehyde derivatives.

Full text of DOI:10.1016/j.tet.2012.06.043

Tetrahedron 68 (2012) 7568e7574
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Organocatalytic asymmetric a-benzoyloxylation of a-branched aldehydes
and enals: a useful approach to oxygenated quaternary stereocenters
*
Nicolas Demoulin, Olga Lifchits, Benjamin List
€
€
Max-Planck-Institut fur Kohlenforschung, Kaiser-Wilhelm-Platz 1, 45470 Mulheim an der Ruhr, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 5 April 2012
Received in revised form 19 May 2012
Accepted 8 June 2012
Available online 21 June 2012
Direct asymmetric a-benzoyloxylation of a-branched aldehydes and a-branched enals via enamine and
dienamine catalysis was used to construct quaternary oxygenated stereocenters with good yields and
moderate to good enantioselectivity. This method uses an inexpensive and readily available cinchona
alkaloid-derived primary amine as the catalyst, benzoyl peroxide as the oxygen source, and stoichio-
metric amounts of the aldehyde substrates, providing simple metal-free access to valuable protected 2-
hydroxyaldehyde derivatives.
Keywords:
Asymmetric catalysis
Organocatalysis
Ó 2012 Published by Elsevier Ltd.
Enamine activation
Dienamine activation
Benzoyloxylation
1. Introduction
a useful, readily available and inexpensive reagent for the asym-
metric
In the previously reported aldehyde a-benzoyloxylations, bulky
secondary amine catalysts were used, presumably to avoid catalyst
decomposition by oxidation.2c,5c,6 In contrast, cinchona alkaloid-
derived primary amines, recently used as catalysts for the asymmet-
a-benzoyloxylation of simple a
-unbranched aldehydes.5
Given the prevalence of chiral oxygenated compounds among
natural products and the utility of oxygen functionalities as handles
for chemical transformations, asymmetric oxidation represents an
important area of research. In this context, a number of enantio-
selective metal-based approaches have been developed for the
oxygenation of carbonyl substrates.1 Recently, enamine catalysis
has emerged as a powerful alternative for the direct asymmetric
a
-
ric epoxidation of enones and a-branched enals, demonstrated ex-
cellent enantioselectivity and reactivity towards challenging
substrate classes, even in the presence of a large excess of an oxidant.7
Nevertheless, while numerous secondary amine catalysts have been
a
-
oxygenation of ketones and aldehydes, which avoids the need for
a pre-formation of an enolate equivalent.2 In particular, the sec-
employed in the a
-oxygenation of aldehydes and ketones,2 the use of
ondary amine-catalyzed
-unbranched aldehydes with nitrosobenzene has been established
as a reliable oxidation method.3 Nevertheless, important limita-
tions of this methodology remain. In particular, aminoxylation of
a
-aminoxylation of aliphatic ketones and
primary amines in this transformation has been scarce.8 Based on
these considerations, we have recently developed highly enantiose-
a
lective a-benzoyloxylation of stoichiometric amounts of cyclic ke-
a
-
tones catalyzed by the readily available cinchona alkaloid-derived
primary amines [Eq. 3].8a Most recently, the group of Bencivenni
reported similar results although using an excess of ketones.8c
branched aldehydes invariably results in mixtures of products,
typically favouring the undesired nitrosoaldol reaction and
affording
lectivity [Eq. 1].4 To the best of our knowledge, there are no reports
of organocatalytic enantioselective reactions of -branched alde-
hydes with other oxygen sources, including benzoyl peroxide [Eq.
2], which has been recently introduced by Maruoka and others as
a
-oxygenated products with only modest enantiose-
We further envisioned that asymmetric
the challenging -branched aldehydes can be similarly amenable to
primaryamine catalysis [Eq. 4]. In addition, we considered the use of
-branched enals as nucleophilic -carbon synthons via dienamine
activation, which would afford the corresponding products with
additional -unsaturation [Eq. 5]. Herein, we report our progress
on the asymmetric -benzoyloxylation of stoichiometric amounts of
-branched aldehydes and -branched enals, which gives access to
valuable aldehydes containing a quaternary stereocenter.
a-benzoyloxylation of
a
a
a
a
b,g
a
a
a
* Corresponding author. Fax: þ49 208 306 2999; e-mail address: list@mpi-
muelheim.mpg.de (B. List).
0040-4020/$ e see front matter Ó 2012 Published by Elsevier Ltd.
http://dx.doi.org/10.1016/j.tet.2012.06.043

N. Demoulin et al. / Tetrahedron 68 (2012) 7568e7574
α-Aminoxylation vs. nitrosoaldol of α-branched aldehydes
7569
enantioselectivity of the reaction to 70:30 er (entry 10). Decreasing
the concentration was detrimental for the conversion without af-
fecting the enantioselectivity (entries 11e12). Other related
cinchona-derived amines 11e14 were evaluated without any acid
co-catalyst and gave similar results compared to the catalyst 10
(entries 13e16).
O
*
H
Ph
N
R1 R2
O
O
N
R2
cat.
*
O
*
+
H
(1)
NHPh
OH
H
H
R1
R1 R2
O
Ph
N
O:N = 1:0.6-1:99
er 72.5:27.5
≤
Unknown
*
O
O
N
H
cat.
