Substrate-controlled stereoselectivity in the Yamamoto aldol reaction

Article abstract of DOI:10.1039/c2ob26185e

The Yamamoto aldol reaction is a vinylogous aldol reaction that relies on bulky aluminium-based Lewis acids. These activate both the aldehyde as well as become part of the enolate moiety. The report discloses the first detailed study on the substrate-controlled Yamamoto aldol reaction in which 2,3-syn and 2,3-anti disubstituted aldehydes serve as the stereodirecting elements. The "size" of the substituent in the β-position strongly determines the facial selectivity of enolate addition to the aldehyde. Large substituents favour formation of 1,3-syn diols while slim alkynyl groups preferentially lead to 1,3-anti products. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c2ob26185e

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 7721
www.rsc.org/obc
PAPER
Substrate-controlled stereoselectivity in the Yamamoto aldol reaction†
Nadin Schläger and Andreas Kirschning*
Received 20th June 2012, Accepted 1st August 2012
DOI: 10.1039/c2ob26185e
The Yamamoto aldol reaction is a vinylogous aldol reaction that relies on bulky aluminium-based Lewis
acids. These activate both the aldehyde as well as become part of the enolate moiety. The report discloses
the first detailed study on the substrate-controlled Yamamoto aldol reaction in which 2,3-syn and 2,3-anti
disubstituted aldehydes serve as the stereodirecting elements. The “size” of the substituent in the
β-position strongly determines the facial selectivity of enolate addition to the aldehyde. Large substituents
favour formation of 1,3-syn diols while slim alkynyl groups preferentially lead to 1,3-anti products.
Introduction
Aldol reactions play an integral part in total synthesis programs
of polyketide-based natural products.1 Many methods have been
developed for the enantioselective construction of α-alkyl/
α-hydroxy and β-hydroxy carboxylic acid derivatives. Vinylo-
gous reactions are especially powerful as they open up opportu-
nities to build up extended polyketide-type fragments in a single
step (Scheme 1).2 A key method among the many vinylogous
aldol reactions is based on Mukaiyama’s use of silylenol ethers
2a (VMAR). In fact, a large number of asymmetric conditions
have been reported to date that include chiral auxiliaries (R2) as
well as chiral Lewis acids (L.A.) depending on the requirements
regarding starting materials and products.2,3 Classically, silylenol
ethers are preferred for the VAR but also lithium enolates 2b or
in one case the enolate generated in the presence of a preformed
chiral erbium (Er) Lewis acid 2c have been employed. In con-
trast, Yamamoto’s aldol reaction is less well established in
natural product synthesis.4 Previously, the method was employed
Scheme 1 Methods of vinylogous aldol reactions.
by Paterson5 et al. in the total synthesis of the callipeltoside
aglycone and by Sammakia and Abramite6 who used the Yama-
Commonly, 2 equiv. of the aluminium-based Lewis acid 7 are
being added in order to achieve metal exchange at the enolate
moiety 9 as well as activation of the aldehyde (A) (Scheme 1).
This scenario guarantees that enolate 9 will be sterically
masked at the α-position and instead will couple to the activated
aldehyde A at the remote terminus of the vinylogous enolate
moiety.
moto protocol for macrolactonisation with preformed esters. The
Yamamoto aldol reaction relies on the use of bulky Lewis acids
such as the C3-symmetric aluminium-tris-2,6-diphenylphenoxide
7 (ATPH) and lithium tetramethylpiperidine 8 (LTMP) as the
base, the latter being responsible for generating the lithium
enolate e.g. from unsaturated esters 5 or 6, respectively.
As part of our total synthesis program towards elansolid A
(10), a polyketide-based antibiotic from Chitinophaga sancti,7
we envisaged a Yamamoto aldol reaction for preparing the frag-
ment C1–C11 11 (Scheme 2). We planned to use the chirality
present in aldehyde 12 (representing C7–C11 of the eastern frag-
ment 11) to control the absolute configuration of the newly
formed stereogenic centre at C7 in the reaction with the enolate
derived from (2E,4E)-ethyl-4-methylhexa-2,4-dienoate (6).
Importantly, facile access to diene ester 6 is achieved by Wittig
Institute of Organic Chemistry and Center of Biomolecular Drug
Research (BMWZ), Schneiderberg 1b, 30167 Hannover.
E-mail: andreas.kirschning@oci.uni-hannover.de;
Fax: +49 (0)511 7623011; Tel: +49 (0)511 7624614
†Electronic supplementary information (ESI) available: Detailed pro-
cedures and relevant analytical data are given in the ESI. See DOI:
10.1039/c2ob26185e
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7721–7729 | 7721

View Article Online
auxiliary and oxidation of the intermediate alcohol to yield the
corresponding aldehydes. Experimental details and relevant
analytical data of all aldehydes are given in the accompanying
ESI.† In comparison to Yamamoto’s original protocol we
slightly modified his procedure4 by coordinating the aldehyde
with ATPH and separately forming the ester enolate in a second
flask. Then, the activated aldehyde A was added to the ATPH
enolate like 9b. This modified procedure avoids the risk of enoli-
sation and hence in the present cases epimerisation of the
aldehyde.
