Diastereoselective hydroformylation of 2-substituted allylic o-DPPB-Esters-On the origin of 1,2-asymmetric induction

Article abstract of DOI:10.1002/chem.200390044

2-Substituted secondary alcohol o-DPPB esters (o-DPPB = orthodiphenylphosphanylbenzoyl) have been prepared and their o-DPPB-directed diastereoselective hydroformylation examined. It was found that the diastereoselectivity increased as a function of the steric demand of the substituents both at the stereogenic center and in the alkene 2-position. Hydrolytic cleavage of the o-DPPB group afforded - via the lactols 29 - the corresponding lactones 30, the relative configurations of the vicinal stereogenic centers of which were ascertainable by 2D-NOESY spectroscopy. In addition, a crystal structure analysis of the hydroformylation product 2d provided further confirmation of the relative configuration. Replacement of the ester carbonyl group of the o-DPPB by a methylene unit resulted in significantly worse diastereoselectivity in the course of the hydroformylation (34→35), which indicates a decisive role for the ester carbonyl function. All the experimental observations were combined in a model of the origin of the 1,2-asymmetric induction during the title reaction. The key feature is the consideration of diastereomeric trigonal-bipyramidal rhodium-hydrido-olefin complexes I and II, capable on the basis of the Hammond postulate of acting as good models for the transition states of the selectivity-determining hydrometalation step. Investigation of these complexes by force-field methods indicated good correlation between theoretically predicted and experimentally determined diastereoselectivities.

Full text of DOI:10.1002/chem.200390044

FULL PAPER
Diastereoselective Hydroformylation of 2-Substituted Allylic
o-DPPB-Esters–On the Origin of 1,2-Asymmetric Induction**
Bernhard Breit,* Golo Heckmann, and Stephan K. Zahn[a]
Abstract: 2-Substituted secondary alco-
hol o-DPPB esters (o-DPPB ˆ ortho-
diphenylphosphanylbenzoyl) have been
prepared and their o-DPPB-directed
diastereoselective hydroformylation ex-
amined. It was found that the diaster-
eoselectivity increased as a function of
the steric demand of the substituents
both at the stereogenic center and in the
alkene 2-position. Hydrolytic cleavage
of the o-DPPB group afforded–via the
lactols 29–the corresponding lactones
30, the relative configurations of the
vicinal stereogenic centers of which
were ascertainable by 2D-NOESY spec-
troscopy. In addition, a crystal structure
analysis of the hydroformylation prod-
uct 2d provided further confirmation of
the relative configuration. Replacement
of the ester carbonyl group of the o-
DPPB by a methylene unit resulted in
significantly worse diastereoselectivity
in the course of the hydroformylation
(34 ! 35), which indicates a decisive
role for the ester carbonyl function. All
the experimental observations were
combined in a model of the origin of
the 1,2-asymmetric induction during the
title reaction. The key feature is the
consideration of diastereomeric trigo-
nal-bipyramidal rhodium-hydrido-olefin
complexes I and II, capable on the basis
of the Hammond postulate of acting as
good models for the transition states of
the selectivity-determining hydrometa-
lation step. Investigation of these com-
plexes by force-field methods indicated
good correlation between theoretically
predicted and experimentally deter-
mined diastereoselectivities.
Keywords: asymmetric synthesis
¥
catalysis ¥ catalyst directing group
À
¥ C C coupling ¥ hydroformylation
methallylic and homomethallylic alcohols. As an ideal cata-
lyst-directing group we identified the ortho-diphenylphospha-
nylbenzoyl function (o-DPPB), attached to a hydroxy group
of the substrate through an ester linkage (see Scheme 1). In
Introduction
The hydroformylation of alkenes is an attractive C/C bond-
forming reaction, since it is not only an atom-economic
addition reaction but also provides the aldehyde function,
which itself is an ideal platform for further synthetic
operations.[1] However, hydroformylation has not yet been
of frequent use in organic synthesis, although it is one of the
industrially most important reactions relying on homogenous
catalysis.[2] This discrepancy is primarily due to the difficulty in
control of selectivity, in particular stereoselectivity, in the
course of the hydroformylation reaction.
PPh2
O
O
Rh-cat.
O(o-DPPB)
O(o-DPPB)
hydroformylation
+
R
R
R
(63-99%)
Me
Me
O
Me
O
1
syn-2
anti-2
We[3] and others[4] recently devised solutions to this
selectivity problem, relying on specifically introduced sub-
strate-bound catalyst-directing groups. Such a function en-
forces attractive substrate catalyst interaction and enables
efficient substrate-directed hydroformylation of, for instance,
up to
:
96
4
Scheme 1. Diastereoselective hydroformylation of metallylic o-DPPB
esters (o-DPPB ˆ ortho-diphenylphosphinobenzoate).
the case of the hydroformylation of methallylic o-DPPB
systems 1, diastereoselectivity of up to 96:4 in favor of the syn-
aldehyde 2 was observed.[5] The reaction has also been used as
a key step for the construction of stereotriad building blocks
for polyketide synthesis,[6] and additionally, was able to serve
as a key step in Domino-type reaction sequences.[7]
[a] Prof. Dr. B. Breit, G. Heckmann, Dr. S. K. Zahn
Albert-Ludwigs-Universit‰t Freiburg
Institut f¸r Organische Chemie und Biochemie
Albertstrasse 21, 79104 Freiburg (Germany)
Fax : (‡49)761-203-8715
E-mail: bernhard.breit@organik.chemie.uni-freiburg.de
[**] Substrate-Directed Diastereoselective Hydroformylations, Part 4; for
Part 3 see: B. Breit, M. Dauber, K. Harms, Chem. Eur. J. 1999, 5,
2819 2827.
However, previous studies with this system have shown that
diastereoselectivity of the o-DPPB-directed hydroformyla-
Chem. Eur. J. 2003, 9, No. 2
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0425 $ 20.00+.50/0
425

B. Breit et al.
FULL PAPER
tion is a function of the nature of the substituent at the
stereogenic center of the methallylic o-DPPB ester 1, with
secondary alkyl substituents showing the highest selectivi-
ties.[5] To learn more about the scope of this stereoselective
hydroformylation and to obtain insights into the origin of 1,2-
asymmetric induction, we have initiated studies on the
influence of the substituent R2 on the diastereoselectivity of
the o-DPPB-directed hydroformylation described (see
Scheme 2). The results of these investigations, together with
a model for the origin of 1,2-asymmetric induction in the
course of the o-DPPB-directed hydroformylation, are report-
ed here.
HCHO
Me2NH2Cl
H2O (pH 7)
RMgX, Et2O,
0 °C
OH
O
O
R1
(40-64 %)
(59-95%)
R2'
R2'
2
R'
3: R2 = iPr
4: R2 = tBu
7: R2 = iPr, R1 = Et
8: R2 = iPr, R1 = Bn
9: R2 = iPr, R1 = Ph
10: R2 = tBu, R1 = Et
11: R2 = tBu, R1 = Bn
5: R2 = iPr
6: R2 = tBu
o-DPPBA,
1.1 equiv DCC, 1 equiv DMAP
CH2Cl2, 25 °C
O(o-DPPB)
R1
7-11
(81-94%)
R2'
12: R2 = iPr, R1 = Et
13: R2 = iPr, R1 = Bn
14: R2 = iPr, R1 = Ph
15: R2 = tBu, R1 = Et
16: R2 = tBu, R1 = Bn
Scheme 3. Preparation of 2-substituted allylic o-DPPB esters.
Scheme 2. Influence of the 2-substituent R2 on the diastereoselectivity of
2-substituted allylic o-DPPB esters.
quent esterification with ortho-diphenylphosphinobenzoic
acid by the DCC/DMAP coupling protocol[8] furnished the
o-DPPB esters in good to excellent yields. Stoichiometric
amounts of DMAP as the acylation catalyst were essential to
ensure good yields for the esterification step.
Preparation of 2-alkyl-substituted allylic alcohol o-DPPB
esters: The synthesis of 2-alkyl-substituted allylic alcohol
derivatives started from aldehydes 3 and 4 (Scheme 3).
Mannich reactions provided enals 5 and 6, and chemoselective
addition of the corresponding Grignard reagent gave the
desired 2-substituted allylic alcohol derivatives 7 11. Subse-
Hydroformylation of 2-alkyl-substituted allylic alcohol o-
DPPB esters: To preclude alkene isomerization as a potential
side reaction, mild reaction conditions (608C, 20 bar CO/H2,
1:1) for the hydroformylation of the 2-alkyl-substituted allyl-
o-DPPB esters were selected.