Table 1
R2
OBz
(2)
H
H
Evaluation of the reaction conditions for the asymmetric benzoyloxylation of alde-
hyde 1a
R1 R2
BzO OBz
R1
Asymmetric α-benzoyloxylation of cyclic ketones
O
O
OBz
H2N
BzO OBz
cat.
*
(3)
R
R
n
n
α-Benzoyloxylation of α-branched aldehydes
O
O
H2N
BzO OBz
R2
cat.
R2
*
(4)
(5)
H
H
R1 OBz
R1
α-Benzoyloxylation of α-branched enals
O
O
R2
H2N
BzO OBz
cat.
*
H
H
R1 OBz
R1 R2
Entry
Catalyst (amine, acid)
Concn [M]
Yieldb [%]
erc
1
2
3
4
5
6
7
8
5d
6
7
8
1
1
1
1
1
2
2
2
1
1
1
0.5
1
2
2
2
6
8
d
2.
a
-Benzoyloxylation of
a-branched aldehydes
d
47
49
25
90
90
93
78f
83
80
70
80
85
85
90
42:58
39:61
19.5:80.5
50:50
65:35
56:44
70:30
52:48
65:35
65:35
68:32
64:36
65:35
35:65
We began our investigation using the commercially available
9
racemic 2-(4-tert-butylbenzyl)propionaldehyde (1a) as a model
substrate and carried out the benzoyloxylation reaction with
a slight excess (1.2 equiv) of anhydrous benzoyl peroxide (2) in THF
at room temperature and 10 mol % of the radical inhibitor 2,6-di-
tert-butyl-4-methylphenol (BHT, 4) to prevent possible benzoyl
radical side reactions. Initial catalyst studies (Table 1) confirmed
our expectation that the commonly used secondary amines are
ineffective catalysts, presumably due to decomposition.5c,6 Thus,
10 mol % of (S)-proline (5) and the pyrrolidine derivative 6 gave the
desired product 3a with 6% and 8% conversion, respectively, after
39 h, which did not increase over the next 5 days (entries 1e2). In
the case of pyrrolidine-based catalysts 7 and 8, the latter being used
10$Cl3CCO2H
10
10$(PhO)2PO2H
9
10$(S)-TRIPe
10
11
12
13
14
15
16
10 (R)-TRIPe
10
10
11
12
13
14
a
BHT¼2,6-di-tert-butyl-4-methylphenol (radical scavenger).
b
c
Determined by 1H NMR spectroscopy using an internal standard.
Determined by chiral-phase HPLC analysis after reduction with NaBH4.11
The reaction was carried out in DMSO to solubilize the catalyst.
by Maruoka for a highly enantioselective
a-benzoyloxylation of
d
e
unbranched aldehydes,5c,9 both gave better conversion after 5 days
and a similar enantioselectivity of about 60:40 er (entries 3e4).
Interestingly, the JørgenseneHayashi catalyst 9 gave the best result
in terms of enantioselectivity with 80.5:19.5 er albeit a low con-
version of 25% after 5 days (entry 5). In contrast, the previously
TRIP¼3,30-bis(2,4,6-triisopropylphenyl)-1,10-bi-2-naphthol
cyclic
mono-
phosphate.
f
Isolated yield of the product.
established cinchona amine salt [10$TCA]8a catalyzed the
a-ben-
In view of these preliminary results, we decided to pursue the
benzoyloxylation of -branched aldehydes further using primary
amine catalysis, which furnished the desired products with the best
yields. Given the modest enantioselectivity observed in the -ben-
a-
a
zoyloxylation of the aldehyde 1a with an excellent NMR yield of
90% in 15 h, albeit in nearly racemic form (entry 6). However, when
we excluded the acid co-catalyst, the reaction still proceeded with
the same conversion, but a significantly higher enantioselectivity
(entry 7). Although these results suggested that an external acid
additive was unnecessary, we hypothesized that the primary
a
zoyloxylation of aldehyde 1a under various conditions, we turned
our attention to the mechanism of the reaction. It is well known that
in reactions operating under enamine catalysis, effective control of
the enamine geometry is crucial for enantioselectivity,12 since (E)
and (Z) enamine isomers typically furnish opposite enantiomers of
the product (Scheme 1). Under thermodynamic conditions of
organocatalysis, substrates with a low steric and/or electronic bias
might form equilibrium mixtures of (E) and (Z) enamine in-
termediates, which are similar in energy. Under these circum-
stances, unless one of the enamine isomers reacts at a significantly
higher rate, even the most facially selective catalysts can produce
racemic or nearly racemic products. Therefore, it is not surprising
that almost all highly enantioselective methodologies employing
amine-catalyzed benzoyloxylation of
a-branched aldehydes still
involves a Brønsted acid in the catalytic cycle, and that this role is
fulfilled by the in situ generated benzoic acid after the first catalytic
turnover. Based on these considerations, we screened several
Brønsted acid co-catalysts, including both enantiomers of the chiral
phosphoric acid TRIP in the hope of influencing the enantiose-
lectivity of the reaction.10 Although both the achiral phosphoric
acid diphenyl ester (entry 8) and (R)-TRIP (entry 9) resulted in
reduced enantioselectivity of the product 3a, (S)-TRIP appeared to
be a ‘matched’ combination with the catalyst 10 and increased the

7570
N. Demoulin et al. / Tetrahedron 68 (2012) 7568e7574
Table 2
enamine catalysis with
a
-branched aldehydes rely on sterically bi-
-aryl substituents.12b In
Evaluation of the reaction conditions for the asymmetric benzoyloxylation of 2-
phenylpropanal 1b
ased substrates, such as aldehydes with
this respect, aldehyde 1a represents a less biased substrate, as the
electronically and sterically similar -methyl substituent and the
a
a
methylene group of the main chain likely result in (E) and (Z) en-
amine intermediates B and C having very similar energies.
