Germanium(II)-mediated reductive cross-aldol reaction of bromoaldehydes with aldehydes: NMR studies and ab initio calculations

Article abstract of DOI:10.1021/jo800904u

(Chemical Equation Presented) A highly practical reductive cross-aldol reaction of α-bromoaldehydes with various aldehydes has been developed using Ge(II)Cl₂ to produce aldehyde germanium(IV) aldolates, which were directly transformed to variou

Full text of DOI:10.1021/jo800904u

Germanium(II)-Mediated Reductive Cross-Aldol Reaction of
Bromoaldehydes with Aldehydes: NMR Studies and ab Initio
Calculations
Shin-ya Tanaka,† Nobuo Tagashira,† Kouji Chiba,‡ Makoto Yasuda,† and Akio Baba*,†
Department of Applied Chemistry and Center for Atomic and Molecular Technologies (CAMT), Graduate
School of Engineering, Osaka UniVersity, 2-1 Yamadaoka, Suita, Osaka 565-0871, Japan, and Science and
Technology System DiVision, Ryoka Systems Inc., 1-28-38 Shinkawa, Chuo-ku, Tokyo 104-0033, Japan
baba@chem.eng.osaka-u.ac.jp
ReceiVed April 30, 2008
A highly practical reductive cross-aldol reaction of R-bromoaldehydes with various aldehydes has been
developed using Ge(II)Cl2 to produce aldehyde germanium(IV) aldolates, which were directly transformed
to various multifunctionalized compounds. A remarkable change in stereoselectivity depended on the
R-bromoaldehydes employed; secondary R-bromoaldehydes gave syn selectivities, while tertiary
R-bromoaldehydes accomplished the synthesis of anti-selective aldol products with a quaternary carbon
center. NMR studies and X-ray analysis strongly suggested the formation of germanium enolate in the
reaction of R-bromoaldehyde 2h with GeCl2-dioxane. Detailed mechanistic studies, including NMR
analysis and ab initio calculations, revealed the generation of stable germanium aldolates, which was
due to the remarkably low Lewis acidity of the germanium(IV).
SCHEME 1. Cross-Aldol Reaction of Metal Enolates with
Aldehydes
Introduction
A stereoselective aldol reaction is one of the most funda-
mental and powerful methods in organic synthesis. Owing to
their broad significance, considerable efforts have been devoted
to the development of the methodology.1 However, most of
them deal with the reactions of metal enolates derived from
ketones or esters with aldehydes (Scheme 1, eq i and ii).
The reaction using aldehyde-enolates (Scheme 1, eq iii) is
still a challenge, because the formyl group in the produced
aldolates (ꢀ-metaloxyaldehydes) suffers from several undes-
ired over-reactions (e.g., further reaction with enolates,
dehydration, and oligomerization).2
There are a few examples of the stereoselective cross-aldol
reaction using a metal enolate prepared from an aldehyde. An
anti-selective cross-aldol reaction has been achieved by titanate-
type aldehyde enolates.3 The first diastereo- and enantioselective
cross-aldol reaction between aldehydes was achieved using
trichlorosilyl enolates with a chiral Lewis base catalyst.4,5 More
* To whom correspondence should be addressed. Tel.: +81-6-6879-7384.
Fax.: +81-6-6879-7387.
† Osaka University.
(3) Yachi, K.; Shinokubo, H.; Oshima, K. J. Am. Chem. Soc. 1999, 121,
9465–9466.
‡ Ryoka Systems Inc.
(1) (a) Selected reviews: ComprehensiVe Organic Syntheses; Trost, B. M.,
Ed.; Pergamon Press: Oxford, U.K., 1991; Vol. 2, pp 99-319. (b) Mahrwald,
R. Chem. ReV. 1999, 99, 1095–1120. (c) Modern Aldol Reactions; Mahrwald,
R., Ed.; Wiley-VCH: Weinheim, Germany, 2004; Vols. 1, 2.
(2) Breit, B.; Demel, P.; Gebert, A. Chem. Commun. 2004, 11, 4–115.
(4) Denmark, S. E.; Ghosh, S. K. Angew. Chem., Int. Ed. 2001, 40, 4759–
4762.
(5) See, also; (a) Denmark, S. E.; Bui, T. Proc. Nat. Acad. Sci. U.S.A. 2004,
101, 5439–5444. (b) Denmark, S. E.; Bui, T. J. Org. Chem. 2005, 70, 10190–
10193. (c) Denmark, S. E.; Bui, T. J. Org. Chem. 2005, 70, 10393–10399.
6312 J. Org. Chem. 2008, 73, 6312–6320
10.1021/jo800904u CCC: $40.75 2008 American Chemical Society
Published on Web 07/17/2008

Germanium(II)-Mediated ReductiVe Aldol Reaction
TABLE 1. Effect of Reductants on the Reductive Cross-Aldol
Reactiona
(Table 1). Among the metals examined, only GeCl2-dioxane
effectively afforded the cross-aldol product 3aa (entry 1). In
contrast, Zn, SmI2, CrCl2, InCl, and In, generally known as
favorable reductants, gave a complex mixture that probably
involved over-reacted products (entries 2-6). The use of
SnCl2 resulted in a recovery of the starting bromoaldehyde
2a (65%), which indicates insufficient reduction ability of
SnCl2 (entry 7).