Hydroformylation occurred smoothly at 20 bar syngas
pressure with a catalyst loading of 0.7 mol% [Rh(CO)2acac]
in the presence of 2.8 mol% of triphenylphosphite. The
addition of a co-ligand was found to be beneficial with respect
to yield and diastereoselectivity in the course of the o-DPPB-
directed hydroformylation reaction. Aldehydes 2a d and
22 26 were formed in good to excellent yields. In each case
the syn diastereomer was obtained as the major product, with
diastereoselectivities of up to 99:1 (Scheme 4, Table 1).
Abstract in German: 2-Substituierte sekund‰re Allylalkohol o-
DPPB Ester (o-DPPB ˆ ortho-diphenylphosphanylbenzoyl)
wurden pr‰pariert und deren o-DPPB-dirigierte diastereose-
lektive Hydroformylierung untersucht. Dabei wurde gefunden,
dass die Diastereoselektivit‰ten mit steigendem sterischen
Anspruch der Substituenten sowohl am stereogenen Zentrum
als auch in der 2-Position des Alkens zunahmen. Durch
hydrolytische Abspaltung der o-DPPB-Gruppe gelangte man
¸ber die g-Lactole 29 zu den entsprechenden g-Lactonen 30.
An diesen konnte die relative Konfiguration der vicinalen
stereogenen Zentren durch 2D-NOESY Spektroskopie aufge-
kl‰rt werden. Dar¸berhinaus lieferte eine Kristallstrukturana-
lyse des Hydroformylierungsproduktes 2d eine weitere Ab-
sicherung der relativen Konfiguration. Der Ersatz der Ester-
0.7 mol% [Rh(CO)2acac]
2.8 mol% P(OPh)3
20 bar H2/CO (1 : 1)
toluene, 60°C, 30-40 h
O(o-DPPB)
R2
O(o-DPPB)
O(o-DPPB)
carbonylgruppe
der
o-DPPB-Funktion
durch
eine
R1
R1
R1
+
Methyleneinheit f¸hrte zu einer deutlich schlechteren Diaste-
reoselektivit‰t im Verlauf der Hydroformylierung (34 ! 35),
was der Estercarbonylfunktion eine entscheidende Rolle zu-
weist. Alle experimentellen Beobachtungen wurden in einem
Modell zum Ursprung der 1,2-asymmetrischen Induktion im
Verlauf der Titelreaktion zusammengefasst. Kerngedanke ist
die Betrachtung der diastereomeren trigonal-bipyramidalen
Rhodium-Hydrido-Olefin-Komplexe I und II, die aufgrund
des Hammond-Postulats als gute Modelle f¸r den ‹bergangs-
zustand des Selektivit‰ts-bestimmenden Hydrometallierung-
steilschrittes dienen kˆnnen. Untersuchungen dieser Komplexe
mit Kraftfeldmethoden ergaben eine gute Korrelation zwischen
theoretisch erwarteter und experimentell gefundener Diaste-
reoselektivit‰t.
R2
O
R2
O
(71-99%)
up to 99 : 1
syn-2a-d, 22-26
1a-d, 17-21
anti-2a-d, 22-26
Scheme 4. Hydroformylation of 2-substituted allylic o-DPPB esters.
Determination of relative configuration: In the case of the
aldehydes 2a d, derived from methallylic alcohol, the
relative configurations of the hydroformylation products
had been determined previously, by transformation into the
corresponding g-lactones.[5] In addition, we were able to
obtain an X-ray crystal structure analysis of aldehyde 2d,
which further confirms the stereochemical assignment of
aldehydes 2a d. The structure of 2d in the solid state is
depicted in Figure 1.
426
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0426 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 2

Diastereoselective Hydroformylation
425 434
Table 1. Results from the substrate-directed diastereoselective hydro-
formylation of 2-substituted allylic o-DPPB esters.
To clarify the relative configuration of lactone cis-30
unambiguously by use of NOE experiments, it became
necessary to have access to the trans diastereomer of 30. A
non-directed hydroformylation of the allylic alcohol 9 was
decided upon. From previous experience, this was expected to
furnish the lactols 29 without significant diastereoselectivity.
Subjection of allylic alcohol 9 to standard hydroformylation
conditions (see Scheme 6) for 48 h gave a 50% conversion
and a theoretical yield of 50% for the g-lactols 29. Oxidation
with PCC revealed a 64:33 mixture of the cis and trans
diastereomers of lactone 30.
Entry
Allylic
o-DPPB ester
R1
R2
Product
(yield/%[a]
dr
)
(syn:anti[b]
)
1
2
3
4
5
6
7
8
9
1a
1b
1c
1d
17
18
19
20
21
Et
Me
Me
Me
Me
iPr
iPr
iPr
tBu
tBu
2a (99)
2b (75)
2c (98)
2d (97)
22 (81)
23 (96)
24 (97)
25 (71)
26 (95)
60:40[c]
72:28[c]
90:10[c]
96:4[c]
84:16
92:8
99:1
94:6
99:1
Bn
Ph
iPr
Et
Bn
Ph
Et
Bn
[a] Isolated yield after column chromatography. In all cases the branched
regioisomer could not be detected. [b] Determined by NMR spectroscopic
analysis of the crude product. [c] For hydroformylations run at 908C
diastereomeric ratios obtained for 2a d were 73:27, 80:20, 92:8, and 96:4,
respectively.
Scheme 6. Hydroformylation of allylic alcohol 9 and oxidation to give the
g-lactones cis- and trans-30.
Two-dimensional NOESY NMR spectra showed distinct
NOE contacts for both diastereomeric g-lactones 30, which
allowed the relative configurations to be determined. Thus,
cis-30 showed a NOE contact between protons attached to C4
and C5 (Figure 2), whereas trans-30 revealed a NOE contact
between the proton at C5 and the methine unit of the
isopropyl substituent.
Figure 1. Structure of 2d in the solid state.
To ensure the relative configurations of the 2-isopropyl-
and 2-tert-butyl-substituted derivatives, aldehyde 24 was
selected for transformation into the corresponding g-lactone
(see Scheme 5). The derivatization sequence commenced with
protection of aldehyde 24 as the dimethylacetal 27. Alkaline
hydrolysis with potassium hydroxide in ethanol gave alcohol
28, together with the ortho-diphenylphosphinobenzoic acid, in
quantitative yield. Deprotection of the dimethyl acetal in 28
occurred upon treatment with 80% acetic acid to furnish the
g-lactols 29, which were smoothly oxidized in the presence of
PCC to furnish the cis-g-lactone 30.
Figure 2. Observed NOE contacts for cis- and trans-30, illustrated with the
corresponding MM2 minimized geometries.
Comparison of the NMR spectra of the aldehydes 22 26
revealed characteristic differences between the syn (major)
and the anti (minor) diastereomers. For instance, the 13C NMR
resonances of the isopropyl methine carbon atoms are shifted
to higher field in syn-22 24 than in anti-22 24 (see Table 2).
The same holds for the proton NMR resonances of the
aldehyde protons, which are shifted to higher field in syn-22,
23, and 25 than in anti-22, 23, and 25. With derivatization of 24
to the cis-g-lactone 30 as a point of reference, NMR
spectroscopy allows the assignment of syn and anti config-
urations for aldehydes 22 26.
Scheme 5. Derivatization of aldehyde 24 to the g-lactone cis-30.
Chem. Eur. J. 2003, 9, No. 2
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0427 $ 20.00+.50/0
427

B. Breit et al.
FULL PAPER
Table 2. Comparison of selected NMR data of syn and anti aldehydes 22
26.
(Figure 3, Table 1, entry 8). Optimum results were obtained
for derivatives 24 (R1 ˆ Ph, R2 ˆ iPr) and 26 (R1 ˆ Bn, R2 ˆ
tBu) (Figure 3, Table 2, entries 7 and 9), the minor anti
diastereomer being undetectable in each case.
In general, an increase in the steric demand of either of the
substituents R1 and R2 significantly increased the diastereo-
selectivity of the hydroformylation of 2-substituted allylic o-
DPPB esters.