Entry Catalyst (amine, acid) Solvent
Temp [ꢀC] Yieldb [%] erc
1
2
3
4
5
6
7
8
9
10
THF
THF
THF
THF
THF
THF
THF
THF
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
83
95
76e
95
95
95
95
81e
95
95
95
95
95
83
39g
43:57
76.24d
84:16
10$(S)-TRIP
10$(R)-TRIP
14$(R)-TRIP
11
11$(S)-TRIP
11$(R)-TRIP
10
10
10
10
10
10
10
42:58f
22:78f
76:24
82:18
15:85f
73:27
9
CHCl3
PhMe
MeCN
Et2O
1,4-Dioxane rt
THF
THF
10
11
12
13
14
15
73:27
62:38
66.5:33.5
72:28
76.5:23.5
84:16
Scheme 1. Mechanistic considerations with the sterically unbiased substrate 1a (ar-
bitrary stereochemical assignment).
5
ꢁ15
Thus, in order to estimate more accurately the facial selectivity
of cinchona-derived primary amine catalysts in the
a-benzoylox-
a
BHT¼2,6-di-tert-butyl-4-methylphenol (radical scavenger).
ylation of -branched aldehydes, we evaluated the more sterically
a
b
c
Determined by 1H NMR spectroscopy using an internal standard.
biased 2-phenyl acetaldehyde 1b as a substrate (Table 2). Gratify-
ingly, in the absence of an acid co-catalyst, both quinine-derived
catalyst 10 (entry 2) and cupreine-derived catalyst 11 (entry 6)
catalyzed the formation of the corresponding product 3b with
complete conversion and an excellent NMR yield. In agreement
with our expectations, compared to 3a, the product 3b was also
obtained with a higher enantioselectivity using either catalyst,
whereby amines 10 and 11 gave similar results (75:25 er). In-
terestingly, in this case, the HayashieJørgensen catalyst 9 gave the
product with good conversion but a low enantioselectivity of 57:43
er (entry 1). Once again, an external acid additive was found to be
unnecessary for the reaction to proceed, but chiral acids affected
the enantioselectivity. Thus, (S)-TRIP was found to be a ‘matched’
combination with the catalyst 10 as before (entry 3), and (R)-TRIP
was a ‘matched’ combination with the pseudoenantiomeric amine
catalyst 14 (entry 5). In contrast, (R)-TRIP clearly appeared to be
a ‘mismatched’ combination with the catalyst 10, resulting in ero-
sion of the product’s enantiomeric excess (entry 4) presumably due
to the conflicting sense of selectivity of the two chiral catalysts. The
situation proved different with the cupreine-derived catalyst salts
(entries 7e8). While (S)-TRIP in combination with the cupreine-
derived catalyst 11 increased the enantioselectivity of the re-
action to 82:18 er (entry 7), to our surprise, (R)-TRIP was also
a ‘matched’ combination and furnished the opposite enantiomer of
the product 3b with the highest er of 85:15.13
Determined by chiral-phase HPLC analysis after reduction with NaBH4.11
d
The absolute (R)-configuration of the product 3b was established by comparison
of the optical rotation with the literature value after global reduction to 2-
phenylpropane-1,2-diol with LiAlH4.1f,11
e
Isolated yield of the product.
Product obtained with the opposite configuration.
The reaction was carried out for 7 days.
f
g
generated, which might form a more active catalyst salt 10$2BzOH.
An attempt to quench the benzoic acid formed in situ by adding
solid K2CO3 resulted in decrease of the enantioselectivity.11
100
90
80
70
60
50
40
30
20
10
0
0
2
4
6
8
10
Time, h
12
14
16
18
20
Following a screen of solvents, THF was kept as the optimal
reaction medium (entries 2 and 9e13). Decreasing the temperature
improved the enantioselectivity of the reaction to the detriment of
the conversion (entries 1 and 14e15).
Fig. 1.
a-Benzoyloxylation of 1b with catalyst 10: a plot of conversion over time,
showing autocatalytic behaviour.
Using catalyst 10 without any Brønsted acid additive (Table 2,
entry 2), we also followed the conversion of the aldehyde 1b over
time by using 1H NMR spectroscopy. Plotting the results, we
obtained a sigmoidal curve characteristic to autocatalytic reactions,
which contained an initial lag period until approximately 20%
conversion of the starting material (Fig. 1). This autocatalytic be-
haviour supports our hypothesis on the role of benzoic acid, which
is generated in situ during the reaction and participates in catalysis.