entry
reductant
conditions
yield (%)b
1
2
3
4
5
6
7c
GeCl2-dioxane (1 equiv)
Zn (1 equiv)
Sml2 (2 equiv)
CrCl2 (3 equiv)
InCl (1 equiv)
rt, 2 h
60
0
0
0
0
68 °C, 2 h
-78 °C, 2 h
rt, 14 h
rt, 3 h
68 °C, 2 h
rt, 2 h
Next, we optimized the reaction conditions for the addition
of benzaldehyde 1a and bromoaldehyde 2b. (Table 2). As the
generated aldol (ꢀ-hydroxyaldehyde) was unstable for isolation,
MeOH-quenching was performed to afford the ꢀ-hydroxyl
dimethyl acetal 4ab as an aldol equivalent. The reaction in THF
gave the product 4ab in moderate yield (entry 1), while only
the debromination of 2b was observed in DMF to give a
considerable amount (59%) of 3-phenylpropanal (entry 2). Less
basic solvents, Et2O and dioxane, hardly afforded the product,
and bromoaldehyde 2b was recovered (entries 3 and 4). The
addition of a catalytic amount of Bu4NBr, which was previously
reported, drastically improved the yield in dioxane solvent (entry
5). The use of PPh3 was found to give higher diastereose-
lectivity than Bu4NBr (entry 6). These results indicate that
it is very important to use solvents or additives with
appropriate coordination ability. The use of 1.3 equiv of
bromoaldehyde 2b and GeCl2-dioxane effectively raised the
yield (entry 7). Moreover, employing 1.5 equiv of 2b and
GeCl2-dioxane and the slow addition of 2b at 0 °C increased
the product yield to 92% (entry 8). Although the use of SnCl2
with Bu4NBr (10 mol %) promoted the consumption of
bromoaldehyde 2b, in contrast to the results obtained when
the reaction was carried out without Bu4NBr (Table 1 entry
7), only a complicated mixture was obtained (entry 9).
In (1 equiv)
SnCl2 (1 equiv)
0
0
a Reaction conditions: 1a (0.6 mmol), reductant, 2a (0.6 mmol), and
THF (2 mL). b Yield determined from 1H NMR spectrum. c Recovery,
2a (65%).
recently, the aldehyde enolates with a bulky silyl group were
found to be effective for the aldol addition of aldehyde enolates.6
Aldehyde-derived encarbamates (as an alternative to metal
enolate) achieved diastereo- and enantioselective addition to
aldehydes.7 In contrast to systems that employ a prepared
nucleophile such as metal enolate, the direct aldol reaction
between aldehydes have received increasing attention.8 The
direct systems may be ideal in terms of atom efficiency, but a
homoaldol reaction can become a problem. We recently
disclosed a reductive cross-aldol reaction using R-bromoalde-
hydes and aldehydes, in which germanium(II) was employed
as a reductant.9,10 This system was operationally simple and
required no isolation of reactive metal enolates.11 Herein, we
report a detailed and systematic investigation of the Ge(II)-
mediated system, including improvement of the reaction condi-
tions, NMR experiments, X-ray analysis, and ab initio calcu-
lations to clarify a key feature of the system. We also
demonstrate successive transformations of the produced ger-
manium aldolates to give a variety of functionalized compounds
in a one-pot treatment.
Under optimized conditions, we explored the scope of
secondary R-bromoaldehydes and aldehydes (Table 3). High
yields were obtained in the reactions with both aromatic
aldehydes bearing electron-withdrawing groups and those bear-
ing donating groups (entries 2-7). Aliphatic aldehyde 1f,
however, gave only a modest yield under the optimized
conditions (condition A: 1 mol % of PPh3, 0 °C) (entry 8). In
this case, the use of 5 mol % of PPh3 at room temperature
(condition B) increased the yield to 75% (entry 9). This
procedure achieved highly reliable results in the reaction with
aliphatic aldehydes bearing an R-hydrogen, with the exception
Results and Discussion
1. Reductive Cross-Aldol Reaction of Secondary r-Bro-
moaldehydes with Aldehydes. Our investigations started with
a search of low-valent metals for the reductive cross-aldol
reaction between benzaldehyde 1a and 2-bromoheptanal 2a
TABLE 2. Reductive Cross-Aldol of Bromoaldehyde 2b with Aldehyde 1aa
entry
reductant
X (equiv)
additive
solvent
conditions
yield (%)b
syn:anti
82:18
1
GeCl2-dioxane
GeCl2-dioxane
GeCl2-dioxane
GeCl2-dioxane
GeCl2-dioxane
GeCl2-dioxane
GeCl2-dioxane
GeCl2-dioxane
SnCl2
1.0
1.0
1.0
1.0
1.0
1.0
1.3
1.5
1.0
none
none
none
none
THF
DMF
Et2O
dioxane
dioxane
dioxane
dioxane
Et2O
rt, 2 h
rt, 2 h
rt, 2 h
rt, 2 h
rt, 1 h
rt, 1 h
rt, 1 h
0 °C, 4 h
rt, 2 h
36
0
9
2c
3d
4e
5
80:20
2
Bu4NBr (5 mol %)
PPh3 (5 mol %)
PPh3 (5 mol %)
PPh3 (1 mol %)
Bu4NBr (10 mol %)
71
71
89
92
0
87:13
91:9
89:11
91:9
6
7
8f
9g
Et2O
a Reaction conditions: 1a (0.6 mmol), GeCl2-dioxane (X equiv), 2b (X equiv), additive, and solvent (2 mL). b Yield determined from 1H NMR
spectrum. c Recovery, 3-phenylpropanal (59%). d Recovery, 3-phenylpropanal (35%) and 2b (23%). e Recovery, 3-phenylpropanal (44%) and 2b
(10%). f 2b was slowly added for 30 min. g Recovery, 2b (7%).