Aldol reactions with 2,3-syn aldehydes
Scheme 2 Structure of elansolid A (10), retrosynthesis of the C1–C11
fragment 11 and synthesis of dienyl ester 6.7
We initiated our studies with the 2,3-syn-aldehydes 15 and 16 in
which R1 is a phenyl substituent and R2 either represents the
MOM protecting group that in principle is able to exert chelation
control or the sterically demanding TBS group (Table 1). In all
cases, the Yamamoto aldol reaction using the enolate derived
from 6 preferentially gave the 7,8-syn,8,9-syn diastereomers 17a
and 18a, respectively. The outcome of these reactions can be
rationalised if one takes Evan’s transition state model into
account. Here, the nucleophile attacks the aldehyde via an anti-
periplanar transition state D, thereby minimizing all steric and
dipole–dipole interactions to yield the Felkin–Anh product.16
The stereochemical outcome of the aldol reaction was eluci-
dated by utilising Rychnovsky’s acetonide method.17 For that
purpose, the protecting groups were removed (MOM: HCl–
EtOH; TBS: TBAF·3H2O–THF) and the resulting diols were
transformed into the cyclic acetonides 19a and 19b (2,2-
dimethoxypropane–PPTS–CH2Cl2), respectively. Relevant and
olefination of 2-methyl-but-2-enal 13 with ylide 14.8 So far, no
studies on the substrate controlled Yamamoto aldol reaction have
been conducted. Only recently, this issue was independently
addressed for the VMAR by List9 and Denmark.10
Here, we provide a report on the substrate-controlled, diastereo-
selective Yamamoto aldol reaction which eventually leads to the
synthesis of protected C1–C11 fragment 11 of elansolid A 10.
Results and discussion
One of the first reported examples of the Yamamoto aldol reac-
tion was the selective 1,4-alkylation of α,β-unsaturated carbonyl
compounds with Grignard reagents.11 The bulky, most likely
monomeric Lewis acid 7 does not provide space for chelating
two carbonyl groups at the same time, as is commonly the case
for smaller aluminium-based Lewis acids.12 The monomeric
character of 7 compensates for the fact that the electron donating
phenol substituents cause reduced Lewis acid activity.3,11
Table 1 Yamamoto aldol reaction with 2,3-syn aldehydes 15 and 16
and antiperiplanar transition state D for the formation of all-syn products
(MOM = methoxymethyl, TBS = tert-butyldimethylsilyl)
The first asymmetric examples were disclosed by Yamamoto
et al. using chiral esters while the use of chiral aldehydes is
unexplored.13 In view of the fact that our studies are part of the
total synthesis program towards the elansolids we particularly
focussed our efforts on aldehydes of the general type B and C,
respectively. We studied the influence on the stereochemical
outcome of the vinylogous aldol reaction with respect to three
structural parameters: (a) the side chain R1, (b) the protecting
group of the hydroxyl group R2 and (c) the relative stereo-
chemistry of the α- as well as β-substituents (Fig. 1).
All α,β-syn-configured aldehydes B were prepared according
to Evans’s protocol14 while α,β-anti-configured aldehydes C
were accessed by the Masamune15 aldol reaction. In both syn-
thetic sequences we first blocked the newly formed hydroxyl
group which was followed by reductive cleavage of the chiral
Isolated
yield (%)
Diastereomeric ratio
7,8-syn/7,8-anti
Entrya
Aldehyde
Temp.
1
2
3
4
15
15
16
16
−78 °C
0 °C
88
89
49
90
5 : 1
2 : 1
2 : 1
−78 °C
0 °C
1.5 : 1
a Conditions: (i) 6 (2.0 equiv.), 7 (2.2 equiv.), toluene, −78 °C, 30 min;
(ii) aldehyde 15/16 (1.0 equiv.), 7 (2.0 equiv.), toluene, −78 °C, 30 min;
(iii) 8 (2.3 equiv.), THF, −78 °C, added to solution (i), 30 min; (iv)
solution of (ii) added to solution (iii), −78 °C or 0 °C, 16 h.
Fig. 1 General presentation of 2,3-syn B and 2,3-anti aldehydes C.
7722 | Org. Biomol. Chem., 2012, 10, 7721–7729
This journal is © The Royal Society of Chemistry 2012

View Article Online
1
diagnostic H- and 13C-NMR data for both diastereoisomers are
for R1. Aldehydes 24 and 25 preferentially afforded the 7,8-syn,
8,9-syn Yamamoto aldol products 30a and 31a when reacted
with enolate 9b (Table 2, entries 5–7). In the case of the TBS-
protection the syn-selectivity turned out to be excellent at
−78 °C (Table 2, entry 7). Steric hindrance exerted by the
t-butyl group probably hampered the facile approach of the
enolate 9b so that we also conducted the reaction at 0 °C
(Table 2, entry 7). The yield improved for the TBS-protected
aldehyde 25 but the facial selectivity dropped. It can be assumed
that particularly the bulky substituent R1 strongly favours tran-
sition state D (Scheme 1).
However, these results suggest that the anti-selective Yama-
moto aldol reaction would be difficult to achieve. Thus, use of
this protocol for the preparation of the eastern fragment 11 of
elansolid A 10 has to be regarded as a challenge.18 Nevertheless,
we first tested the TBS-substituted alkyne 32 in combination
with a MOM protection (Table 3). Under the standard conditions
at −78 °C the Yamamoto aldol reaction yielded the aldol pro-
ducts 33a,b, this time however, with a small preference for the
7,8-anti,8,9-syn diastereomer 33b (Table 3, entries 2 and 3). This
promising trend became more pronounced when the reaction
temperature was raised (Table 3, entries 4–6). However,
higher temperatures were accompanied by formation of degra-
dation products and at 50 °C selectivity was lost completely
(Table 3, entry 7).
listed in Fig. 2.