In particular, at 20% conversion, 20 mol % of benzoic acid is
Since the kinetic study confirmed that one or several molecules
of acid participate in the catalytic cycle of the reaction, we were
encouraged to further test chiral Brønsted acid additives to improve
the enantioselectivity of the process. In addition, the results
obtained with the substrate 1b pointed out the influence of the
substrate’s steric bias on the enantioselectivity of the reaction. To
understand the relative influence of these factors, we decided to
perform a comprehensive analysis of various substrates, screening

N. Demoulin et al. / Tetrahedron 68 (2012) 7568e7574
7571
Table 3
them against different acid co-catalysts and amine catalysts 10 and
Evaluation of different substrates 1 in the
11 and various acid co-catalysts
a
-benzoyloxylation with catalysts 10 and
11. As substrates, we chose aldehydes including a protected alcohol
functionality in the main chain to generate an electronic bias,
which might also serve as a coordination point in the catalytic as-
sembly (Scheme 2a). A screening matrix using substrates 1aee,
amine catalysts 10 and 11 and different chiral acid co-catalyst is
summarized in Table 3.
Entryc
R
Amine Acid co-catalystb
None
(S)-TRIP
(R)-TRIP
1
2
10
11
68:32 (90%) 70:30 (78%)c
52:48 (83%)
68:32 (80%) 63.5:36.5 (30%) 50:50 (59%)
3
4
10
11
76:24 (100%) 84:16 (76%)d
76:24 (95%) 82:18 (95%)
42:58 (95%)
15:85 (83%)d
5
6
7
10
10
10
78:22 (76%) d (0%)
70:30 (50%)
82:18 (88%) 90:10 (72%)
72:28 (63%) 88:12 (90%)
57.5:42.5 (43%)
39.5:60.5 (75%)
8
11
76:24 (95%) 84:16 (90%)
23:77 (85%)
a
BHT¼2,6-di-tert-butyl-4-methylphenol (radical scavenger).
b
c
Scheme 2. Possible involvement of the oxygen moiety in substrates
1 as a co-
For a screen of other chiral acids, see Supplementary data.
Yields of products 3 (in brackets) were determined by 1H NMR spectroscopy
ordination site in the catalyst assembly.
with an internal standard, and er values were determined by chiral-phase HPLC
after reduction with NaBH4.
d
Isolated yield of the product.
Concomitant formation of methacrylic aldehyde by elimination of the protected
In agreement with our expectations, all sterically and/or elec-
tronically biased substrates 1bee afforded the corresponding
products with a higher enantioselectivity than the model substrate
1a in the absence of an acid co-catalyst (column 1, entries 3e8 vs
1e2). Gratifyingly, employing (S)-TRIP as the acid co-catalyst with
either catalyst 10 or 11 improved the enantioselectivity of all re-
actions (column 2, entries 1e8). In contrast, (R)-TRIP was optimal
only when combined with the cupreine-derived catalyst 11 (col-
umn 3, entries 4 and 8), furnishing the opposite enantiomer of the
products 3.
e
b
-hydroxy substituent was observed.
8e11) and aliphatic (entries 1 and 4e7) substrates cleanly un-
derwent -benzoyloxylation, affording the desired products with
a
good yields and useful enantioselectivities at the newly created
quaternary stereocenter. Importantly, the opposite enantiomer of
products 3 could be accessed with comparable efficiency using (R)-
TRIP combined with the cupreine-derivative 11 or the quinidine
derivative 14 (entries 3 and 6). Comparison of the ortho- and para-
methoxy-substituted substrates 1g and 1h (entries 8e9) un-
derscores the importance of an oxygen moiety as a potential co-
ordination site in the catalyst assembly. Indeed, a methoxy group in
the ortho position of the aromatic ring of aldehyde 1g dramatically
improved the enantioselectivity of the reaction (entry 8) compared
to its para-substituted analogue (entry 9). To avoid the problem of
irreversible elimination observed in substrates 1ced, we tested
tetrahydro-3-furancarbaldehyde (1f), which contains an oxygen
Unfortunately, both substrates 1c (entry 3) and 1d (entry 4)
were prone to elimination of the protected b-hydroxy substituent,
leading to the formation of methacrylic aldehyde in various
amounts.11 Presumably, this formal retro-oxa-Michael reaction was
promoted by enamine catalysis with further activation of the
leaving group by the acid co-catalyst (Scheme 2a). Interestingly, the
amount of elimination product observed with substrate 1c (entry 5)
strongly depended on which enantiomer of the acid co-catalyst
TRIP was employed. In the ‘mismatched’ case with (R)-TRIP, the
elimination side product amounted to 46% of the total yield,
whereas with (S)-TRIP, it was the only product of the reaction
(92%). These results are in agreement with our hypothesis that the
oxygen moiety in the substrate might contribute as a coordination
site in the catalyst assembly, which leads to both the desired re-
atom in the
However, this substrate proved to be poorly reactive, with con-
version below 15%,11 presumably due to a competitive
-alkoxy
b-position but cannot undergo fragmentation (entry 7).
b
elimination/cyclization equilibrium (Scheme 2b). To improve con-
version, we used the catalytic salt [10$3TCA] that we previously
found to be more active.8a While this salt gave better conversion,
only very low enantioselectivity was observed (entry 7), which
could not be improved with a chiral acid co-catalyst.11 It is note-
worthy that, in contrast to the substrates 1cee,g, the oxygen moiety
in the substrate 1f is locked in the tetrahydrofuran backbone and
therefore might not be in the vicinity of the reactive site of the
action pathway forming a-benzoyloxylated product 3 and the un-
desired elimination leading to methacrylic aldehyde as the side
product (Scheme 2a).