J. Org. Chem. Vol. 73, No. 16, 2008 6313

Tanaka et al.
TABLE 3. Reaction of Various Aldehydes and Secondary Bromoaldehydesa
a Condition A: Slow addition of 2 (0.9 mmol) in Et2O for 30 min to the mixture of 1 (0.6 mmol), GeCl2-dioxane (0.9 mmol), PPh3 (0.006 mmol),
and Et2O, 0 °C, 4 h. Condition B: 1 (0.6 mmol), GeCl2-dioxane (0.72 mmol), 2 (0.72 mmol), PPh3 (0.03 mmol), and dioxane, rt, 1 h. Condition C:
Slow addition of GeCl2-dioxane (0.78 mmol, DME solution) for 30 min to the mixture of 1h (0.6 mmol), 2b (0.78 mmol), PPh3 (0.03 mmol), and
b
DME, rt, 1 h. Yield determined from 1H NMR spectrum. Values in parentheses indicate isolated yield in different batch reactions in different scale
c
d
e
(see Supporting Information). Solvent: DME. Slow addition of 2 for 15min. PPh3 (1 mol %) was used.
of 4-pentenal (1h) (entries 8-15). The low yield from 1h was
improved by the slow addition of GeCl2-dioxane in DME
(condition C) (entry 13). Notably, secondary aldehyde 1i also
provided the corresponding cross-aldol products (entries 14 and
15).12 In the reactions of R-bromoaldehyde 2a with aldehydes
1j and 1k, which have a Csp2-halide moiety, GeCl2-dioxane
selectively reacted with 2a to give the desired aldol products
(entries 16 and 17).
of an intramolecular Reformatsky reaction,13 instead of
GeCl2-dioxane gave only a complicated mixture.
The bis-aldehyde 5 provided the cyclic aldol derivative 6 in
high yield (eq 1). As far as we know, this type of intramolecular
reductive aldol reaction (bromoaldehyde + CHO) has not been
previously reported. The use of SmI2, a well-known promoter
The highly practical, one-pot, large-scale synthesis (100
mmol) of cross-aldol product 4aa was demonstrated (eq 2).
Bromination14 of the aldehyde 1l was followed by a reductive
cross-aldol reaction with another aldehyde 1a, providing the
6314 J. Org. Chem. Vol. 73, No. 16, 2008

Germanium(II)-Mediated ReductiVe Aldol Reaction
TABLE 4. Reductive Cross-Aldol of Tertiary Bromoaldehyde 2d with Aldehyde 1ea
entry
solvent
additive
none
PPh3 (5 mol %)
none
none
none
conditions
anti:syn
yield (%)b
1
2
3
4
dioxane
dioxane
THF
THF
THF
rt, 2 h
rt, 2 h
rt, 2 h
0 °C, 2 h
0 °C, 6 h
0 °C, 6 h
80:20
83:17
88:12
92:8
94:6
95:5
44
83
73
67
83
89
5
6c
THF
none
a Reaction conditions: 1e (0.6 mmol), additive, 2d (0.6 mmol), GeCl2-dioxane (0.6 mmol), and solvent (2 mL). b Yield determined from 1H NMR
spectrum. c Run using 0.72 mmol of 2d and GeCl2-dioxane.
TABLE 5. Diastereoselective Construction of Quaternary Carbon Centera
entry
time
X (equiv)
1
R1
2
R2
product
yield (%)b
syn:anti
1
2
3
4
5
6
6
6
12
12
4
12
12
4
9
10
2
4
2
1.2
1.2
1.2
1.2
1.0
1.0
1.0
2.0
1.2
1.0
1.2
1.0
1.0
1e
1e
1m
1n
1o
1p
1g
1h
1e
1o
1e
1a
1o
4-NO2-C6H4
4-NO2-C6H4
4-CN-C6H4
4-CF3-C6H4
PhCH2CH2
(CH3)2CHCH2
BnOCH2CH2
CH2dCHCH2CH2
4-NO2-C6H4
PhCH2CH2
2d
2d
2d
2d
2d
2d
2d
2d
2e
2e
2f
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
4ed
4ed
4md
4nd
4od
4pd
4gd
4hd
4ee
4oe
4ef
4ag
4og
89
80c
83(73)
77(71)
73
66
64
85(56)
84(80)
70(77)
91
5:95
6:94
8:92
8:92
7:93
8:92
9:91
9:91
5:95
10:90
28:72
-
7
8d
9
2-Naph
2-Naph
t-BuC6H4CH2
Me
10
11e,f
12e,g
13e,h
4-NO2-C6H4
Ph
PhCH2CH2
2g
2g
83
93
Me
-
a Reaction conditions: 1 (0.6 mmol), 2 (X equiv), GeCl2-dioxane (X equiv), and THF (2 mL). b Yield determined from 1H NMR spectrum. Value in
parentheses indicates isolated yield in different batch reactions in different scale (see Supporting Information). c Isolated yield on 10 mmol scale.
d Addition of 1h to the premixed solution of 2d and GeCl2-dioxane. e Reaction temperature: rt. f Obtained as a ꢀ-hydroxyaldehyde. g Reaction
conditions: 1 (0.6 mmol), 2 (0.6 mmol), GeCl2-dioxane (0.6 mmol), PPh3 (0.03 mmol), and dioxane (2 mL). h Reaction conditions: 1 (0.6 mmol), 2
(0.6 mmol), GeCl2-dioxane (0.6 mmol), PPh3 (0.06 mmol), and dioxane (2 mL).
pure adduct 4aa in 69% isolated yield (18.4 g) after column
chromatography.