This method of assignment was also used to unravel the
stereochemical preference of all other Yamamoto aldol reactions
described below. Details are found in the ESI.†
This study was extended to aldehydes 20–25 that differ in the
substituent R1 (methyl, vinyl, t-butyl) and the protecting group
R2 (MOM, TBS). When R1 equals methyl or vinyl the stereo-
control was only low to moderate again with preference for the
7,8-syn diastereomers 26b–29b, irrespective of the protecting
group chosen (Table 2, entries 1–4). However, facial selectivity
considerably improved when the bulky t-butyl group was chosen
Fig. 2 Diagnostic 1H- and 13C-NMR data of acetonides 7,8-syn,8,9-
syn 19a and 7,8-anti,8,9-syn 19b according to Rychnovsky et al.17
Interestingly, when we exchanged the Lewis acid for coordi-
nation to the aldehyde from ATPH 7 to the sterically less
Table 2 Yamamoto aldol reaction with 2,3-syn aldehydes 20–25
(stereochemical assignment in analogy to Fig. 2; see ESI†)
Table 3 Yamamoto aldol reaction with alkynyl-substituted 2,3-syn
aldehyde 32 (stereochemical assignment in analogy to Fig. 2; see ESI†)
Isolated
yield (%)
Ratio 33a/
33b
Entrya Conditions
Diastereomeric ratio
7,8-syn/7,8-anti
Entrya
Aldehyde
Isolated yield (%)
1
2
3
4
5
6
7
8
32 not precomplexed
18
51
80
65
61
58
77
84
2 : 1
1 : 1.5
1 : 1.5
1 : 2.5
1 : 3
1 : 3
1 : 1
Standard, −78 °C, 105 min
Standard, −78 °C, 16 h
Standard, −40 °C, 75 min
Standard, 0 °C, 16 h
Standard, rt, 16 h
Standard, 50 °C, 2 h
Standard except 32 was
complexed with 34, −78 °C, 16 h
One pot, −78 °C, 16 h ref. 4
1
2
3
4
5
6
7
20
21
22
23
24
25
25b
72
80
91
90
69
62
90
26a : 26b = 3 : 1
27a : 27b = 2 : 1
28a : 28b = 1 : 1
29a : 29b = 2 : 1
30a : 30b = 6 : 1
31a : 31b > 10 : 1
31a : 31b = 6 : 1
2.5 : 1
9
93
1 : 1
a Standard conditions: (i) 6 (2.0 equiv.), 7 (2.2 equiv.), toluene, −78 °C,
30 min; (ii) aldehydes 20–25 (1.0 equiv.), 7 (2.0 equiv.), toluene,
−78 °C, 30 min; (iii) 8 (2.3 equiv.), THF, −78 °C added to solution (i),
30 min; (iv) solution (ii) added to solution (iii), −78 °C or 0 °C, 16 h.
b Reaction temperature 0 °C.
a Standard conditions: 6 (2.0 equiv.), 7 (2.2 equiv.), toluene, −78 °C,
30 min; (ii) 32 (1.0 equiv.), 7 (2.0 equiv.), toluene, −78 °C, 30 min; (iii)
8 (2.3 equiv.), THF, −78 °C, added to solution (i), 30 min; (iv) solution
(ii) added to solution (iii), −78 °C or 0 °C, 16 h.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7721–7729 | 7723

View Article Online
demanding Lewis acid methyl-aluminium-bis-2,6-diphenylphen-
oxide (MAPH, 34) the diastereoselectivity was reversed again
favouring the all-syn product 33a (Table 3, entry 8). We also
probed the one pot procedure originally reported by Yamamoto
et al.4 but did not encounter improved yields or selectivities.
Finally, precomplexation of the aldehyde and formation of inter-
mediate A (Scheme 1) is crucial because otherwise the aldol
reaction only proceeded in very low yield (Table 3, entry 1).
With these promising results in hand we next studied the
influence of different protecting groups on the β-hydroxyl group
of the aldehyde on the facial selectivity of the aldol reaction
(Table 4). Besides the p-methoxybenzyl (PMB) group which in
our hands turned out not to be suited for the Yamamoto aldol
reaction, we chose four different silyl groups. Our results reveal
that silyl protecting groups exhibit similar or even better 7,8-anti
stereoselectivities compared to the MOM group. Best results
were obtained for the TBS group at −78 °C (Table 4, entries 4, 8
and 9). In some cases, particularly when the reaction temperature
Scheme 3 Syntheses of acetonides 46a,b and 47 and diagnostic 1H-
and 13C-NMR-data. Reagents and conditions: a. TBAF·3H2O, THF,
0 °C, 25 min, 96% (69% for 46b + 27% for diastereomers 45b/separated
afterwards by flash chromatography); b. 2,2-dimethoxypropane, PPTA,
CH2Cl2, rt, 3 h, 85% for 46a and 95% for 46b; c. TBAF·3H2O, THF,
0 °C, 25 min, 96%; d. Lindlar catalyst (3 mol% Pd), H2, CH2Cl2, rt,
105 min, 75%.
Table 4 Yamamoto aldol reaction with alkynyl-substituted 2,3-syn
aldehydes 35–39 (TES = triethylsilyl, TIPS = triisopropylsilyl, TPS =
triphenylsilyl) (stereochemical assignment in analogy to Fig. 2; see
ESI†)
was kept above 0 °C, by-products were detected which most
likely are isomers that have Z-stereochemistry in the diene unit.