Using the best catalyst salt 10$(S)-TRIP, we explored the scope of
the asymmetric
a-benzoyloxylation with respect to a-branched
aldehyde substrates (Table 4). Both aromatic (entries 2e3 and

7572
N. Demoulin et al. / Tetrahedron 68 (2012) 7568e7574
catalyst assembly for potential coordination (Scheme 2b). In-
terestingly, the naphthyl substrate 1j gave good enantioselectivity
without any acid co-catalyst, which was not improved with a chiral
acid co-catalyst (entry 11).11 Aldehydes with bulkier substituents
3. a-Benzoyloxylation of a-branched aldehydes
In parallel with our investigations on the
-branched aldehydes via enamine catalysis, we also explored
a related transformation of -unsaturated -branched aldehydes
a-benzoyloxylation of
a
than a methyl group in the
tion of this method, as they undergo
a
position currently represent a limita-
-benzoyloxylation with poor
a,
b
a
a
via dienamine catalysis. In particular, we envisioned that enals 15,
which readily condense with the catalyst 10 to form the iminium
species A, can tautomerize to dienamine B and as such react with
benzoyl peroxide (Scheme 4). Having vinylogous character, dien-
conversion (<40%), possibly owing to increased steric congestion at
the reactive site.11
amines can act either as a- or g-carbon nucleophiles (Scheme 3).
Although examples of both are frequently encountered in amino-
catalytic reactions, factors, which govern this dual behaviour are
poorly understood, and systematic studies of regioselectivity in
dienamine catalysis have been scarce.14 Based on literature ana-
lysis,14d it is tempting to speculate that, while the regioselectivity of
Table 4
Scope of
a
-benzoyloxylation of a-branched aldehydes 1 with the catalyst salt 10$(S)-
TRIPa
unbranched
-unsaturated carbonyls is case-specific,14aef
branched a,b-unsaturated carbonyl substrates prefer to avoid
a,
b
a or g-
forming a quaternary stereocenter if given a choice, and select the
most unencumbered position for a nucleophilic attack.14gei Indeed,
aside from a single example recently disclosed by Jørgensen,14a all
Entry
1
1
3
Yield [%]
78
erc
reported
g
-branched
a
,
b
-unsaturated carbonyl compounds appear
-branched ones seem to
g
-functionalization14b,h or simply fail to react.14d
-functionalization,14d,g,i while
a
1a
70:30
to favour
a
favour
2
1b
1b
76f
83
84:16d
12:88
3e
4
1d
1e
1e
1f
57f
78
79
41
91.5:8.5
88:12
11:89
44:56
Scheme 3. Benzoyloxylation of enals 15 via dienamine catalysis.
5
6g
7h
As a model substrate, we chose the commercially available cyclic
enal 15a. In order to identify a suitable catalytic system, we
screened various amine catalysts at 10 mol % loading together with
different acid co-catalysts (Table 5). Substrate 15a was used with
a slight excess of anhydrous benzoyl peroxide (1.2 equiv) in THF at
1 M concentration and room temperature, in the presence of the
radical inhibitor BHT (4). Carrying out the reaction with the achiral
amine catalyst meso-18 and an equal amount (10 mol %) of Brønsted
8
1g
1h
1i
76
70
77
84:16
acid TCA (entry 1) afforded a mixture of
a and g products 16a and
17a in a modest yield, favouring the -product with an 84:16 ratio.
a
9
71:29
Excluding the acid co-catalyst (entry 2) shut down the reaction,
presumably because the acid is required to catalyze iminium-
dienamine tautomerization (Scheme 3). Employing the chiral ver-
sion of the catalyst (R,R)-18 with TCA gave the product in a similar
10
79.5:20.5
yield and a slightly lower a/g ratio, and only very low enantiose-
lectivity (entry 3). Pairing this catalyst with the chiral phosphoric
acid TRIP failed to improve the enantioselectivity, although oppo-
site enantiomers of the major product 16a were observed with (R)-
11i
1j
69
75:25
and (S)-TRIP (entries 4e5). A screen of a L-
-amino acids11 identified
tert-leucine as a promising catalyst in presence of 10 mol % TCA,
which gave an improved conversion of 74% and 66:34 er (entry 6).
When we switched the catalyst to cupreine-derived amine 11 to-
gether with 30 mol % TCA, both conversion and enantiomeric ex-
cess of the product 16a were diminished, although the a/g ratio was
improved (entry 7). Unfortunately, pairing amine 11 with either
enantiomer of the BINOL-derived chiral phosphoric acid 19 failed to
a
Reactions performed on 0.2 mmol or 0.4 mmol scale.