Table 5 shows the results of the stereoselective construction
of quaternary carbon center using tertiary R-bromoaldehydes.
Electron deficient aromatic aldehydes and primary aliphatic
aldehydes both provided the aldol products in high anti
selectivities (entries 1-10).16 In the reaction with bromoalde-
hyde 2f, a smaller steric difference between the two substituents
at R position (t-BuC6H4CH2 vs Me) lowered the selectivity
(entry 11). R,R-Dimethyl bromoaldehyde 2g furnished products
with no stereogenic quaternary center in high yields (entries 12
and 13).
3. Observation of Reactive Species. To gain insight into
the active species, an NMR measurement was performed on
the mixture of bromoaldehyde 2a and GeCl2-dioxane in THF-
d8 without aldehyde. Unfortunately, decomposition of 2a into
complicated mixture presumably because of homocoupling and/
or over reactions was observed. Therefore, we prepared 2h
whose enol form could be stabilized and observed for NMR
study illustrated in eq 3.17
2. Reductive Cross-Aldol Reaction of Tertiary r-Bro-
moaldehydes with Aldehydes. To broaden the scope of this
methodology, tertiary R-bromoaldehydes were treated. The
diastereocontrolled construction of a quaternary carbon center
by cross-aldol reaction was limited.15
Initially, we tested the reaction of tertiary bromoaldehyde 2d
with p-nitrobenzaldehyde 1e (Table 4). Surprisingly, anti-aldol
4ed was predominantly obtained, whereas secondary bromoal-
dehydes gave syn-selective adducts (Table 3). The addition of
a catalytic amount of PPh3 improved the yield from 44% to
83% (entries 1 and 2). The use of THF solvent avoided the
addition of PPh3 and gave both a good yield and selectivity
(entry 3). Higher yield and diastereoselectivity were obtained
by lowering the temperature with a longer reaction time (entries
4 and 5). Finally, the use of a slight excess amount of
bromoaldehyde 2d and GeCl2-dioxane provided a satisfactory
result (entry 6, 89% yield, 95:5 selectivity).
1
Figure 1 shows the H and 13C NMR spectra of germanium
1
enolate 7. Significantly, in the H NMR spectrum no peak
J. Org. Chem. Vol. 73, No. 16, 2008 6315

Tanaka et al.
provides strong evidence of the formation of germanium enolate
from R-bromoaldehyde.
4. Mechanistic Insight. A possible reaction path is depicted
in Scheme 2. GeCl2-dioxane reacts with R-bromoaldehyde 2
to generate germanium(IV) enolate 9. Basic species such as THF
or PPh3 coordinate the germanium enolate to increase the
nucleophilicity.18 Enolate 9 then reacts with aldehyde 1 to give
germanium aldolate 10, which is obtained as dimethylacetal 4
after MeOH workup. One of the most difficult features of the
cross-aldol reaction using the aldehyde enolate is that the
produced metal aldolate 10suffers from undesired over-
reactions. The use of SnCl2 with a catalytic amount of
Bu4NBr (instead of GeCl2-dioxane) gave a complicated
mixture, perhaps because of expected over-reactions of tin
aldolate (Table 2, entry 9).
A possible mechanism for the avoidance of over-reactions
might involve the formation of halohydrin 11 from aldolate 10,
as reported by Denmark and co-workers.19 In our case, however,
NMR analysis of the reaction between 1a and 2a suggested no
formation of halohydrin, and only germanium aldolate 10aa was
observed as the reaction intermediate (eq 6, see Supporting
Information).
FIGURE 1. (a) 1H NMR and (b) 13C{1H} NMR spectra of germanium
enolate 7. (c) 13C{1H} NMR and 13C off-resonance decoupled spectra
of 7.
corresponding to a formyl group around 9-10 ppm was
observed, and several singlet signals appeared around 7 ppm,
which would correspond to vinyl protons (H1) of germanium
enolate 7 (Figure 1a). Similarly, the 13C NMR spectrum showed
several signals around 150 ppm (Figure 1b). The off-resonance
decoupled spectra of the signals around 150 ppm exhibited
doublet multiplicities (Figure 1c), and therefore, the signals
corresponded to vinyl carbon (C1). These results suggest the
formation of several kinds of germanium enolates. We envision
a halogen exchange taking place on initially generated germa-
nium enolate 7a to form different kinds of germanium enolates
7b (eq 4).
After a workup of 10aa by H2O, the 13C chemical shift of
the carbonyl group in 10aa (202 ppm) moved downfield to
205 ppm (3aa), which indicated a weaker interaction between
the carbonyl group and the germanium moiety in 10aa than
that of the hydrogen bonding in 3aa. This observation
surprised us, since a strong interaction between the carbonyl
group and the germanium(IV) moiety was expected. We
focused next on a detailed investigation of the Lewis acidity
of germanium(IV).
First, we monitored the interaction between several Lewis
acids and the formyl moiety of heptanal by 13C NMR and
IR (Table 6).20 As expected, significant downfield shifts in
13C NMR and a marked decrease in the carbonyl stretching
frequency in the IR spectrum for a carbonyl group were
observed in the interaction with TiCl4 and SnCl4 (entries 2
Next, germanium enolate 7 was treated with various ligands
in order to stabilize it. When 7 was treated with 4-t-
butylpyridine, the resulting pale yellow precipitates were
soluble in several organic solvents, such as THF, CH2Cl2,
and CHCl3. The NMR spectra (in CDCl3) indicated the
formation of the pyridine adduct of germanium enolates (see
Supporting Information). X-ray analysis of the single crystals
obtained by recrystallization from THF/hexane indicated the
formation of a hexa-coordinate germanium enolate bearing two
vinyloxy moieties 8 (eq 5). An ORTEP view of 8 is shown in the
Supporting Information. Although we cannot discuss the
structural details, because of unsatisfactory refinement, this
(6) Boxer, M. B.; Yamamoto, H. J. Am. Chem. Soc. 2006, 128, 48–49.