1
These isomers showed almost identical H NMR spectra except
that H5 and neighbouring protons showed major chemical shift
differences compared to the desired aldol products. Noteworthy,
the use of the Lewis acid MAPH 34 yielded similar yields and
selectivities with aldehyde 37 as did the standard Lewis acid
ATPH 7. The newly formed stereogenic center at C7 was also
assigned using Rychnovsky’s acetonide method.17 However,
slim substituents like the alkynyl group are problematic to
impose a defined conformation on the dioxolane ring.19 A few
standard transformations yielded acetonides 46a and 46b,
respectively (Scheme 3). While the NMR data collected for 46a
are clearly diagnostic for the expected chair conformation of
1,3-syn diols, the other diastereomer 46b gave ambiguous
results. In order to secure the stereochemistry, 46b was hydro-
genated to the corresponding alkene 47 and the NMR data were
compared with the dioxolane generated from aldol products 29
(see ESI†) unequivocally confirming the 1,3-anti relationship of
the two alkoxy-functionalities.
The preferred formation of the 7,8-anti diastereomers in the
Yamamoto aldol reactions of aldehydes that bear a silyl-substi-
tuted alkynyl group in the β-position requires one to consider a
transition state alternative to D that supposedly is responsible for
the formation of the 7,8-syn products. We propose that the alde-
hyde preferentially adopts the conformation E in which all steri-
cally demanding structural elements (a. Lewis acid ATPH
opposite to the siloxy group and b. the alkynyl-bound TBS
group opposite to the activated carbonyl group) are as far apart
from each other as possible (Scheme 4). Thus, the alkynyl
moiety directs the TBS group away from the reaction centre. In
fact, this TBS group influences the preferred conformation in the
transition state because we observed no stereoselectivity for
Isolated
yield (%) 7,8-syn : 7,8-anti
Diastereomeric ratio
Entrya Aldehyde Conditionsb
c
1
35
36
36
37
37
37
37
37
37
38
38
39
39
−78 °C, 16 h
−78 °C, 16 h 55
0 °C, 16 h
—
—
2
3
4
5
41a : 41b = 1 : 1
41a : 41b = ∼1 : 1.5d
42a : 42b = 1 : 3
42a : 42b = 1 : 2e
42a : 42b = 1 : 2e
—
42a : 42b = 1 : 3.5
42a : 42b = 1 : 3.5
43a : 43b = 1 : 2.5
43a : 43b = 1 : 2.5
44a : 44b = 1 : 2
44a : 44b = 1 : 2.5
9
−78 °C, 16 h 89
0 °C, 16 h 62
rt, 16 h 41
−78 °C, 16 h n.r.
−78 °C, 16 h 74
−78 °C, 16 h 70
−78 °C, 16 h 61
0 °C, 16 h 40
−78 °C, 16 h 78
0 °C, 2 h 50
6
7f
8g
9h
10
11
12
13
a The enantiomer with respect to the general structure was employed
instead. b Standard conditions: 6 (2.0 equiv.), 7 (2.2 equiv.), toluene,
−78 °C, 30 min; (ii) 35–39 (1.0 equiv.), 7 (2.0 equiv.), toluene, −78 °C,
30 min; (iii) 8 (2.3 equiv.), THF, −78 °C, added to solution (i), 30 min;
(iv) solution (ii) added to solution (iii), −78 °C or 0 °C, 16 h.
c ∼1 : 1 mixture and complex mixture of by-products along with
decomposition products. d About 25% Z-isomers. e About 15%
Z-isomers. f AlMe3 as the Lewis acid. g MAPH 34 as the Lewis acid.
h One pot procedure.4
7724 | Org. Biomol. Chem., 2012, 10, 7721–7729
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 5 Yamamoto aldol reaction of enolate derived from ethyl tiglate
(47) with 2,3-syn aldehydes (stereochemical assignment in analogy to
Fig. 2; see ESI†)
Scheme 4 Postulated transition state for the formation of 7,8-anti aldol
products from β-alkynyl substituted aldehydes.
Diastereomeric ratio
5,6-syn : 5,6-anti
Entrya
Aldehyde
Isolated yield (%)
1
2
3
4
20
21
32
37
72
80
70
87
51a : 51b = 2.5 : 1
52a : 52b = 2.5 : 1
53a : 53b = 4 : 1
54a : 54b = 1 : 2.5
a Standard conditions: 6 (2.0 equiv.), 7 (2.2 equiv.), toluene, −78 °C,
30 min; (ii) 20, 21, 32 or 37 (1.0 equiv.), 7 (2.0 equiv.), toluene,
−78 °C, 30 min; (iii) 8 (2.3 equiv.), THF, −78 °C, added to solution (i),
30 min; (iv) solution (ii) added to solution (iii), −78 °C or 0 °C, 16 h.
Scheme 5 Aldol reaction of enolate derived from diene 6 and 2,3-syn
aldehyde 48 (stereochemical assignment in analogy to Fig. 2; see ESI†).
Conditions: (i) 6 (2.0 equiv.), 7 (2.2 equiv.), toluene, −78 °C, 30 min;
(ii) 48 (1.0 equiv.), 7 (2.0 equiv.), toluene, −78 °C, 30 min; (iii) 8 (2.3
equiv.), THF, −78 °C, added to solution (i), 30 min; (iv) solution (ii)
added to solution (iii), −78 °C, 16 h.
of a three step procedure from ester 6 that included reduction and
Wittig olefination (see ESI†). The Yamamoto aldol reaction with
2,3-syn aldehyde 37 furnished the desired product as a mixture
of diastereomers 56 in excellent yield with the 9,10-anti product
56b being the major isomer (Scheme 6).
These examples clearly demonstrate that the preferred confor-
mation of the complexed aldehyde A (see Scheme 1) is respon-
sible for the stereochemical outcome of the substrate-controlled
Yamamoto aldol reaction.