BHT¼2,6-di-tert-butyl-4-methylphenol (radical scavenger).
Determined by chiral-phase HPLC analysis.11
b
c
d
The absolute (R)-configuration of the product 3b was established by comparison
of the optical rotation with the literature value after global reduction to 2-
phenylpropane-1,2-diol with LiAlH4.1f,11
e
Using catalyst salt [11$(R)-TRIP].
Methacrylic aldehyde was observed as a side product.
Using catalyst salt [14$(R)-TRIP].
f
improve the enantioselectivity and
highest selectivity and enantioselectivity were once again
obtained with the quinine-derived catalyst 10 and TCA as the acid
a/g ratios (entries 8e9). The
g
h
a/g
Using catalyst salt [10$3TCA].
i
Using no acid co-catalyst.11

N. Demoulin et al. / Tetrahedron 68 (2012) 7568e7574
7573
Table 6
co-catalyst (entry 10). Excluding the acid from this reaction resul-
ted in no conversion of the starting material (entry 11).
Scope of
a
-benzoyloxylation of
a
,
b
-unsaturated
a
-branched aldehydes 15bef with
the catalyst salt [10$3TCA]a
Table 5
Identification of a suitable catalytic system for the benzoyloxylation of enal 15a
Entry
1
15
16
16/17c
Yield (16) [%]
erd (16)
15a
90:10
65
79.5:20.5
2e
3
15a
15b
15c
15d
15e
15f
90:10
>99:1
>99:1
>99:1
>99:1
>99:1
78
23.5:76.5
Entry
Amine
Acid (mol %)
16a/17a
Yieldb (16a) [%]
erc (16a)
74 (11%)
74
74:26 (41.5:58.5)
76.5:23.5
1
2
3
4
5
6
7
8
meso-18
meso-18
(R,R)-18
(R,R)-18
(R,R)-18
(S)-t-Leu
11
11
11
10
10
TCA (10)
None
86:14
d
54
0
d
d
TCA (10)
(R)-TRIP (10)
(S)-TRIP (10)
TCA (10)
TCA (30)
(R)-19 (20)
(S)-19 (20)
TCA (30)
None
80:20
80:20
76:24
80:20
89:11
85:15
85:15
89:11
d
45
37
33
74
68
60
57
74
0
54:46
43:57
54:46
66:34
63:37
56:44
50:50
76:24
d
4
5
51f
74:26
9
10
11
a
BHT¼2,6-di-tert-butyl-4-methylphenol (radical scavenger).
Determined by 1H NMR spectroscopy using an internal standard.
Determined by chiral-phase HPLC analysis.
6g
68 (30%)
65
73:27 (42:58)
55:45h
b
c
7
Unfortunately, extensive screening of the Brønsted acid co-
catalysts failed to improve the enantioselectivity of the reaction.11
Following a screen of solvent, concentration and reaction time,11
we found that carrying out the reaction in THF at 0.5 M concen-
tration at room temperature for 48 h gave the best results. Using
a
Reactions performed on 0.2 mmol or 0.4 mmol scale.
b
c
BHT¼2,6-di-tert-butyl-4-methylphenol (radical scavenger).
The a/g
ratio was determined by analysis of the crude reaction mixture by 1H
NMR spectroscopy.
these conditions, we explored the scope of the asymmetric
benzoyloxylation with respect to -unsaturated -branched al-
dehyde substrates (Table 6). Both cyclic (entries 1e2) and acyclic
(entries 3e6) substrates cleanly underwent -benzoyloxylation,
affording the desired products with good yields and useful enan-
tioselectivities. The opposite enantiomer of 16a could be accessed
with comparable efficiency using the quinidine-derived catalyst
salt 14 (entry 2). Interestingly, the acyclic substrates 15bef reacted
a-
d
Determined by chiral-phase HPLC analysis.11
Using catalyst salt [14$(R)-TRIP].
Incomplete conversion was observed and 40% of the starting aldehyde 15d was
a
,b
a
e
f
a
also isolated.
g
The (E) geometry of the double bond in 16e was confirmed by NOESY.11
The reaction was carried out at 0 ꢀC for 48 h.
h
with greater a-selectivity than the cyclic substrate 15a (>99:1 vs
90:10). However, although products 17bef were not observed by
1H NMR spectroscopy analysis of the crude reaction mixture, they
were sometimes detected on TLC and isolated after purification on
silica gel with eroded enantioselectivity compared to their
a-
Scheme 4. Benzoyloxylation of linear enone 20 with amine catalyst 10.