(7) Matsubara, R.; Kawai, N.; Kobayashi, S. Angew. Chem., Int. Ed. 2006,
45, 3814–3816.
6316 J. Org. Chem. Vol. 73, No. 16, 2008

Germanium(II)-Mediated ReductiVe Aldol Reaction
SCHEME 2. A Possible Reaction Course
TABLE 6. Effect of Metal Halides on δ(13C) or ν/cm-1 of
As shown in Figure 2, the germanium center of 13in had a
tetrahedral geometry similar to that of 13out ( O1GeCl1 +
Cl1GeCl2 + Cl2GeO1: 325.4° in 13in, 331.2° in 13out), and
similar charge values on carbonyl carbons (C3) were observed
in both structures (13in, 0.49397; 13out, 0.48693). In addition,
germanium aldolate 13in had a significantly long Ge-O distance
(Ge-O2, 4.163 Å). These results strongly suggest that the
formyl group did NOT interact with the germanium center, even
in 13in. In contrast, a strong coordination of the carbonyl oxygen
to the tin center in 15in (Figure 2c) is apparent because of a
much shorter Sn-O2 distance than the Ge-O2 distance
(Sn-O2, 2.379 Å) and because of the construction of a trigonal
bypiramidal geometry of the tin center ( O1SnCl1 +
Cl1SnCl2 + Cl2SnO1 ) 353.1°). The coordination makes
the charge of the carbonyl carbon (C3) much more positive
than that of 15out, which has a noncoordinated tetrahedral
geometry in the tin center (15in, 0.54074; 15out, 0.48802).
Carbonyl Group in Heptanala
entry metal halide δ(13C) (ppm) ∆δ(13C) (ppm) ν (cm-1) ∆ν (cm-1
)
1
2
3
4
none
TiCl4
SnCl4
GeCl4
202.61
219.13
216.89
201.70
0
1728
1674
1668
1728
0
-54
-60
0
+16.52
+14.28
-0.91
a NMR, heptanal and metal halide (2.4 equiv) in CDCl3. IR, heptanal
and metal halide (1.0 equiv) in CCl4.
SCHEME 3. Theoretical Calculation of Cross-Aldol
Reaction of Trihalogenated Germanium/Tin Enolates 12/14
with Benzaldehyde 1aa
Apparently, tin aldolate 15in is more electrophilic than 15out
,
and this is probably the reason for the over-reactions. In
contrast, germanium aldolate 13 does not take a coordinated
model regardless of the direction of the carbonyl group, and
thus, the carbonyl group is not activated.
a Energy values are calculated values.
and 3). However, no such phenomena were observed in the
case of GeCl4 (entry 4). These results show that the Lewis
acidity of GeCl4 is quite low, as compared with TiCl4 and
SnCl4.
Next, ab initio calculations were performed on the cross-aldol
reactions to compare germanium enolate 12 and tin enolate 14
(Scheme 3).21,22 The produced metal aldolates 13 and 15 were
optimized in two situations, in which carbonyl oxygen was
placed either inside (13in or 15in) or outside (13out or 15out
)
relative to the metal center.
The ∆E value (eq c, -20 kcal/mol) of the formation of 15in
indicated a stronger interaction between the tin center and formyl
oxygen than that of 15out (eq d, -14 kcal/mol). On the other
hand, in the reaction of germanium enolate 12 with benzalde-
hyde 1a, the formations of 13in and 13out were equally
exothermic (eq a, ∆E -15 kcal/mol; eq b, ∆E -14 kcal/mol).
To further clarify the nature of these metal aldolates, the
optimized structures of germanium aldolates (13in, 13out) and
tin aldolates (15in, 15out) were investigated, as shown in Fig-
ure 2.
FIGURE 2. Optimized structures of germanium aldolates (a, 13in; b,
13out) and tin aldolates (c, 15in; d, 15out). Selected bond distances (Å)
and NBO charges: [13in] Ge-O1, 1.767; Ge-O2, 4.163; Ge-Cl1,
2.153; Ge-Cl2, 2.145; Ge-Br, 2.296; C3, 0.49397. [13out] Ge-O1,
1.771; Ge-Cl1, 2.155; Ge-Cl2, 2.147; Ge-Br, 2.286; C3, 0.48693.
[15in] Sn-O1, 2.380; Sn-O2, 2.379; Sn-Cl1, 2.366; Sn-Cl2, 2.367;
Sn-Br, 2.518; C3, 0.54074. [15out] Sn-O1, 1.984; Sn-Cl1, 2.353;
Sn-Cl2, 2.347; Sn-Br, 2.471; C3, 0.48802.
J. Org. Chem. Vol. 73, No. 16, 2008 6317

Tanaka et al.
SCHEME 4
13out, -1.398 eV). The energy values mean that there is
significantly low reactivity of the formyl group in germanium
aldolate 13.