Yamamoto aldol reactions of β-alkynyl-substituted aldehydes
without an alkynyl-bound TBS group (Scheme 5; for the prep-
aration of 48: see ESI†). In essence, conformation E supports a
different angle of approach for nucleophiles. Attack from the
backside is hindered by the silyloxy group while the front face is
less hindered for nucleophilic attack.
When a smaller coordinating Lewis acid such as MAPH 34 is
chosen and the size of the hydroxyl protection is reduced, as is
the case for the MOM group, transition state D prevails and the
Yamamoto aldol reaction becomes syn-selective (see Table 3,
entry 8).
Aldol reactions with 2,3-anti aldehydes
In a second series of experiments we investigated the use of 2,3-
anti aldehydes 57–64 in the Yamamoto aldol reaction with
enolate 9b (Table 6). Commonly, aldol products 65 to 72 were
obtained in good to very good yields. Similar to the results
obtained with 2,3-syn aldehydes 24 and 25 (Table 2, entries 5, 7
and 8) the bulky t-butyl group (R1) provided best selectivities
yielding 7,8-syn,8,9-anti aldol products 69b and 70b, respect-
ively, as major isomers (Table 6, entries 5 and 6) which may be
formed via transition state F. Here, reinforcing effects of Felkin–
Anh control and preferred 1,3-anti diol formation are operative.16
To a lesser extent the smaller methyl side chain follows the same
trend (Table 6, entries 3 and 4). When R1 equals phenyl or the
TBS-substituted alkynyl group, diastereocontrol is lost or rever-
sal of selectivity has to be encountered (Table 6, entries 1, 2, 7
and 8). Here, transition state G may play a more dominant role
in which the substituents in the β-position are rotated to new pos-
itions, because not R1 but the protected hydroxyl group become
Based on these results, we studied the Yamamoto aldol reac-
tion employing two different enolates, the first one generated
from ethyl tiglate (50). It is a simpler analogue of enolate 9b that
derives from ester 6. For comparison reasons we chose four 2,3-
syn aldehydes 20, 21, 32 and 37 (Table 5). The Yamamoto aldol
reaction furnished aldol products 51–54 with good isolated
yields and the expected stereoselectivities except with aldehyde
32 which are in the range of enolate 9b (see Table 2, entries
1 and 2, and Table 4, entries 8 and 9). Only for aldehyde 32
selectivities changed towards 53a, giving similar selectivities as
with MAPH as the Lewis acid. This might be due to the fact that
MOM is not stabilising transition state E as good as TBS and
with the sterically less demanding enolate preferences are
changing.
Likewise, we employed the extended enolate derived from
triene 55. Ester 55 was prepared in overall 64% yield by means
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7721–7729 | 7725

View Article Online
Therefore, also aldehydes 73 and 74, both lacking α-branching,
were employed in the Yamamoto aldol reaction (Scheme 7). In
both cases, the 7,9-syn diastereomers were formed as major pro-
ducts. Due to the lack of the methyl group in the α-position,
only the alkoxy group in the β-position exhibits stereochemical
control. In the proposed transition state H the interaction
between the side chain (here the small methyl group) and the
bulky ATPH that is complexed to the carbonyl group is mini-
mised. The OR group is proposed to be the sterically most
demanding group and is therefore placed in the antiperiplanar
position with respect to the carbonyl group.
Scheme 6 Aldol reaction of enolate derived from triene 52 and 2,3-syn
aldehyde 37 (stereochemical assignment in analogy to Fig. 2; see ESI†).
Conditions: (i) 55 (2.0 equiv.), 7 (2.2 equiv.), toluene, −78 °C, 30 min;
(ii) 37 (1.0 equiv.), 7 (2.0 equiv.), toluene, −78 °C, 30 min; (iii) 8 (2.3
equiv.), THF, −78 °C, added to solution (i), 30 min; (iv) solution (ii)
added to solution (iii), −78 °C, 16 h.
Table 6 Yamamoto aldol reactions with 2,3-anti aldehydes 57–64 and
proposed transition states F (leading to 7,8-syn,8,9-anti products) and G
(leading to 7,8-anti,8,9-anti products)16 (stereochemical assignment in
analogy to Fig. 2; see ESI†)
Scheme 7 Aldol reactions with β-alkoxy aldehydes 73 and 74 and pos-
tulated transition state H16 (stereochemical assignment in analogy to
Fig. 2; see ESI†). Conditions: (i) 6 (2.0 equiv.), 7 (2.2 equiv.), toluene,
−78 °C, 30 min; (ii) 73/74 (1.0 equiv.), 7 (2.0 equiv.), toluene, −78 °C,
30 min; (iii) 8 (2.3 equiv.), THF, −78 °C, added to solution (i), 30 min;
(iv) solution (ii) added to solution (iii), −78 °C, 16 h.
Diastereomeric ratio
7,9-syn : 7,9-anti
Entrya
Aldehyde
Isolated yield (%)
1
2
3
4
5
6
7
8
57
58
59
60
61
62
63
64
64
79
87
79
63
77
58
88
65a : 65b = 1 : 1
66a : 66b = 2 : 1
67a : 67b = 1 : 2.5
68a : 68b = 1 : 2
69a : 69b > 1 : 10
70a : 70b = 1 : 7
71a : 71b = 2 : 1
72a : 72b = 1 : 1
Aldol reactions with ketone 78
Finally, we chose alcohol 77 that is available en route towards
aldehyde 60 and initiated proton-induced silyl migration to form
the corresponding secondary alcohol which then was oxidised to
the ketone 78 using Dess–Martin periodinane (DMP)
(Scheme 8). Stereocontrol was found to be minute in the follow-
ing Yamamoto aldol reaction that yielded a diastereomeric
mixture of (7S)-79a and (7R)-79b with minor preference for the
Felkin–Anh product (7S : 7R = 1.5 : 1).
a Standard conditions: 6 (2.0 equiv.), 7 (2.2 equiv.), toluene, −78 °C,
30 min; (ii) 57–64 (1.0 equiv.), 7 (2.0 equiv.), toluene, −78 °C, 30 min;
(iii) 8 (2.3 equiv.), THF, −78 °C, added to solution (i), 30 min; (iv)
solution (ii) added to solution (iii), −78 °C or 0 °C, 16 h.
sterically more demanding. Now, the nucleophile 9b accesses the
carbonyl group from the other face.