substituted counterparts (entries 3 and 6). A potential explanation
for this observation could be a silica gel-mediated allylic rear-
rangement of the benzoyl group.15 Substrates bearing an alkyl
group in the
a position (entries 1e6) reacted with very similar
Considering the highly selective
reported by Melchiorre and colleagues,14b we were surprised by the
clear -selectivity of our benzoyloxylation reaction, despite very
similar catalyst systems (Scheme 5a). Wishing to study this be-
haviour, we repeated the amination reaction of 15a, and could re-
produce the yield and enantiomeric ratio of the product, which was
formed with very high selectivity. However, the reported structure
of the product was found to be erroneous. Indeed, 2D-NMR spec-
troscopy and X-ray analysis allowed us to unambiguously assign
g-amination of 15a recently
enantioselectivities (74:26e79.5:20.5 er) and required 48 h to
proceed to completion, except substrate 15d, which was benzoy-
loxylated with incomplete conversion (entry 5). In contrast, the
a
aldehyde 15f that bears a phenyl group in the
a position was very
reactive and proceeded even without any acid co-catalyst, although
with low enantioselectivity (entry 7). The high reactivity of 15f
might be attributed to a facile dienamine formation assisted by
conjugation with the phenyl group. Unfortunately, lowering the
reaction temperature to 0 ꢀC with this substrate failed to signifi-
cantly improve the enantioselectivity.11
We also tested enone 20 under the optimal conditions. However,
thissubstrateprovedtobeinerttobenzoyloxylation, andonlystarting
material was recovered after 48 h of reaction time (Scheme 4).
the product as the
configuration was secured from the X-ray structure (Scheme 5b).11
Interestingly, although the -selectivity was confirmed to be con-
sistent in both amination and benzoyloxylation of enals 15a, the
former reaction furnished the product with much higher
a-aminated compound 21b, whose absolute
a
a

7574
N. Demoulin et al. / Tetrahedron 68 (2012) 7568e7574
References and notes
1. For recent reviews, see: (a) Fukuda, T.; Katsuki, T. Tetrahedron Lett. 1996, 37,
€
4389; (b) Adam, W.; Fell, R. T.; Stegmann, V. R.; Saha-Moller, C. R. J. Am. Chem.
Soc. 1998, 120, 708; (c) Janey, J. M. Angew. Chem., Int. Ed. 2005, 44, 4292; (d)
Smith, A. M. R.; Hii, K. K. Chem. Rev. 2011, 111, 1637; (e) DeBergh, J. R.; Spivey, K.
M.; Rea, J. M. J. Am. Chem. Soc. 2008, 130, 7828; (f) Gourdet, B.; Lam, H. W.
Angew. Chem., Int. Ed. 2010, 49, 8733.
2. For recent reviews of organocatalyzed asymmetric a-oxygenation of carbonyls,
see: (a) Xu, L.-W.; Li, L.; Shia, Z.-H. Adv. Synth. Catal. 2010, 352, 243; (b) Vilaivan,
T.; Bhanthumnavin, W. Molecules 2010, 15, 917; (c) Guillena, G.; Ramon, D. J.
Curr. Org. Chem. 2011, 15, 296.
3. For seminal publications, see: (a) Brown, S. P.; Brochu, M. P.; Sinz, C. J.; Mac-
Millan, D. W. C. J. Am. Chem. Soc. 2003, 125, 10808; (b) Hayashi, Y.; Yamaguchi, J.;
Hibino, K.; Shoji, M. Tetrahedron Lett. 2003, 44, 8293; (c) Zhong, G. Angew. Chem.
, Int. Ed. 2003, 42, 4247; (d) Boegevig, A.; Sunden, H.; Cordova, A. Angew. Chem.,
Int. Ed. 2004, 43, 1109; (e) Cordova, A.; Sunden, H.; Boegevig, A.; Johansson, M.;
Himo, F. Chem.dEur. J. 2004, 10, 3673; (f) Hayashi, Y.; Yamaguchi, J.; Sumiya, T.;
Hibino, K.; Shoji, M. J. Org. Chem. 2004, 69, 5966; (g) Hayashi, Y.; Yamaguchi, J.;
Sumiya, T.; Shoji, M. Angew. Chem., Int. Ed. 2004, 43, 1112; (h) Momiyama, N.;
Torii, H.; Saito, S.; Yamamoto, H. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 5374.
4. (a) Guo, H.-M.; Cheng, L.; Cun, L.-F.; Gong, L.-Z.; Mi, A.-Q.; Jang, Y.-Z. Chem.
Commun. 2006, 428; (b) Kim, S.-G.; Park, T.-H. Tetrahedron Lett. 2006, 47, 9067.
Scheme 5. (a) Selectivity in the amination of enal 15a as reported,14b and after re-
assignment by us (b). X-ray: red¼oxygen, blue¼nitrogen, grey¼carbon,
white¼hydrogen.
5. (a) For a general, racemic a-benzoyloxylation of aldehydes and ketones using
N-methyl-O-benzoylhydroxylamine reagent, see: Beshara, C. S.; Hall, A.; Jen-
kins, R. L.; Jones, K. L.; Jones, T. C.; Killeen, N. M.; Taylor, P. H.; Thomas, S. P.;
Tomkinson, N. C. O. Org. Lett. 2005, 7, 5729; (b) Gotoh, H.; Hayashi, Y. Chem.
Commun. 2009, 3083; (c) Kano, T.; Mii, H.; Maruoka, K. J. Am. Chem. Soc. 2009,
131, 3450; (d) Vaismaa, M. J. P.; Yau, S. C. Y.; Tomkinson, N. C. O. Tetrahedron Lett.