These NMR studies and ab initio calculations suggest that
the formyl group of germanium(IV) aldolate 13 does not take
a coordinated model, due to the remarkably low Lewis acidity
of germanium(IV). Therefore, the formyl group of germa-
nium(IV) aldolate 13 is NOT activated by the germanium(IV)
moiety. This result is in marked contrast to that with tin(IV)
aldolates, in which the carbonyl group is highly activated by
the tin(IV) moiety. The unique character of germanium makes
possible an efficient system for the cross-aldol reaction between
aldehydes, with no undesired over-reactions. The other important
point is that the resulting germanium aldolates were more
sterically hindered than the starting aldehydes in most cases.
The steric difference avoids over reactions well because the
present system was quite sensitive to the steric environment.
In fact, the substrate 1i showed lower reactivity (Table 3, entry
14), and ButCHO did not give the product. This situation enabled
the present system to be successful without contamination of
over-reaction products.
5. Change in Stereoselectivity. As noted in Tables 3 and
5, the syn-selective cross-aldol reaction using secondary R-bro-
moaldehydes is in sharp contrast to the anti predominance of
tertiary R-bromoaldehydes (Scheme 4). To explain the interest-
ing change in selectivity, open transition models23 are strongly
(8) (a) Mahrwald, R.; Costisella, B.; Gu¨ndogan, B. Tetrahedron Lett. 1997,
38, 4543–4544. (b) Mahrwald, R.; Costisella, B.; Gu¨ndogan, B. Synthesis 1998,
262–264. (c) Northrup, A. B.; MacMillan, D. W. C. J. Am. Chem. Soc. 2002,
124, 6798–6799. (d) Co´rdova, A. Tetrahedron Lett. 2004, 45, 3949–3952. (e)
Northrup, A. B.; Mangion, I. K.; Hettche, F.; MacMillan, D. W. C. Angew. Chem.,
Int. Ed. 2004, 43, 2152–2154. (f) Mase, N.; Tanaka, F.; Barbas, C. F., III Angew.
Chem., Int. Ed. 2004, 43, 2420–2423. (g) Thayumanavan, R.; Tanaka, F.; Barbas,
C. F., III Org. Lett. 2004, 6, 3541–3544. (h) Mangion, I. K.; Northrup, A. B.;
MacMillan, D. W. C. Angew. Chem., Int. Ed. 2004, 43, 6722–6724. (i) Wang,
W.; Li, H.; Wang, J. Tetrahedron Lett. 2005, 46, 5077–5079. (j) Hayashi, Y.;
Aratake, S.; Okano, T.; Takahashi, J.; Sumiya, T.; Shoji, M. Angew. Chem., Int.
Ed. 2006, 45, 5527–5529. (k) Kano, T.; Yamaguchi, Y.; Tanaka, Y.; Maruoka,
K. Angew. Chem., Int. Ed. 2007, 46, 1738–1740. (l) Hayashi, Y.; Itoh, T.;
Aratake, S.; Ishikawa, H. Angew. Chem., Int. Ed. 2008, 47, 2082–2084.
(9) For pioneering work on low-valent germanium-mediated reductive C-
C bond formations, see:(a) Hashimoto, Y.; Kagoshima, H.; Saigo, K. Tetrahedron
Lett. 1994, 35, 4805–4808. (b) Kagoshima, H.; Hashimoto, Y.; Oguro, D.; Saigo,
K. J. Org. Chem. 1998, 63, 691–697.
(10) (a) Yasuda, M.; Tanaka, S.-y.; Baba, A. Org. Lett. 2005, 7, 1845–1848.
(b) Tanaka, S.-y.; Yasuda, M.; Baba, A. Synlett 2007, 1720–1724.
(11) The reductive cross-aldol reactions between aldehydes, not concerned
with stereoselectivity or giving moderate selectivity, have been reported. (a) Kato,
J.; Mukaiyama, T. Chem. Lett. 1983, 1727–1728. (b) Maeda, K.; Shinokubo,
H.; Oshima, K. J. Org. Chem. 1998, 63, 4558–4560.
FIGURE 3. LUMOs and next LUMOs for germanium and tin aldolates:
(a) 13in, (b) 13out, (c) 15in, (d) 15out
.
(12) Unfortunately, the reaction with tertiary aldehyde (pivalaldehyde) gave
a complicated mixture.
Figure 3 displays the MO diagrams (LUMO and the next
LUMO) and the energy values of the germanium and tin
aldolates. For tin aldolates 15, while the LUMO of 15out is
dominated by orbitals of the tin atom (Figure 3d), the coordina-
tion model 15in has the carbonyl π* orbital in the LUMO, which
is at a very low energy level (Figure 3c, -2.825 eV). On the
other hand, the LUMOs of both germanium aldolates 13in and
13out are not on the corresponding carbonyl groups (Figures 3a
and b). The carbonyl π* orbitals are involved in the next
LUMOs, which are both at quite high energy (13in, -1.151 eV;
(13) Molander, G. A.; Etter, J. B.; Harring, L. S.; Thorel, P.-J. J. Am. Chem.
Soc. 1991, 113, 8036–8045.
(14) Bellesia, F.; Ghelfi, F.; Grandi, R.; Pagnoni, U. M. J. Chem. Res. (S)
1986, 428–429.
(15) For recent examples of the cross-aldol reaction concerned with the
diastereoselective construction of the quaternary carbon center, see: (a) Burke,
E. D.; Gleason, J. L. Org. Lett. 2004, 6, 405–407. (b) Kera¨nen, M. D.; Eilbracht,
P. Org. Biomol. Chem. 2004, 2, 1688–1690. (c) Adhikari, S.; Caille, S.; Hanbauer,
M.; Ngo, V. X.; Overman, L. E. Org. Lett. 2005, 7, 2795–2797. (see also ref 8f
and 8i)
(16) Unfortunately, secondary aliphatic aldehydes gave no cross-aldol
products.