For structure elucidation the TBS group was removed at
which point both diastereomers were separated chromatographi-
cally. The 7R-diol was transformed into the corresponding aceto-
nide (7R)-80 (Scheme 9). NMR data that included analysis of
nOe-experiments proved the 7R-configuration. Particularly the
Aldol reactions with β-alkoxy aldehydes
All discussion on the proposed transition states D–G did not
centre on the role of the α-methyl group in the aldehyde moiety.
7726 | Org. Biomol. Chem., 2012, 10, 7721–7729
This journal is © The Royal Society of Chemistry 2012

View Article Online
3
coupling constant J8,9a = 11.7 Hz supported the antiperiplanar
under certain conditions. Clearly, the enolate although containing
the large ATPH moiety does not influence the stereochemical
outcome of the vinylogous aldol reaction with 2,3-disubstituted
aldehydes. It is the preferred conformation of the aldehyde in the
transition state which is activated by ATPH that determines the
facial selectivity of the Yamamoto aldol reaction. The “size” of
the substituent in the β-position strongly influences aldehyde
conformation and determines the facial selectivity. Large substi-
tuents favour formation of 1,3-syn diols while slim alkynyl
groups preferentially lead to 1,3-anti products. The current study
led to the successful synthesis of the eastern fragment of elanso-
lid A 10 by utilising the substrate-controlled Yamamoto aldol
reaction.
orientation of H8 and H9a that can only be expected for the 7R-
diastereomer (conformation I). It needs to be pointed out that
conformation II is also feasible, yet H8 and H9 cannot adopt an
3
axial–axial orientation responsible for a large J8,9a coupling
constant. Furthermore, it should be unlikely for the sterically
demanding R group to be in the axial orientation although both
methyl groups have to be axial instead. In the case of the (7S)-
diastereomer both protons are either oriented equatorially or
adopt an equatorial–axial orientation with subsequent smaller
coupling constants (J).20 In fact, this is the first reported appli-
cation of the Yamamoto aldol reaction with a ketone. Still, this
preliminary result needs to be studied in more detail with respect
to the scope and control of selectivity.
Further studies will have to be directed towards the complexa-
tion of the aldehyde moiety with even larger Lewis acids than
ATPH or with bulky chiral Lewis acids.
Conclusion
We found that the substrate-controlled asymmetric Yamamoto
aldol reaction is feasible with respect to yields and selectivities
Experimental
General remarks
Unless otherwise stated, all chemicals and solvents were pur-
1
chased in per analysis quality and used as received. H NMR
spectra were recorded at 400 MHz or 500 MHz and 13C NMR
spectra were recorded at 100 MHz or 125 MHz with a BRUKER
Avance 400, a DPX 400 or a DRX 500. Chemical shift values of
NMR data are reported as values in ppm relative to (residual
undeuterated) the solvent signal as an internal standard. Multipli-
cities for 1H NMR signals are described using the following
abbreviations: s = singlet, d = doublet, t = triplet, q = quartet, m
= multiplet; where appropriate with the addition of b = broad.
13C multiplicities refer to the resonances in the off-resonance
decoupled spectra and were elucidated using the distortionless
enhancement by the polarisation transfer (DEPT) spectral editing
technique. Multiplicities for 13C NMR signals are reported using
the following abbreviations: q = quaternary (CR4), t = tertiary
Scheme 8 Synthesis of ketone 78 and Yamamoto aldol reaction with
enolate 9b derived from ester 6. Reagents and conditions: (a) silica gel;
(b) DMP, NaHCO3, CH2Cl2, rt, 1 h, 64%; (c) (i) 6 (2.0 equiv.), 7 (2.2
equiv.), toluene, −78 °C, 30 min; (ii) 78 (1.0 equiv.), 7 (2.0 equiv.),
toluene, −78 °C, 30 min; (iii) 8 (2.3 equiv.) to (i), THF, −78 °C, 30 min;
(iv) (ii) to (iii), −78 °C, 16 h.
Scheme 9 Synthesis of acetonide 80b and diagnostic NMR data for determining the configuration at C7 in (7R)-79b.
This journal is © The Royal Society of Chemistry 2012 Org. Biomol. Chem., 2012, 10, 7721–7729 | 7727

View Article Online
(R3CH), s = secondary (R2CH2) and p = primary (RCH3). Mass
spectra were obtained with a type LCT (ESI) (Micromass)
equipped with a lockspray dual ion source in combination with a
WATERS Alliance 2695 LC system, or with a type Q-TOF premier
(Micromass) spectrometer (ESI mode) in combination with a
WATERS Acquity UPLC system equipped with a WATERS BEH
C18 1.7 μm (SN 01473711315545) column (solvent A: water +
0.1% (v/v) formic acid; solvent B: MeOH + 0.1% (v/v) formic
acid; flow rate = 0.4 mL min−1; gradient (t [min]/solvent B [%]):
(0 : 5) (2.5 : 95) (6.5 : 95) (6.6 : 5) (8 : 5)). Ion mass signals (m/z)
are reported as values in atomic mass units. Optical rotations
were measured on a Perkin-Elmer polarimeter type 341 or 241 in
a quartz glass cuvette at l = 589 nm (Na D-line). The optical
rotation is given in [° ml g−1 dm−1] with c = 1 corresponding to
10 mg ml−1. Flash-chromatography was done with silica
gel (Acros, particle size 35–70 μm) by applying moderate
procedure. The crude product was purified by flash chromato-
graphy (petroleum ether : ethyl acetate = 30 : 1 → 10 : 1) and
furnished alcohol 7,8-syn,8,9-syn-31 (dr: >10 : 1, 46 mg,
0.11 μmol; 62%) as a yellow oil.