Com-
13C NMR shifts for R2
1H NMR shift
pound
of aldehyde proton
syn
anti
syn
anti
22
23
24
28.0 (CH)
19.1, 21.2
[CH(CH3)2]
28.0 (CH)
19.4, 21.3
[CH(CH3)2]
27.0 (CH)
17.1, 21.3
[CH(CH3)2]
32.8 [C(CH3)3]
28.6 [C(CH3)3]
33.0 [C(CH3)3]
28.8 [C(CH3)3]
28.8 (CH)
18.9, 20.7
[CH(CH3)2]
29.4(CH)
19.2, 20.7
9.50
9.53
9.18
9.64
9.70
9.66
[CH(CH3)2]
Model for 1,2-asymmetric induction: To understand these
trends in the diastereoselectivity and to provide a more
detailed insight into the factors governing 1,2-asymmetric
induction, one has to consider the rate and selectivity-
determining step of the hydroformylation, which is the
hydrometalation step.[9] According to the generally accepted
mechanism of the hydroformylation, the hydrometalation
occurs from a trigonal-bipyramidal rhodium(i)-hydrido-olefin
complex. Stereoelectronic reasons cause the alkene to align
coplanar with the rhodium hydride bond to allow for
efficient orbital overlap. The hydrometalation is known to
be exothermic.[10] Hence, according to Hammond×s postu-
late,[11] the transition state is early, making the starting
hydrido alkene rhodium complex a good model for the
corresponding hydrometalation transition state.
How, though, do o-DPPB esters of 2-substituted allylic
alcohols bind with a trigonal-bipyramidal rhodium center?
From previous experiments we already know that the
presence of the phosphine function is essential to produce
diastereoselectivity in the course of the hydroformylation
reaction.[5] Hence, binding of the phosphine to the catalyti-
cally active rhodium center throughout the selectivity-deter-
mining step is reasonable. The ester linkage should adopt the
energetically preferred conformation expected for an ester of
a secondary alcohol, based on minimization of allylic 1,3-
strain (see Figure 4; consider for example the o-DPPB ester
conformation in the X-ray structure plot of 1d depicted in
Figure 1). This would reduce the number of degrees of
freedom, and hence should lead to a highly ordered transition
state for the hydrometalation step.
25
26
n. d.
9.42
9.40
n.d. ˆ not detectable.
Discussion
The hydroformylation of 2-substituted allyl alcohol o-DPPB
esters proceeded in all cases with good to excellent yields at
lowtemperatures (60 8C) and within 30 48 h. In each case the
syn diastereomer was formed as the major product. For the
methallylic ester derivatives 1a d the diastereoselectivity for
the reactions run at 608C was in some cases somewhat lower
than obtained for the same reactions run at 908C (see
Scheme 4, Table 1).[5] The diastereoselectivities showa clear
dependence on the size of the substituents both at the
stereogenic center (ˆ R1) and at the 2-position of the allylic
alcohol system (ˆ R2). Thus, a combination of the sterically
least demanding substituents–R1 ˆ Et and R2 ˆ Me–gave
the lowest diastereomer ratio for 2a, at syn/anti 60:40
(Figure 3, Table 1, entry 1). However, an increase in the steric
demand of R1–as an isopropyl substituent–with R2 main-
tained as Me increased the diastereoselectivity to 96:4 for 2d
(Figure 3, Table 1, entry 4). Similarly, maintenance of R1 as Et
while increasing the size of the R2 substituent to a tert-butyl
group increased diastereoselectivity from 60:40 to 94:6 for 25
To probe this role of the o-DPPB linkage, a substrate in
which the carbonyl group of the ester was replaced by a CH2
unit was needed (!31). Such a benzyl ether linkage should
lack the conformational preference of the o-DPPB ester
linkage present in o-DPPB substrates 1 (Figure 4).
Methylene Group
flexible segment
Carbonyl Group
"fixed" segment
R
H
R
O
O
PPh2
PPh2
O
H
H
H
1
31
A
1,3 minimized
Figure 4. Proposed influence of the linkage of the catalyst-directing group
on the conformational properties of the alkene substrates.
Synthesis of a corresponding benzyl ether substrate was
achieved by starting from ortho-diphenylphosphinobenzoic
acid (Scheme 7). Reduction with lithium aluminum hydride
and oxidation with hydrogen peroxide gave the known
Figure 3. Graphical summary of the results of the o-DPPB-directed
hydroformylation of 2-substituted allylic o-DPPB esters.
428
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0428 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 2

Diastereoselective Hydroformylation
425 434
– Na+
,THF
Since rhodium(i)-hydrido-alkene complexes are too reac-
O
Ph
tive to be isolated and investigated experimentally,[13]
a
Ph
CO2H
PPh2
i, ii, iii
Br
Me
O
theoretical study of the preferred binding mode of the o-
DPPB substrates was undertaken. For this reason we em-
ployed a modified version of Casey and Whiteker×s concept of
the natural bite angle for bidentate ligands.[14] This method
was originally developed to determine natural bite angles of
diphosphine ligands. We extrapolated this method to the
bidentate phosphine/alkene rhodium complexes A of the o-
DPPB esters 1. The force-field parameters used are given in
Table 3. The potential energy as a function of the bite angle is
shown in Figure 6.[15]
Me
2 h, RT
86%
P(O)Ph2
P(O)Ph2
32
33
Ph
O
HSiCl3, NEt3
Me
toluene, 140 °C, 45 h
PPh2
77%
34
Scheme 7. Preparation of benzyl ether 34: i) LiAlH4; ii) H2O2; iii) PBr3.
phosphane oxide alcohol.[12] Treatment with phosphorus
bromide furnished bromide 32, which could be coupled to
produce the allylic ether 33 by use of the Williamson
etherification protocol. Reduction of the phosphane oxide
proceeded upon treatment with trichlorosilane in the pres-
ence of triethylamine at higher temperatures to give the
desired test substrate 34.
Table 3. Force-field parameters used to determine the natural bite angle bn
of methallylic o-DPPB esters 1.
Rh P bond
d ˆ 230 pm, k ˆ 440 nmÀ1
À
À
Rh Colefin bonds
d ˆ 200 pm (locked)
P-Rh-Colefin angle
k ˆ 0
r0 (Rh) (van der Waals)
e (hardness factor)
200 pm
0.255
Next, ortho-diphenylphosphinobenzyl allylic ether 34 was
subjected to rhodium-catalyzed hydroformylation under our
standard conditions at 908C and 20 bar syngas pressure.
Formation of the two diastereomeric aldehydes 35 was
observed, in
a syn/anti diastereomer ratio of 68:32
(Scheme 8). The corresponding hydroformylation with the
o-DPPB 1c substrate at 908C occurred with a much higher
syn/anti diastereoselectivity of 92:8. This showed that the
ester linkage was essential for achievement of high levels of
diastereoselectivity.
PPh2
PPh2
PPh2
0.7 mol% [Rh(CO)2acac]
2.8 mol% P(OPh)3
O
X
O
O
X
X
toluene, CO/H2 1:1, 20 bar
+
Ph
Ph
Ph
90 °C, 24 h
Me
O
Me
Me
O
34: X = 2H
1c: X = O
(64%)
(99%)
syn-35
anti-35
:
68
32
syn-2c
anti-2c
:
92
8
Scheme 8. Hydroformylation of benzyl ether 34 and o-DPPB ester 1c:
influence of the carbonyl group.
Figure 6. Relative energies of complex A as a function of the C'olefin-Rh-P
bite angle.
What remained to be clarified was the coordination mode
of the o-DPPB allylic esters within the trigonal-bipyramidal
geometry of the starting hydrido-olefin-rhodium(i) complexes.
Three possible bidentate binding modes of olefin and
phosphine had to be considered: a) equatorial equatorial,
b) axial equatorial, or c) axial axial coordination (Fig-
ure 5).
Clearly, the preferred bite angle bn for substrates 1 is close
to 1208. Going either to 908 (required for axial equatorial
binding) or to 1808 (required for diaxial binding) is associated
with a large energy penalty. Hence, allylic o-DPPB esters
should adopt the equatorial equatorial binding mode as a
bidentate ligand in trigonal-bipyramidal rhodium(i) com-
plexes.
L
As a result, only two diastereomeric rhodium(i) alkene
complexes leading to the corresponding hydrometalation
transition states remained to be considered: complex I,
characterized by minimization of allylic 1,2-strain, and com-
plex II, which shows a minimization of allylic 1,3-strain within
the allylic part of the molecule (Scheme 9, Figure 7). The
L
L
Rh
L
L
Rh
L
Rh
equatorial-axial
equatorial-equatorial
axial-axial
Figure 5. Possible binding modes of a bidentate ligand in a trigonal-
bipyramidal rhodium(i) complex.
Chem. Eur. J. 2003, 9, No. 2
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0429 $ 20.00+.50/0
429

B. Breit et al.
FULL PAPER
The diastereoselectivities of the o-DPPB-directed hydro-
formylation of 2-substituted allylic alcohol derivatives are
thus governed by a repulsive interaction between substituents
R1 and R2 (Scheme 9). This repulsive interaction is minimized
in alkene complexes I (and in the corresponding transition
state) because both substituents have dihedral angles of ca.