6318 J. Org. Chem. Vol. 73, No. 16, 2008

Germanium(II)-Mediated ReductiVe Aldol Reaction
SCHEME 5. Open Transition State Models
SCHEME 6. One-Pot Conversion of Germanium Aldolate
7oga
methodology makes possible effective routes for multifunction-
alized compounds without isolation and purification of aldol
products.
Germanium aldolate 19, generated from 1o and 2g, was
treated with LiAlH4 to give diol 20 in high yield (eq a). A
reductive amination26 provided the aminoalcohol 21 in 74%
yield (eq b). A Wittig reaction proceeded stereoselectively to
give only Z-isomer 22 (eq c). The one-pot methodology can be
applied to Hosomi-Sakurai allylation27 to give 1,3-anti diol
23 (eq d). Generally, the Lewis acid promoted allylation of aldol
(ꢀ-hydroxyaldehyde) requires the protection of the hydroxy
group.28 In our method, fortunately, the trihalogenated germa-
nium group acts as a temporary protecting group.
a Reagents and conditions: (a) LiAlH4, rt, 1 h; (b) aniline, Bu2SnHCl,
HMPA; (c) CH3COCHPPh3, 60 °C, 24 h; (d) allyltrimethylsilane, TiCl4,
rt, 30 min.
A similar one-pot methodology was applied to the reduction
and reductive amination of aldolate 24 to give 1,3-diol 25 and
1,3-amino alcohol 26 with a diastereocontrolled quaternary
carbon center (Scheme 7).
invoked as the transition state model, since the carbonyl
oxygen-metal interacted cyclic transition model seems to be
difficult owing to the low Lewis acidity of germanium(IV).
Conclusion
Possible open transition state models are shown in Scheme
5. When secondary R-bromoaldehydes are used, the syn
selectivity can be rationalized by the antiperiplanar model 16
(Scheme 5a). In the case of tertiary bromoaldehydes, however,
the synclinal transition model 1824 would be more favorable
than the antiperiplanar model 17 (Scheme 5b), because of the
minimization of an unfavorable steric interaction between the
aldehyde oxygen and the methyl group (the order of steric
hindrance; Csp3 > Csp2 > H). In an interesting finding, the
diastereoselectivity was dramatically changed from syn to anti
when tertiary R-bromoaldehydes were used instead of secondary
ones.
In summary, we demonstrated the Ge(II)-mediated reductive
cross-aldol reaction of R-bromoaldehydes with aldehydes, in
which a sharp change in stereochemistry was accomplished;
secondary R-bromoaldehydes gave syn aldol products, while
anti selectivities were obtained from tertiary ones. The synthetic
utility of the produced germanium aldolates was demonstrated
by one-pot transformations to other functionalized products.
NMR study revealed the formation of several kinds of trihalo-
genated germanium enolates from R-bromoaldehyde and
GeCl2-dioxane. Furthermore, treatment with tert-butylpyridine
gave the tert-butylpyridine complex of germanium enolate,
which was analyzed by X-ray crystallography. Ab initio
calculations revealed that the formyl moiety of the produced
germanium aldolate is not activated by the germanium(IV)
moiety. This unique characteristic makes possible a highly
reliable method for the cross-aldol reaction of various bromoal-
dehydes with aldehydes, with no undesired over-reactions.
6. One-Pot Transformation of Germanium Aldolates.
After achieving the effective formation of germanium aldolate,
the next objective is the successive transformation of the
resulting formyl moiety (Schemes 6 and 7).25 This one-pot
(17) Unfortunately, no cross-aldol products were obtained in the reaction of
bromoaldehyde 2h with any aldehydes.
J. Org. Chem. Vol. 73, No. 16, 2008 6319

Tanaka et al.
SCHEME 7a
a Reagents and conditions: (a) NaBH4, Na2CO3 aq; (b) aniline, Bu2SnHCl, HMPA.
Experimental Section
Culture, Sports, Science and Technology of the Japanese
Government. We also acknowledge financial support from the
Tokuyama Science Foundation. S.T. expresses his special thanks
for Research Fellowships of the Japan Society for JSPS Research
Fellowships for Young Scientists.
Representative Procedure: Synthesis of 4gb. 2-Bromo-3-
phenylpropanal 2b (0.74 mmol) was added to a stirred suspension
of 3-benzyloxypropanal 1g (0.60 mmol), PPh3 (0.028 mmol), and
GeCl2-dioxane (0.72 mmol) in dioxane (2 mL) at room temper-
ature. After the mixture was stirred for 1 h at room temperature,
methanol (8 mL) was added. The resulting solution was stirred for
an additional 1 h at room temperature, and then aqueous NaHCO3
(saturated; 10 mL) was added. The mixture was extracted with Et2O/
hexane (4/1, three times), dried (MgSO4), and evaporated to give
the crude product 4gb (66%, syn/anti ) 91:9). The crude product
was purified by silica gel chromatography (hexane/EtOAc ) 95/5
to 0/100) to afford the pure product 4gb as a pale yellow liquid
(0.31 mmol, 51%, syn/anti ) 91:9). Data for syn-4gb: IR, (neat)
Supporting Information Available: Experimental proce-
dures, spectroscopic details of new compounds, listing of
absolute energies and geometries for calculated species, and
X-ray data for 8, 4ee, and disilylated compound 27 from 23
(CIF). This material is available free of charge via the Internet
at http://pubs.acs.org.