Rf = 0.23 (PE : EE = 10 : 1); [α]2D0 = −13.0 (c = 1.0, CHCl3);
1H-NMR (400 MHz, CDCl3, CHCl3 = 7.26 ppm): δ 7.33 (d,
1H, J = 15.7 Hz, H-3), 5.96 (dd, 1H, J = 7.5, 7.2 Hz, H-5), 5.82
(d, 1H, J = 15.7 Hz, H-2), 4.20 (q, 2H, J = 7.3 Hz, H-14), 3.55
(ddd, 1H, J = 6.2, 6.2, 6.2 Hz, H-7), 3.33 (d, 1H, J = 2.4 Hz,
H-9), 2.40–2.48 (m, 1H, H-6a), 2.28–2.39 (m, 1H, H-6b),
1.79–1.90 (m, 1H, H-8), 1.80 (s, 3H, H-12), 1.54 (brs, 1H, OH),
1.29 (t, 3H, J = 7.3 Hz, H-15), 0.88–0.96 (m, 12H, H-11 +
H-13), 0.86 (s, 9H, TBS), 0.08 (s, 3H, TBS), 0.05 (s, 3H, TBS)
ppm; 13C-NMR (100 MHz, CDCl3, CDCl3 = 77.16 ppm):
δ 167.6 (q, C-1), 149.2 (t, C-3), 137.9 (t, H-5), 135.0 (q, C-4),
116.4 (t, C-2), 80.2 (t, C-9), 75.4 (t, C-7), 60.4 (s, C-14), 39.8
(t, C-8), 37.2 (q, C-10), 35.3 (s, C-6), 26.7 (p, C-11), 26.5
(p, TBS), 18.9 (q, TBS), 14.5 (p, C-15), 12.6 (p, C-12), 10.2
(p, C-13), −2.8 (p, TBS), −4.0 (p, TBS) ppm; HRMS (ESI):
m/z: calculated for C23H45O4Si: 413.3087 [M + H]+, found:
413.3075 [M + H]+.
pressure. Preparative HPLC was operated at
a MERCK
HITACHI LaChrome HPLC (Pump L7150 or L7100, Interface
D-7000, Diode Array Detector L-7450) respectively at a BECK-
MANN system Gold HPLC (Solvent Module 125, Detector
166). Solvents, columns, operating procedures and retention
times are given with the corresponding experimental and analyti-
cal data.
Acknowledgements
The work was funded by the Deutsche Forschungsgemeinschaft
(Ki 397/16-1) and the Fonds der Chemischen Industrie.
General procedure for the Yamamoto-aldol reaction
Solution A: 2,6-Diphenylphenol (6.6 equiv.) was dissolved in
toluene (c = 0.28 mol l−1, with respect to AlMe3) and AlMe3
(c = 2 mol l−1 in toluene, 2.2 equiv.) was slowly added over
30 min after which time the yellow solution was stirred for
30 min at rt and then cooled to −78 °C. The ester (2.0 equiv.)
was dissolved in toluene (c = 1 mol l−1) and slowly added. The
resulting solution was stirred for 30 min at −78 °C. Solution B
(LTMP-solution 8): 2,2,6,6-Tetramethyl-piperidine (2.3 equiv.)
was dissolved in THF (c = 0.19 mol l−1) and cooled to −78 °C.
n-BuLi (c = 2.5 mol l−1 in hexane, 2.3 equiv.) was added drop-
wise. The resulting solution was stirred for 20 min at −78 °C
and slowly added to solution A. The resulting mixture was
stirred for 40 min at −78 °C. Solution C: 2,6-Diphenylphenol
(6.0 equiv.) was dissolved in toluene (c = 0.28 mol l−1 with
respect to AlMe3) and AlMe3 (c = 2 mol l−1 in toluene, 2.0
equiv.) was slowly added over 45 min. Afterwards the solution
was stirred for 30 min at rt and then cooled to −78 °C. The alde-
hyde (1.0 equiv.) was dissolved in toluene (c = 1 mol l−1) and
added dropwise. The solution was stirred for 30 min at −78 °C.
Solution C was added over 10 min to solution A and the
resulting reaction mixture was stirred at −78 °C overnight. The
reaction was terminated by addition of aq. NH4Cl, warmed up
to rt and stirred for 3 h after the addition of a solution of Na-K-
tartrate. The layers were separated and the aqueous layer
extracted with EE. The combined, organic phases were dried
with MgSO4 and the solvent was removed under reduced
pressure. The resulting crude product was purified by flash
column chromatography (petroleum ether : ethyl acetate; ratios
are given).
Notes and references
1 For introductory reviews see: (a) J. Li and D. Menche, Synthesis, 2009,
2293–2315; (b) A. Mengel and O. Reiser, Chem. Rev., 1999, 99,
1191–1224.