‡868. Conversely, in alkene complexes II this dihedral angle
is much smaller, at ca. À538. Hence, the two substituents R1
and R2 are in spatial proximity. The repulsive interaction
increases with increasing size of R1 and R2 and results in
higher diastereoselectivity.
O(o-DPPB)
R1
R2
1
R1
O
R1
R2
O
CO
O
R2
H
PPh2
II
PPh2
H
O
H
OC
Rh
H
Rh
I
OC
CO
O(o-DPPB)
O(o-DPPB)
R1
R1
R2
O
R2
O
Conclusion
anti-2
syn-2
Scheme 9. Assumed chelate complexes I and II for hydroformylation of o-
DPPB esters 1.
In conclusion, the hydroformylation of o-DPPB esters of
2-substituted allylic alcohols occurs as a substrate-directed
diastereoselective reaction. The diastereoselectivities can be
interpreted by employing the diastereomeric hydrido-alkene-
rhodium(i) complexes as good models for the corresponding
competing diastereomorphic hydrometalation transition
states. Two-point binding of the substrate through phosphine
and alkene, a preferred conformation induced by the ester
linkage of the o-DPPB group, and minimization of repulsive
interaction of substituents R1 and R2 are the basis and origin
of diastereoselectivity in these reactions. In some cases perfect
diastereoselectivity could be observed, which is now not only
theoretically understandable but also makes this variant of the
hydroformylation a preparatively interesting method for the
construction of building blocks with vicinal stereocenters.
relative axial position of the hydrido ligand is a consequence
of the formation only of the linear regioisomer, as observed
experimentally. Since both alkene complexes should repre-
sent good models for the corresponding diastereomorphic
transition states, inspection of the relative stabilities of
complexes I and II was carried out by use of the MMFF94
force-field, as implemented in the Spartan molecular package.
Table 4 shows the results and Figure 7 displays the calculated
structures for complexes I and II of derivative 1d.
In accordance with the experimental results, complex I,
leading to the syn diastereomer, is the energetically more stable
in all cases. Furthermore, the energy differences between the
two alkene complexes increase with the steric demand of the
substituents R1 and R2. Interestingly, the calculated diaster-
eomer ratios based on these energy differences are very close
to the experimental observations (Table 4).
Experimental Section
General: Reactions were performed
in flame-dried glassware either under
argon (purity >99.998%) or under
nitrogen. The solvents were dried by
standard procedures, distilled, and
stored under nitrogen. All tempera-
tures quoted are uncorrected. 1H,
13C NMR spectra: Bruker ARX 200,
Bruker AC 300, Bruker WH 400, or
Bruker AMX 500, with tetramethylsi-
lane (TMS), chloroform (CHCl3), or
benzene (C6H6) as internal standards.
31P NMR spectra: Bruker WH 400
(161.978 MHz) with 85% H3PO4 as
external standard. Melting points: Dr.
Tottoli (B¸chi) melting point appara-
Figure 7. MMFF-minimized geometries for rhodium complexes I and II derived from o-DPPB ester 1d.
Table 4. MMFF force-field calculations for complexes I and II.
tus. Elemental analyses: CHN-rapid
analyzer (Heraeus). Flash chromatog-
raphy: Si 60 silica gel, E. Merck AG,
Darmstadt, 40 63 mm. Hydroformylation reactions were performed in
100 and 200 mL stainless steel autoclaves equipped with magnetic
stirrers. Gases: carbon monoxide 2.0 (Messer Griesheim), hydrogen 3.0
(Messer Griesheim). The following compounds were prepared by
literature procedures: 3,3-dimethylbutyraldehyde (4),[16] methallylic o-
DPPB esters 1a d,[5] [2-(diphenylphosphinoyl)phenyl] methanol.[12]
R1
R2
EI
EII
[kcalmolÀ1
EII À EI
NI/NII
dr (exptl)
[a]
[kcalmolÀ1
]
]
[kcalmolÀ1
]
Me Me 13.7
15.2
32.2
23.5
34.1
39.3
1.5
1.7
2.8
4.2
1.9
88:12
91:9
98:2
73:27[b]
92:8
96:4
Ph
iPr
Me 30.5
Me 20.7
2-Isopropylpropenal (5): A solution of dimethylammonium chloride
(32.62 g, 0.40 mol) in 37% aqueous formaldehyde (32.5 mL, 0.40 mol)
was adjusted to pH 7 with powdered sodium bicarbonate. After addition
of 3-methylbutyraldehyde (3, 28.42 g, 0.33 mol) the reaction mixture was
heated at 708C for 22 h. The reaction mixture was subsequently steam
tBu Me 29.9
Me tBu 37.4
> 99: < 1 n.d.[c]
94:6
94:6[b]
[a] from [ln(NII/NI) ˆ À(EII À EI)/RT)] at 363 K; [b] For R1 ˆ Et; [c] not
determined.
430
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0430 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 2

Diastereoselective Hydroformylation
425 434
distilled until organic material no longer separated from the distillate. The
distillate (ca. 500 mL) was extracted with diethyl ether (3 Â 150 mL), the
combined organic phases were dried (Na2SO4), and the solvent was
removed in vacuo. Distillation of the residue at normal pressure (15 cm
Vigreux) furnished enal 5 (20.52 g, 64%) as a colorless liquid. B.p. 1098C
(1013 mbar). The analytical and spectroscopic data correspond to those
reported previously.[17]
3-tert-Butyl-1-phenylbut-3-en-2-ol (11): The procedure was analogous to
that used for the preparation of 8. Allyl alcohol 9 (721 mg, 59%) was
obtained from magnesium (201 mg, 8.3 mmol), benzyl chloride (949 mg,
7.5 mmol), and 2-tert-butyl-propenal (6, 673 mg, 6.0 mmol), as a colorless
oil. Rf ˆ 0.33 (petrol ether/tert-butyl methyl ether 4:1); 1H NMR
(300.133 MHz, CDCl3): d ˆ 1.12 [s, 9H; C(CH3)3], 1.66 (brs, 1H; OH),
2.80 (dd, J ˆ 13.8 Hz, 8.9 Hz, 1H; CH2Ph), 2.94 (dd, J ˆ 13.8 Hz, 3.9 Hz,
ˆ
1H; CH2Ph), 4.43 [dd, J ˆ 8.9, 3.9 Hz, 1H; CHOH), 5.15 (s, 1H CH2), 5.30
2-tert-Butylpropenal (6): The procedure used was analogous to that
described for the preparation of enal 5. Treatment of 3,3-dimethylbutyr-
aldehyde (4, 3.51 g, 35.5 mmol), dimethylammonium chloride (3.43 g,
42 mmol), and 37% aqueous formaldehyde solution (3.4 mL, 42 mmol)
gave enal 6 (1.53 g, 40%) as a colorless liquid. B.p. 1248C (1013 mbar). The
analytical and spectroscopic data correspond to those reported previous-
ly.[17]
(s, 1H; CH2), 7.23 7.36 (m, 5H; Ar-H); 13C NMR (75.469 MHz, CDCl3):
ˆ
d ˆ 28.98 (3C), 35.64, 45.39, 71.04, 108.19, 126.38, 128.37(2C), 129.38 (2C),
139.10, 161.11; IR (film): nÄ ˆ 3500(m, b), 2965(s), 2908(s), 1454(m),
1363(m), 1047(m), 1031(m), 908(m), 751(m), 699(s)cmÀ1; MS (EI, 70 eV):
‡
‡
‡
m/z(%): 204 (1) [M] , 113 (25) [M À C7H7] , 92 (100) [C7H8] , 57 (44)
‡
‡
[C(CH3)3] ; found: 204.1516 [M] ; C14H20O calcd 204.1514.
General procedure for synthesis of o-DPPB esters 12 16: o-DPPBA
(1 equiv), DMAP (1 equiv), and DCC (1.1 equiv) were successively added
to a solution of allylic alcohol in CH2Cl2 (0.5m, 1 equiv) and the resulting
mixture was stirred at room temperature until TLC showed complete
consumption of the starting material. The reaction mixture was subse-
quently filtered through a plug of CH2Cl2-wetted Celite and washed with
additional CH2Cl2. An appropriate amount of silica gel was added to the
filtrate, which was then concentrated to dryness. Flash chromatography
with petroleum ether/tert-butyl methyl ether (9:1) provided the o-DPPB
esters 12 16 either as slightly yellowto colorless, highly viscous oils or as
colorless solids.