2009, 50, 3625; (e) Also see: benzoyloxylation of oxindoles using a calcium
phosphate catalyst: Zhang, Z.; Zheng, W.; Antilla, J. C. Angew. Chem., Int. Ed.
2011, 50, 1135.
6. Denney, D. B.; Denney, D. Z. J. Am. Chem. Soc. 1960, 82, 1389.
7. (a) Reisinger, C. M.; Wang, X.; List, B. Angew. Chem., Int. Ed. 2008, 47, 8112; (b)
Wang, X.; Reisinger, C. M.; List, B. J. Am. Chem. Soc. 2008, 130, 6070; (c) Lifchits,
O.; Reisinger, C. M.; List, B. J. Am. Chem. Soc. 2010, 132, 10227.
8. (a) Lifchits, O.; Demoulin, N.; List, B. Angew. Chem., Int. Ed. 2011, 50, 9680; (b)
Qin, L.; Li, L.; Yi, L.; Da, C.-S.; Zhou, Y.-F. Chirality 2011, 23, 527; (c) Jadhav, M. S.;
Righi, P.; Marcantoni, E.; Bencivenni, G. J. Org. Chem. 2012, 77, 2667.
9. Klumpp, D. A.; Aguirre, S. L.; Sanchez, G. V.; de Leon, S. J. Org. Lett. 2001, 3, 2781.
10. Liu, C.; Zhu, Q.; Huang, K.-W.; Lu, Y. Org. Lett. 2011, 13, 2638.
11. See Supplementary data for further details.
12. (a) Fu, N.; Zhang, L.; Li, J.; Luo, S.; Cheng, J.-P. Angew. Chem., Int. Ed. 2011, 50,
11451; (b) Mukherjee, S.; Yang, J. W.; Hoffmann, S.; List, B. Chem. Rev. 2007, 107,
5471.
enantioselectivity. Presumably, the greater steric bulk of the ami-
nating reagent compared to benzoyl peroxide is responsible for the
higher enantiocontrol.
In conclusion, we have developed a practical methodology for
the
enals. Our method delivers benzoyl-protected hydroxyaldehydes 3
and 16 containing an -quaternary stereocenter with good yields
a-benzoyloxylation of a-branched aldehydes and a-branched
a
and useful enantioselectivities up to 91.5:8.5 er. This process is
mediated by readily available primary amine catalyst salts and
employs stoichiometric amounts of various
and
-branched enals. To the best of our knowledge,14a
loxylation of -branched enals via dienamine catalysis is, to date,
the first example of -functionalization of -branched enals. New
a
-branched aldehydes
a
a
-benzoy-
a
a
a
catalysts are being investigated in this transformation and will be
reported in due course.
13. (a) Moteki, S. A.; Han, J.; Arimitsu, S.; Akakura, M.; Nakayama, K.; Maruoka, K.
Angew. Chem., Int. Ed. 2012, 51, 1187; (b) Tian, X.; Cassani, C.; Liu, Y.; Moran, A.;
Urakawa, A.; Galzerano, P.; Arceo, E.; Melchiorre, P. J. Am. Chem. Soc. 2011, 133,
17934.
Supplementary data
14. (a) Albrecht, q.; Dickmeiss, G.; Acosta, F. C.; Rodríguez-Escrich, C.; Davis, R. L.;
Jørgensen, K. A. J. Am. Chem. Soc. 2012, 134, 2543; (b) Bencivenni, G.; Galzerano,
P.; Mazzanti, A.; Bartoli, G.; Melchiorre, P. Proc. Natl. Acad. Sci. U.S.A. 2010, 107,
20642; (c) Bertelsen, S.; Marigo, M.; Brandes, S.; Diner, P.; Jørgensen, K. A. J. Am.
Chem. Soc. 2006, 128, 12973; (d) Stiller, J.; Marques-Lopez, E.; Herrera, R. P.;
Supplementary data that may be helpful in the review process
should be prepared and provided as a separate electronic file. That
file can then be transformed into PDF format and submitted along
with the manuscript and graphic files to the appropriate editorial
office. Supplementary data associated with this article can be found
in the online version, at http://dx.doi.org/10.1016/j.tet.2012.06.043.
These data include MOL files and InChIKeys of the most important
compounds described in this article.
€
Frohlich, R.; Strohmann, C.; Christmann, M. Org. Lett. 2011, 13, 70; (e) Talavera,
G.; Reyes, E.; Vicario, J. L.; Carrillo, L. Angew. Chem., Int. Ed. 2012, ; (f) Utsumi, N.;
Zhang, H.; Tanaka, F.; Barbas, C. F., III. Angew. Chem., Int. Ed. 2007, 46, 1878; (g)
Han, B.; Xiao, Y.-C.; He, Z.-Q.; Chen, Y.-C. Org. Lett. 2009, 11, 4660; (h) Bergon-
zini, G.; Vera, S.; Melchiorre, P. Angew. Chem., Int. Ed. 2010, 49, 9685; (i) Ram-
achary, D. B.; Ramakumar, K. Eur. J. Org. Chem. 2011, 14, 2599.
15. Serra-Muns, A.; Guerinot, A.; Reymond, S.; Cossy, J. Chem. Commun. 2010, 4178.