2 For reviews on this topic see: (a) G. Casiraghi, L. Battistini, C. Curti,
G. Rassu and F. Zanardi, Chem. Rev., 2011, 111, 3076–3154;
(b) S. V. Pansare and E. K. Paul, Chem.–Eur. J., 2011, 17, 8770–8779;
(c) T. Brodmann, M. Lorenz, R. Schäckel, S. Simsek and M. Kalesse,
Synlett, 2009, 174–192; (d) S. E. Denmark and J. R. Heemstra Jr., J. Org.
Chem., 2007, 72, 5668–5688; (e) M. Kalesse, Top. Curr. Chem., 2005,
244, 43–76; (f) S. E. Denmark and J. R. Heemstra Jr., Angew. Chem.,
2005, 117, 4760–4777, (Angew. Chem., Int. Ed., 2005, 44, 4682–4698).
3 (a) T. Mukaiyama, K. Banno and K. Narasaka, J. Am. Chem. Soc., 1974,
96, 7503–7509; (b) T. Mukaiyama, Angew. Chem., 1977, 89, 858–866,
(Angew. Chem., Int. Ed. Engl., 1977, 16, 817).
4 (a) S. Saito, M. Shiozawa and H. Yamamoto, Angew. Chem., 1999, 111,
1884–1886, (Angew. Chem., Int. Ed., 1999, 38, 1769–1771); (b) S. Saito
and H. Yamamoto, Chem. Commun., 1997, 1585–1592.
5 I. Paterson, R. D. M. Davies and R. Marquez, Angew. Chem., 2001, 113,
623–627, (Angew. Chem., Int. Ed., 2001, 40, 603–607).
6 J. A. Abramite and T. Sammakia, Org. Lett., 2007, 9, 2103–2106.
7 (a) H. Steinmetz, K. Gerth, R. Jansen, N. Schläger, R. Dehn,
S. Reinecke, A. Kirschning and R. Müller, Angew. Chem., 2011, 123,
553–557, (Angew. Chem., Int. Ed., 2011, 50, 532–536); (b) R. Dehn,
Y. Katsuyama, A. Weber, K. Gerth, R. Jansen, H. Steinmetz, G. Höfle,
R. Müller and A. Kirschning, Angew. Chem., 2011, 123, 3968–3973,
(Angew. Chem., Int. Ed., 2011, 50, 3882–3887); (c) R. Jansen, K. Gerth,
H. Steinmetz, S. Reinecke, W. Kessler, A. Kirschning and R. Müller,
Chem.–Eur. J., 2011, 17, 7739–7744.
8 (a) G. Vidari, S. Ferriño and P. A. Grieco, J. Am. Chem. Soc., 1984, 106,
3539–3548; (b) M. Ahmed, M. S. Mortensen and G. A. O’Doherty,
J. Org. Chem., 2006, 71, 7741–7746.
9 L. Ratjen, P. García-García, F. Lay, M. E. Beck and B. List, Angew.
Chem., 2011, 123, 780–784, (Angew. Chem., Int. Ed., 2011, 50, 754).
10 S. E. Denmark, T. Kobayashi and C. S. Regens, Tetrahedron, 2010, 66,
4745–4759.
Specifically for the Yamamoto aldol product 31: Aldehyde 25
(48 mg, 0.18 mmol, 1.0 equiv.) was reacted with ester 6 (56 mg,
0.36 mmol, 2.0 equiv.) at −78 °C overnight using the general
7728 | Org. Biomol. Chem., 2012, 10, 7721–7729
This journal is © The Royal Society of Chemistry 2012

View Article Online
11 K. Maruoka, H. Imoto, S. Saito and H. Yamamoto, J. Am. Chem. Soc.,
1994, 116, 4131–4132.
16 D. A. Evans, M. J. Dart, J. L. Duffy and M. G. Yang, J. Am. Chem. Soc.,
1996, 118, 4322–4343 and literature cited therein.
12 (a) D. A. Evans, B. D. Allison and M. G. Yang, Tetrahedron Lett., 1999,
40, 4457–4460; (b) D. A. Evans, B. D. Allison, M. G. Yang and
C. E. Masse, J. Am. Chem. Soc., 2001, 123, 10840–10852.
13 H. Takikawa, K. Ishihara, S. Saito and H. Yamamoto, Synlett, 2004,
732–734.
14 D. A. Evans, J. Bartroli and T. L. Shih, J. Am. Chem. Soc., 1981, 103,
2127–2129.
15 (a) T. Inoue, J. Liu, D. C. Buske and A. Abiko, J. Org. Chem., 2002, 67,
5250–5256; (b) A. Abiko, J. Liu and S. Masamune, J. Am. Chem. Soc.,
1997, 119, 2586–2587.
17 (a) S. D. Rychnovsky and D. J. Skalitzky, Tetrahedron Lett., 1990, 31,
945–948; (b) S. D. Rychnovsky, B. N. Rogers and T. I. Richardson, Acc.
Chem. Res., 1998, 31, 9–17; (c) D. A. Evans, D. L. Rieger and
J. R. Gage, Tetrahedron Lett., 1990, 31, 7099–7100.
18 (a) J. Shao and J. S. Panek, Org. Lett., 2004, 6, 3083–3085; M. G. Organ
and J. Wang, J. Org. Chem., 2002, 67, 7847–7851; (b) A.-F. Salit, C. Meyer,
J. Cossy, B. Delouvrié and L. Hennequin, Tetrahedron, 2008, 64, 6684–6697.
19 S. D. Rychnovsky, B. Rogers and G. Yang, J. Org. Chem., 1993, 58,
3511–3515.
20 M. Karplus, J. Am. Chem. Soc., 1963, 85, 2870–2871.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7721–7729 | 7729