2-Isopropylpent-1-en-3-ol (7): A solution of 2-isopropylpropenal (5, 2.75 g,
8 mmol) in ether (8 mL) was added dropwise over 15 min to a 1m solution
of ethyl magnesium bromide in diethyl ether (35 mL, 35 mmol). The
reaction mixture was stirred at room temperature for 3 h; followed by
addition of saturated aqueous NH4Cl solution (30 mL). The reaction
mixture was extracted with tert-butyl methyl ether (3 Â 100 mL). The
combined organic layers were dried (MgSO4), the solvent was removed in
vacuo, and the residue was purified by distillation (15 cm Vigreux) to give
allyl alcohol 7 (2.21 g, 62%) as a colorless oil. B.p. 70 718C (24 25 mbar);
1H NMR (300.133 MHz, CDCl3): d ˆ 0.88 (pseudo t, J ˆ 7.4 Hz, 3H;
CH2CH3), 1.01 [d, J ˆ 7.2 Hz, 3H; CH(CH3)2], 1.04 [d, J ˆ 7.1 Hz, 3 H ;
(1RS)-(Æ)-1-Ethyl-2-isopropyl-prop-2-enyl 2-(diphenylphosphanyl)ben-
zoate (12): o-DPPB ester 12 (1.84 g, 88%) was obtained from 2-isopro-
pylpent-1-en-3-ol (7, 0.64 g, 5.0 mmol), o-DPPBA (1.53 g, 5.0 mmol),
DMAP (0.61 g, 5.0 mmol), and DCC (1.08 g, 5.3 mmol), as a colorless
CH(CH3)2], 1.42 1.66 (m, 2H; CH2CH3), 1.71 (brs, 1H; OH), 2.23 [m, 1H;
3
ˆ
CH(CH3)2], 3.98 (pseudo t, J ˆ 6.1 Hz, 1H; CHOH), 4.87 (s, 1H, CH2),
4.98 (s, 1H; CH2); 13C NMR (75.469 MHz, CDCl3): d ˆ 9.99, 22.70, 23.15,
ˆ
28.83, 30.18, 75.75, 106.97, 158.88; analytical data correspond to those
reported previously.[18]
1
solid. M.p. 77 788C; H NMR (300.133 MHz, CDCl3): d ˆ 0.80(pseudo t,
J ˆ 7.4 Hz, 3H; CH2CH3), 0.99 [d, J ˆ 6.8 Hz, 3H; CH(CH3)2], 1.01 [d, J ˆ
6.9 Hz, 3H; CH(CH3)2], 1.64 (m, 2H; CH2CH3), 2.20 [m, 1H; CH(CH3)2],
3-Isopropyl-1-phenyl-but-3-en-2-ol (8): A solution of 2-isopropylpropenal
(5, 3.93 g, 40 mmol) in diethyl ether (5 mL) was added at 08C to a solution
of benzylmagnesium chloride [prepared from benzyl chloride (6.33 g,
50 mmol) and magnesium (1.34 g, 55 mmol)] in diethyl ether (25 mL). The
reaction mixture was stirred at room temperature for 4 h, followed by
addition of saturated aqueous NH4Cl (30 mL). The resulting precipitate
was removed by filtration over Celite. The organic phase was separated and
the aqueous phase was extracted with tert-butyl methyl ether (3 Â 40 mL).
The solvent was removed in vacuo and the residue was purified by
distillation to give allyl alcohol 8 (7.19 g, 94%) as a 10:1 mixture of 1,2- and
1,4-addition product, which was used as such in the subsequent esterifica-
tion step. B.p. 80 848C (1 mbar); 1H NMR (300.133 MHz, CDCl3): d ˆ
1.13 [d, J ˆ 6.9 Hz, 6H; CH(CH3)2], 1.15 [d, J ˆ 6.7 Hz, 6H; CH(CH3)2],
1.77 (brs, 1H; OH), 2.36 [m, 1H; CH(CH3)2], 2.75 (dd, J ˆ 13.7, 8.8 Hz, 1H;
CH2), 2.98 (dd, J ˆ 13.6, 4.2 Hz, 1H; CH2), 4.32 (m, 1H; CHOH), 4.99 (m,
ˆ
ˆ
4.89 (m, 1H; CH2), 5.00 (m, 1H; CH2), 5.30 (pseudo t, J ˆ 6.6 Hz, 1H;
CHOR), 6.89 6.92 (m, 1H; Ar-H), 7.25 7.38 (m, 12H; Ar-H), 8.06 8.13
(m, 1H; Ar-H); 13C NMR (75.469 MHz, CDCl3): d ˆ 9.89, 22.55, 22.78,
26.57, 30.41, 78.28, 109.28, 128.10, 128.29 128.45 (6Aryl-C), 130.46 (d,
J(C,P) ˆ 2.6 Hz), 131.67, 133.87 (d, J(C,P) ˆ 20.7 Hz, 2C), 133.97 (d,
J(C,P) ˆ 20.8 Hz, 2C), 134.26, 134.83 (d, J(C,P) ˆ 18.9 Hz), 138.10 (d,
J(C,P) ˆ 11.7 Hz), 138.19 (d, J(C,P) ˆ 12.2 Hz), 140.49 (d, J(C,P) ˆ
27.9 Hz), 153.86 (C2'), 165.95 (d, 3J(C,P) ˆ 2.4 Hz); 31P NMR
(161.978 MHz, CDCl3): d ˆ À4.2; elemental analysis calcd (%) for
C27H29O2P (416.5): C 77.85, H 7.03; found C 77.86, H 6.98.
(1RS)-(Æ)-1-Benzyl-2-isopropyl-prop-2-enyl 2-(diphenylphosphanyl)ben-
zoate (13): o-DPPB ester 13 (2.20 g, 92%) was obtained from 2-isoprop-
yl-4-phenylbut-1-en-3-ol (8, 1.05 g, 5.5 mmol), o-DPPBA (1.53 g,
5.0 mmol), DMAP (0.61 g, 5.0 mmol), and DCC (1.08 g, 5.3 mmol), as a
highly viscous, slightly yellowoil. 1H NMR (300.133 MHz, CDCl3): d ˆ 0.94
[d, J ˆ 6.9 Hz, 3H; CH(CH3)2], 0.97 [d, J ˆ 6.9 Hz, 3H; CH(CH3)2], 2.15
1H; CH2), 5.15 (m, 1H; CH2), 7.21 7.37 (m, 5H; Ar-H); 13C NMR
(75.469 MHz, CDCl3): d ˆ 22.35, 22.98, 30.73, 43.41, 74.79, 107.29, 125.94,
128.35 (2C), 129.34 (2C), 138.58, 158.33 (C3).
ˆ
ˆ
ˆ
[m, 1H; CH(CH3)2], 2.87 2.99 (m, 2H; CH2Ph), 4.86 (s, 1H; CH2), 4.96
ˆ
(s, 1H; CH2), 5.53 (pseudo t, J ˆ 6.7 Hz, 1H; CHOR), 6.87 6.90 (m, 1H;
2-Isopropyl-1-phenyl-prop-2-en-1-ol (9): The procedure was analogous to
that used for the preparation of 8. Allyl alcohol 9 (5.71 g, 81%) was
obtained from magnesium (1.34 g, 55 mmol), bromobenzene (7.85 g,
50 mmol), and 2-isopropylpropenal (5, 3.93 g, 40 mmol) as a colorless oil.
B.p. 85 908C (1 mbar); 1H NMR (300.133 MHz, CDCl3): d ˆ 0.95 [pseu-
do d, J ˆ 6.6 Hz, 6H; CH(CH3)2], 1.96 (brs, 1H; OH), 2.08 [m, 1H;
Ar-H), 7.14 7.30 (m, 17H; Ar-H), 7.97 8.02 (m, 1H; Ar-H); 13C NMR
(75.469 MHz, CDCl3): d ˆ 22.25, 22.59, 31.00, 40.61, 77.26, 109.78, 126.41,
128.19 (2C), 128.37 128.55 (7Aryl-C), 129.54 (2C), 130.51(d, J(C,P) ˆ
2.3 Hz), 131.79, 133.98 (d, J(C,P) ˆ 20.8 Hz, 2C), 134.07 (d, J(C,P) ˆ
20.8 Hz, 2C), 134.28, 134.60 (d, J(C,P) ˆ 18.5 Hz), 137.46 (C5), 138.16 (d,
J(C,P) ˆ 11.7 Hz), 138.19 (d, J(C,P) ˆ 12.3 Hz), 140.75 (d, J(C,P) ˆ
27.5 Hz), 153.82, 165.95 (d, J(C,P) ˆ 2.4 Hz); 31P NMR (161.978 MHz,
CDCl3): d ˆ À4.3; elemental analysis calcd (%) for C32H31O2P (478.6): C
80.31, H 6.53; found C 80.28, H 6.68.