1
3525 (OH) cm-1. H NMR: (400 MHz, CDCl3) 7.37-7.14 (m,
JO800904U
10H), 4.51 (s, 2H), 4.23 (d, J ) 4.7 Hz, 1H), 4.12-4.06 (m, 1H),
3.69-3.58 (m, 2H), 3.42 (s, 3H), 3.28 (s, 3H), 3.21-3.17 (brs,
1H), 2.79 (dd, J ) 14.3, 7.9 Hz, 1H), 2.72 (dd, J ) 14.3, 7.1 Hz,
1H), 2.09 (dddd, J ) 7.9, 7.1, 4.7, 2.0 Hz, 1H), 1.89-1.72 (m,
2H). 13C NMR: (100 MHz, CDCl3) 140.6, 138.2, 128.9, 128.3,
128.2, 127.6, 127.4, 125.8, 107.1, 73.1, 68.4, 68.0, 56.1, 54.2, 47.4,
33.9, 30.9; MS: (CI, 200 eV) 313 (M+ + 1 - MeOH, 6), 295 (M+
+ 1 - H2O - MeOH, 84), 264 (20), 263 (100), 235 (40), 191
(51), 189 (51), 173 (32), 165 (94), 161 (21), 149 (50), 148 (25), 91
(73), 75 (21). HRMS: (CI, 200 eV) calcd (C20H25O3), 313.1804
(M+ + 1 - MeOH); found, 313.1803; calcd (C20H23O2), 295.1698
(M+ + 1 - H2O - MeOH); found, 295.1704; calcd (C19H21O2),
281.1542 (M+ - 2MeOH); found, 281.1534. Anal. Calcd for
C21H28O4 (344.4446): C, 73.23; H, 8.19. Found: C, 73.19; H, 8.18.
Data for anti-4gb (selected signals are shown): 1H NMR, (400 MHz,
CDCl3) 4.33 (d, J ) 3.9 Hz, 1H), 3.86 (dddd, J ) 8.9, 4.5, 4.5, 4.5
Hz, 1H), 3.40 (s, 3H), 3.35 (s, 3H). 13C NMR: (100 MHz, CDCl3)
138.2, 129.1, 125.8, 69.4, 68.7, 55.4, 47.6, 35.2, 32.0. Some signals
are obscured by those of the major isomer.
(21) All calculations were performed with Gaussian 03. The method we
employed was the B3PW91 HF/DFT hybrid method. A basis set used was
DGDZVP for all the atoms. Total energies for evaluation corrected with zero
point energies. Solvent effects were included via a PCM model. The dielectric
constant chosen was of diethyl ether, 4.335, and atomic radii with UFF referred
as UA0.
(22) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, Jr., J. A.; Vreven, T.; Kudin, K. N.;
Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci,
B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada,
M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.;
Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian,
H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.;
Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala,
P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski,
V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul,
A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.;
Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision C.02;
Gaussian, Inc.: Wallingford, CT, 2004.
Acknowledgment. This work was supported by a Grant-in-
Aid for Scientific Research from the Ministry of Education,
(23) (a) Yamamoto, Y.; Yatagai, H.; Naruta, Y.; Maruyama, K. J. Am. Chem.
Soc. 1980, 102, 7107–7109. (b) Nakamura, E.; Yamago, S.; Machii, D.;
Kuwajima, I. Tetrahedron Lett. 1988, 29, 2207–2210. (c) Denmark, S. E.; Henke,
B. R. J. Am. Chem. Soc. 1989, 111, 8032–8034.
(18) We previously reported a remarkable effect of a bromide anion on the
activation of tin enolate. Yasuda, M.; Chiba, K.; Ohigashi, N.; Katoh, Y.; Baba,
A. J. Am. Chem. Soc. 2003, 125, 7291–7300. (b) Yasuda, M.; Hayashi, K.; Katoh,
Y.; Shibata, I.; Baba, A. J. Am. Chem. Soc. 1998, 120, 715–721. (c) Yasuda,
M.; Chiba, K.; Baba, A. J. Am. Chem. Soc. 2000, 122, 7549–7555.
(19) Denmark and co-workers observed the formation of chlorohydrin in the
aldol addition of trimethylsilyl enol ether to aldehyde in the presence of SiCl4
(see ref 5b).
(20) (a) Power, M. B.; Bott, S. G.; Atwood, J. L.; Barron, A. R. J. Am. Chem.
Soc. 1990, 112, 3446–3451. (b) Power, M. B.; Bott, S. G.; Clark, D. L.; Atwood,
J. L.; Barron, A. R. Organometallics 1990, 9, 3086–3097.
(24) The same transition state model was suggested by Nakamura et al. in
the TBAF catalyzed cross-aldol reaction of silyl enol ethers (see ref 23b).
(25) For examples of successive transformation of cross-aldol products, see:
(a) Boxer, M. B.; Yamamoto, H. J. Am. Chem. Soc. 2007, 129, 2762–2763. (b)
Boxer, M. B.; Yamamoto, H. Org. Lett. 2008, 10, 453–455. (see also ref 5b).
(26) Suwa, T.; Sugiyama, E.; Shibata, I.; Baba, A. Synthesis 2000, 789–
800.
(27) Hosomi, A.; Sakurai, H. Tetrahedron Lett. 1976, 17, 1295–1298.
(28) Evans, D. A.; Allison, B. D.; Yang, M. G.; Masse, C. E. J. Am. Chem.
Soc. 2001, 123, 10840–10852.
6320 J. Org. Chem. Vol. 73, No. 16, 2008