ˆ
ˆ
CH(CH3)2], 5.01 (m, 1H; CH2), 5.15 (s, 1H; CHOH), 5.22 (m, 1H; CH2),
7.23 7.32 (m, 5H; Ar-H); 13C NMR (75.469 MHz, CDCl3): d ˆ 22.37, 23.10,
30.10, 76.61, 107.71, 126.95 (2C), 127.13, 128.34 (2C), 142.39, 157.76;
elemental analysis calcd (%) for C12H16O (176.3): C 81.77, H 9.15; found C
81.90, H 9.18.
(1RS)-(Æ)-2-Isopropyl-1-phenyl-prop-2-enyl 2-(diphenylphosphanyl)ben-
zoate (14): o-DPPB ester 14 (2.18 g, 94%) was obtained from 2-isoprop-
yl-3-phenylprop-2-en-1-ol (9, 0.88 g, 5.0 mmol), o-DPPBA (1.53 g,
5.0 mmol), DMAP (0.61 g, 5.0 mmol), and DCC (1.08 g, 5.3 mmol), as a
highly viscous, slightly yellowoil. 1H NMR (300.133 MHz, CDCl3): d ˆ 0.97
[d, J ˆ 6.8 Hz, 3H; CH(CH3)2], 1.01 [d, J ˆ 6.9 Hz, 3H; CH(CH3)2], 2.14 [m,
2-tert-Butyl-pent-1-en-3-ol (10): The procedure was analogous to that used
for the preparation of 7. Allyl alcohol 10 (0.90 g, 79%) was obtained from
ethylmagnesium bromide solution in diethyl ether (1m, 10 mL, 10 mmol)
and 2-tert-butylpropenal (6, 0.90 g, 8.0 mmol), as a colorless oil. B.p. 748C
(25 mbar); 1H NMR (300.133 MHz, CDCl3): d ˆ 0.88 (pseudo t, J ˆ 7.4 Hz,
3H; CH2CH3), 1.02 [s, 9H; C(CH3)3], 1.49 1.59 (m, 2H; CH2CH3), 1.86
(brs, 1H; OH), 4.04 (pseudo t, J ˆ 6.6 Hz, 1H; CHOH), 4.95 (s, 1H;
ˆ
ˆ
1H; CH(CH3)2], 5.04 (s, 1H; CH2), 5.14 (s, 1H; CH2), 6.44 (s, 1H;
CHOR), 6.92 6.96 (m, 1H; Ar-H), 7.21 7.35 (m, 17H; Ar-H), 8.15 8.18
(m, 1H; Ar-H); 13C NMR (75.469 MHz, CDCl3): d ˆ 22.17, 22.67, 30.46,
77.75, 109.60, 127.82 (2C), 127.90, 128.14, 128.21 128.48 (6Aryl-C), 128.31
(2C), 130.59 (d, J(C,P) ˆ 2.3 Hz), 131.88, 133.94 (d, J(C,P) ˆ 20.8 Hz, 4C),
134.31, 134.31 (d, J(C,P) ˆ 18.8 Hz), 137.99 (d, J(C,P) ˆ 12.2 Hz), 138.08 (d,
CH2), 5.06 (s, 1H; CH2); 13C NMR (75.469 MHz, CDCl3): d ˆ 10.88
(C5), 29.16 (3C), 31.50, 35.57, 71.52, 107.32, 162.10; analytical data
correspond to those reported previously.[18]
ˆ
ˆ
Chem. Eur. J. 2003, 9, No. 2
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0431 $ 20.00+.50/0
431

B. Breit et al.
FULL PAPER
J(C,P) ˆ 12.2 Hz), 138.61 (C4), 140.99 (d, J(C,P) ˆ 27.8 Hz), 153.69 (C2),
165.29; 31P NMR (161.978 MHz, CDCl3): d ˆ À4.2; elemental analysis
calcd (%) for C31H29O2P (464.6): C 80.15, H 6.29; found C 79.98, H 6.42.
2-(diphenylphosphanyl)benzoate (anti-22): Compound 22 (180 mg, 81%)
was obtained from 17 (208 mg, 0.50 mmol) after 46 h, as a colorless, highly
viscous oil. Diastereomer ratio 84:16 (syn/anti). 1H NMR (300.133 MHz,
CDCl3): d ˆ 0.74 0.86 (m, 9H; 3  CH3), 1.35 1.76 [m, 3H; CH(CH3)2 and
CH2CH3], 2.15 2.38 [m, 3H; CHCH2CHO), 5.11 5.19 (m, 1H; CHOR),
6.94 6.98 (m, 1H; Ar-H), 7.28 7.42 (m, 12H; Ar-H), 8.00 8.10 (m, 1H;
Ar-H), 9.50 (pseudo t, J ˆ 2.0 Hz, 1H; CHO-syn); [minor diastereomer:
9.64 (pseudo t, J ˆ 1.7 Hz, 1H; CHO-anti)]; 13C NMR (75.469 MHz,
CDCl3): d ˆ 10.25, 19.14, 21.23, 23.56, 28.02, 41.65, 41.84, 77.49, 128.11
140.67 (18Aryl-C), 166.23 (d, J(C,P) ˆ 2.9 Hz), 201.93 [minor diastereom-
er: 9.53, 18.88, 20.67, 25.77, 28.82, 41.29, 41.52, 76.83]; 31P NMR
(161.978 MHz, CDCl3): d ˆ À4.1; elemental analysis calcd (%) for
C28H31O3P (446.6): C 75.30, H 7.00; found C 75.15, H 7.08.
(1RS)-(Æ)-2-tert-Butyl-1-ethyl-prop-2-enyl 2-(diphenylphosphanyl)ben-
zoate (15): o-DPPB ester 15 (1.39 g, 81%) was obtained from 2-tert-
butylpent-1-en-3-ol (10, 0.22 g, 4.0 mmol), o-DPPBA (1.22 g, 4.0 mmol),
DMAP (0.61 g, 5.0 mmol), and DCC (1.08 g, 5.3 mmol), as a highly viscous,
slightly yellowoil. 1H NMR (300.133 MHz, CDCl3): d ˆ 0.76 (pseudo t, J ˆ
7.4 Hz, 3H; CH2CH3), 0.98 [s, 9H; C(CH3)3], 1.63 (pseudo quint, J ˆ 7.2 Hz,
3
ˆ
ˆ
2H; CH2CH3), 5.00 (s, 1H; CH2), 5.07 (s, 1H; CH2), 5.42 (t, J ˆ 6.6 Hz,
1H; CHOR), 6.85 6.89 (m, 1H; Ar-H), 7.19 7.37 (m, 12H; Ar-H), 8.01
8.05 (m, 1H; Ar-H); 13C NMR (75.469 MHz, CDCl3): d ˆ 10.36, 28.92, 29.14
(3C), 35.44, 74.40, 109.86, 128.05, 128.26 128.47 (6Aryl-C), 130.49 (d,
J(C,P) ˆ 2.3 Hz), 131.61, 133.80 (d, J(C,P) ˆ 21.1 Hz, 2C), 134.09 (d,
J(C,P) ˆ 21.5 Hz), 134.12, 134.79 (d, J(C,P) ˆ 18.5 Hz, 2C), 138.10 (d,
J(C,P) ˆ 10.0 Hz), 138.24 (d, J(C,P) ˆ 12.5 Hz), 140.47 (d, J(C,P) ˆ
27.4 Hz), 156.37, 165.94; 31P NMR (81.015 MHz, CDCl3): d ˆ À3.1;
elemental analysis calcd (%) for C28H31O2P (430.5): C 78.11, H 7.26; found
C 77.84, H 7.10.
(1R*,2S*)-(Æ)-1-Benzyl-2-isopropyl-4-oxobutyl 2-(diphenylphosphanyl)-
benzoate (syn-23) and (1R*,2R*)-(Æ)-1-benzyl-2-isopropyl-4-oxobutyl
2-(diphenylphosphanyl)benzoate (anti-23): Compound 23 (244 mg, 96%)
was obtained from 18 (239 mg, 0.50 mmol) after 36 h, as a yellow, highly
viscous oil. Diastereomer ratio 92:8 (syn/anti). 1H NMR (300.133 MHz,
CDCl3): d ˆ 0.86 [d, J ˆ 6.8 Hz, 3H; CH(CH3)2-syn], 0.95 [d, J ˆ 6.7 Hz,
3H; CH(CH3)2-syn], 1.79 1.90 [m, 1H; CH(CH3)2], 2.11 2.42 (m, 3H;
CHCH2CHO), 2.76 2.89 (m, 2H; PhCH2), 5.40 5.46 (m, 1H; CHOR),
6.86 6.91 (m, 1H; Ar-H), 7.13 7.44 (m, 17H; Ar-H), 7.82 7.86 (m, 1H;
Ar-H), 9.53 (brs, 1H; CHO-syn); [minor diastereomer: 0.73 (d, J ˆ 6.8 Hz,
3H; CH(CH3)2-anti), 0.77 (d, J ˆ 6.7 Hz, 3H; CH(CH3)2-anti), 9.70 (brs,
1H; CHO-anti]; 13C NMR (75.469 MHz, CDCl3): d ˆ 19.44, 21.31, 28.04,
36.73, 41.74, 41.83, 76.55, 126.47, 128.15, 128.38 (2C), 128.38 128.60
(6Aryl-C), 129.11 (2C), 130.36 (d, J(C,P) ˆ 2.2 Hz), 131.84, 133.91 (d,
J(C,P) ˆ 20.8 Hz, 2C), 134.07 (d, J(C,P) ˆ 21.0 Hz, 2C) (C1', expected at
d ꢀ 134 as a doublet, is obscured by the signals at 133.91 134.14), 134.14,
137.47, 137.59 (d, J(C,P) ˆ 12.8 Hz), 138.18 (d, J(C,P) ˆ 11.5 Hz), 140.61 (d,
J(C,P) ˆ 27.6 Hz), 165.95 (d, J(C,P) ˆ 2.6 Hz), 201.76 [minor diastereomer:
19.20, 20.14, 29.15, 38.93, 40.42, 41.39, 75.73]; 31P NMR (161.978 MHz,
CDCl3): d ˆ À4.1; elemental analysis calcd (%) for C33H33O3P (508.6): C
77.93, H 6.54; found C 77.69, H 6.34.
(1RS)-(Æ)-1-Benzyl-2-tert-butyl-prop-2-enyl 2-(diphenylphosphanyl)ben-
zoate (16): o-DPPB ester 16 (1.03 g, 84%) was obtained from 2-tert-
butyl-4-phenyl-but-1-en-3-ol (11, 511 mg, 2.5 mmol), o-DPPBA (765 mg,
2.5 mmol), DMAP (305 mg, 2.5 mmol), and DCC (540 mg, 2.6 mmol), as a
highly viscous, slightly yellowoil. 1H NMR (300.133 MHz, CDCl3): d ˆ 0.94
ˆ
[s, 9H; CH(CH3)3], 2.91 3.05 (m, 2H; CH2Ph), 5.10 (s, 1H; CH2), 5.21 (s,
ˆ
1H; CH2), 5.71 (pseudo t, J ˆ 6.6 Hz, 1H; CHOR), 6.90 6.94 (m, 1H;
Ar-H), 7.19 7.43 (m, 17H; Ar-H), 8.02 8.06 (m, 1H; Ar-H); 13C NMR
(75.469 MHz, CDCl3): d ˆ 28.96 (3C), 35.48 (C9), 42.01 (C4), 73.51 (C1),
110.87 (C3), 126.32, 128.06 (2C), 128.26 128.50 (7Aryl-C), 129.66 (2C),
130.41 (d, J(C,P) ˆ 2.2 Hz), 131.66, 133.79 (d, J(C,P) ˆ 20.8 Hz, 2C), 133.99,
134.13 (d, J(C,P) ˆ 20.9 Hz, 2C), 134.45 (d, J(C,P) ˆ 18.9 Hz), 137.46,
138.07 (d, J(C,P) ˆ 11.5 Hz), 138.19 (d, J(C,P) ˆ 12.8 Hz), 140.65 (d,
J(C,P) ˆ 27.8 Hz), 155.79, 165.59 (d, J(C,P) ˆ 2.6 Hz); 31P NMR
(81.015 MHz, CDCl3): d ˆ À3.1; elemental analysis calcd (%) for
C33H33O2P (492.6): C 80.46, H 6.75; found C 80.46, H 6.55.
(1R*,2R*)-(Æ)-2-Isopropyl-1-phenyl-4-oxobutyl 2-(diphenylphosphanyl)-
benzoate (syn-24): Compound 24 (239 mg, 97%) was obtained from 19
(232 mg, 0.50 mmol) after 33 h, as a yellow, highly viscous oil. Diastereomer
ratio 99:1 (syn/anti). 1H NMR (300.133 MHz, CDCl3): d ˆ 0.73 (d, J ˆ
7.0 Hz, 3H; CH(CH3)2), 0.88 (d, J ˆ 6.9 Hz, 3H; CH(CH3)2), 1.96 2.20
(m, 3H; CH2 and CH(CH3)2), 2.58 2.63 (m, 1H; CHCH2), 5.89 (d, J ˆ
8.4 Hz, 1H; CHOR), 6.89 6.93 (m, 1H; Ar-H), 7.19 7.35 (m, 17H; Ar-H),
8.07 8.12 (m, 1H; Ar-H), 9.18 (dd, 3J ˆ 1.9, 1.7 Hz, 1H; CHO); 13C NMR
(75.469 MHz, CDCl3): d ˆ 17.06, 21.28, 27.00, 40.44, 43.52, 77.60, 127.51
(2C), 128.11, 128.14, 128.32 (2C), 128.25 128.50 (6Aryl-C), 130.58 (d,
J(C,P) ˆ 2.3 Hz), 131.88, 133.66 (d, J(C,P) ˆ 20.5 Hz, 2C), 133.91 (d,
J(C,P) ˆ 20.8 Hz, 2C), 134.05 (d, J(C,P) ˆ 18.0 Hz), 134.22, 137.61 (d,
J(C,P) ˆ 12.9 Hz), 138.09 (d, J(C,P) ˆ 11.5 Hz), 138.24, 140.56 (d, J(C,P) ˆ
28.0 Hz), 165.58 (d, J(C,P) ˆ 2.5 Hz), 200.60; 31P NMR (161.978 MHz,
CDCl3): d ˆ À4.2; elemental analysis calcd (%) for C32H31O3P (494.6): C
77.71, H 6.32; found C 77.52, H 6.58.
General procedure for hydroformylation of methallylic o-DPPB esters
1a d and 12–16: P(OPh)3 (4.5 mg, 1.4 Â 10À2 mmol) was added at room
temperature to a solution of [Rh(CO)2acac] (0.9 mg, 3.5 Â 10À3 mmol) in
toluene (3 mL) (exclusion of air and moisture), and the mixture was stirred
for 15 min. The corresponding allylic o-DPPB ester (0.5 mmol) was
subsequently added, and the resulting solution was cannulated into a
stainless steel autoclave, followed by rinsing with additional toluene
(2 mL). The autoclave was heated to 608C and then pressurized succes-
sively with 20 bar of a 1:1 mixture of hydrogen and carbon monoxide. The
reaction was monitored by TLC and was stopped after complete
consumption of starting material. The autoclave was allowed to cool to
room temperature, and the reaction mixture was filtered through a small
plug of silica with tert-butyl methyl ether (30 mL). After evaporation of the
solvent in vacuo, the crude product was analyzed by NMR to determine
conversion and diastereomer ratio. Subsequent flash chromatography with
petroleum ether/tert-butyl methyl ether (9:1) provided the corresponding
aldehydes 2a d[5] and 22 26 as highly viscous oils.
(1R*,2S*)-(Æ)-2-tert-Butyl-1-ethyl-4-oxobutyl
2-(diphenylphosphanyl)benzoate (syn-25) and (1R*,2R*)-(Æ)-2-tert-butyl-
1-ethyl-4-oxobutyl 2-(diphenyl-phosphanyl)benzoate (anti-25): Compound
25 (163 mg, 71%) was obtained from 20 (208 mg, 0.50 mmol) after 30 h, as
a colorless, highly viscous oil. Diastereomer ratio 94:6 (syn/anti). 1H NMR
(300.133 MHz, CDCl3): d ˆ 0.74 (pseudo t, J ˆ 7.4 Hz, 3H; CH2CH3), 0.85
[s, 9H; C(CH3)3], 1.41 (m, 2H; CH2CH3), 1.96 [m, 1H; CHC(CH3)3], 2.23
(ddd, J ˆ 16.0, 5.2, 2.0 Hz, 1H; CH2CHO), 2.34 (ddd, J ˆ 16.0, 8.1, 3.3 Hz,
1H; CH2CHO), 5.18 (m, 1H; CHOR), 6.83 6.87 (m, 1H; Ar-H), 7.19 7.35
(m, 12H; Ar-H), 7.90 7.94 (m, 1H; Ar-H), 9.42 (m, 1H; CHO-syn) [minor
diastereomer: 9.66 (brs, 1H; CHO-anti)]; 13C NMR (75.469 MHz, CDCl3):
d ˆ 10.72, 23.82, 28.61 (3C), 32.80, 41.04, 46.77, 77.48, 128.15 128.57
(7Aryl-C), 130.45 (d, J(C,P) ˆ 2.0 Hz), 131.78, 133.79 (d, J(C,P) ˆ 20.5 Hz,
2C), 133.93, 134.06 (d, J(C,P) ˆ 20.9 Hz, 2C) (C1', expected at d ꢀ 134 as a
doublet, is obscured by the signals at 133.93 134.06), 137.73 (d, J(C,P) ˆ
12.7 Hz), 138.23 (d, J(C,P) ˆ 11.3 Hz), 140.69 (d, J(C,P) ˆ 27.5 Hz), 166.00
(d, J(C,P) ˆ 2.9 Hz), 202.56 [minor diastereomer: 201.93 (C4-anti)]; 31P
NMR (81.015 MHz, CDCl3): d ˆ À2.9 [minor diastereomer: À3.6]; ele-
mental analysis calcd (%) for C29H33O3P (460.6): C 75.63, H 7.22; found C
75.56, H 7.30.
Crystal structure analysis of (Æ)-2d: C27H29O3P, Mr ˆ 432.47; monoclinic,
space group: P21/n; a ˆ 12.2840(2), b ˆ 11.6675(2), c ˆ 17.0825(3) ä, b ˆ
103.6533(12)8, Vˆ 2379.14(7) ä3, 1calcd 1.207 gcmÀ3; Z ˆ 4, F(000) ˆ 920,
crystal dimensions: 0.45 Â 0.35 Â 0.2 mm. A total of 11422 reflections were
collected at 100 K with a Nonius Kappa CCD area-detector diffractometer,
by use of w scans in the theta range of 3.16 to 30.038, 6939 reflections were
unique (Rint ˆ 0.0217). The structure was solved by direct methods.[19] Full
matrix, least-squares refinement[20] was based on F2, with all non-hydrogen
atoms anisotropic and hydrogens isotropic. The refinement converged at
R1 ˆ 0.0592 and wR2 ˆ 0.1380 for all data; final GOF 1.012; largest peak
and hole in the final difference Fourier: 0.315 and À0.315 eäÀ3
.
CCDC-194956 contains the supplementary crystallographic data for this
paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/
conts/retrieving.html (or from the Cambridge Crystallographic Data
Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (‡44)1223-
336033; or deposit@ccdc.cam.uk).
(1R*,2S*)-(Æ)-1-Ethyl-2-isopropyl-4-oxobutyl 2-(diphenylphosphanyl)-
benzoate (syn-22) and (1R*,2R*)-(Æ)-1-ethyl-2-isopropyl-4-oxobutyl
432
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0432 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 2

Diastereoselective Hydroformylation
425 434
3
(1R*,2S*)-(Æ)-1-Benzyl-2-tert-butyl-4-oxobutyl 2-(diphenylphosphanyl)-
benzoate (syn-26): Compound 26 (247 mg, 95%) was obtained from 21
(246 mg, 0.50 mmol) after 40 h, as a colorless, highly viscous oil. Diaster-
eomer ratio 99:1 (syn/anti). 1H NMR (300.133 MHz, CDCl3): d ˆ 0.97 [s,
9H; C(CH3)3], 1.99 [ddd, J ˆ 8.7, 5.1, 2.7 Hz, 1H; CHC(CH3)3], 2.34 (ddd,
J ˆ 15.8, 5.1, 1.8 Hz, 1H; CH2CHO), 2.52 (ddd, J ˆ 15.8, 5.1, 3.3 Hz, 1H;
CH2CHO), 2.80 (dd, J ˆ 14.6, 10.4 Hz, 1H; PhCH2), 2.89 (dd, J ˆ 14.6,
2.8 Hz, 1H; PhCH2), 5.55 (ddd, J ˆ 10.3, 2.8, 2.6 Hz, 1H; CHOR), 6.80
6.84 (m, 1H; Ar-H), 7.09 7.33 (m, 17H; Ar-H), 7.71 7.75 (m, 1H; Ar-H),
9.40 (dd, J ˆ 3.8, 1.8 Hz, 1H; CHO); 13C NMR (75.469 MHz, CDCl3): d ˆ
28.79 (3C), 33.02, 36.88, 41.18, 46.93, 76.27, 126.40, 128.14, 128.35 (2C),
128.35 128.59 (6Aryl-C), 128.83 (2C), 130.40, 131.80, 133.53 (d, J(C,P) ˆ
17.1 Hz), 133.86, 133.89(d, J(C,P) ˆ 20.6 Hz, 2C), 134.16 (d, J(C,P) ˆ
20.8 Hz, 2C), 137.54 (d, J(C,P) ˆ 13.2 Hz), 138.36, 138.43 (d, J(C,P) ˆ
10.9 Hz), 140.89 (d, J(C,P) ˆ 27.5 Hz), 166.00 (d, J(C,P) ˆ 2.9 Hz), 202.59
(C4); 31P NMR (81.015 MHz, CDCl3): d ˆ À2.8; elemental analysis calcd
(%) for C34H35O3P (522.6): C 78.14, H 6.75; found C 77.88, H 6.66.
3H; CHCH2), 5.54 (d, J ˆ 6.4 Hz, 1H; CHPh), 7.18 7.35 (m, 5H; Ar-H);
13C NMR (75.469 MHz, CDCl3): d ˆ 19.90, 21.22, 27.49, 31.79, 47.54, 84.27,
126.51 (2C), 128.32, 128.47 (2C), 136.34, 176.83; MS (EI, 70 eV, m/z (%):
‡
204 (29), 107 (60) [C7H7O] , 98 (19), 70 (100), 55 (55), 42 (19), 28 (24)
‡
‡
[CO] ; found 204.1151 [M] (HRMS, EI); C13H16O2 calcd 204.1150.
(4R*,5R*)-(Æ)-4-Isopropyl-5-phenyl-tetrahydrofuran-2-one (cis-30) and
(4R*,5S*)-(Æ)-4-isopropyl-5-phenyl-tetrahydrofuran-2-one
(trans-30):
PPh3 (74.9 mg, 0.286 mmol) was added at room temperature (exclusion
of air and moisture) to a solution of [Rh(CO)2acac] (0.9 mg, 3.5 Â
10À3 mmol) in toluene (3 mL), and the mixture was stirred for 15 min.
Subsequently, 2-isopropyl-1-phenylprop-2-en-1-ol (9, 705 mg, 4.0 mmol)
was added, and the resulting solution was cannulated into a stainless steel
autoclave, followed by rinsing with an additional toluene (2 mL). The
autoclave was heated to 908C and then pressurized successively with 20 bar
of a 1:1 mixture of hydrogen and carbon monoxide. After 48 h under these
conditions, the autoclave was allowed to cool to room temperature, and the
reaction mixture was filtered through a small plug of silica with tert-butyl
methyl ether (30 mL). After evaporation of the solvent in vacuo, the crude
product was purified by flash chromatography with petroleum ether/tert-
butyl methyl ether (4:1) to give lactols 29 (207 mg, 50% based on recovered
starting material 350 mg, 2.0 mmol).
(1R*,2R*)-(Æ)-2-Isopropyl-4-dimethoxy-1-phenylbutyl 2-(diphenylphos-
phanyl)benzoate (27): A mixture of aldehyde 24 (1.30 g, 2.6 mmol), Bayer
Lewatit SC 102 (200 mg), and Na2SO4 (1.85 g, 13.0 mmol) in methanol
(22 mL) was stirred for 3 h at room temperature. The reaction mixture was
filtered through basic alumina. Evaporation of the solvent afforded 27
(1.41 g, >99%) as a yellow, highly viscous oil. 1H NMR (300.133 MHz,
CDCl3): d ˆ 0.83 [d, J ˆ 6.8 Hz, 3H; CH(CH3)2], 0.89 [d, J ˆ 6.8 Hz, 3H;
CH(CH3)2], 1.28 1.37 (m, 1H; CH2), 1.52 1.60 (m, 1H; CH2), 1.96 2.11
[m, 2H; CHCH(CH3)2], 2.99 (s, 3H; OCH3), 3.18 (s, 3H; OCH3), 3.50 [m,
1H; CH(OCH3)2], 5.89 (d, J ˆ 8.4 Hz, 1H; CHOR), 6.93 6.97 (m, 1H; Ar-
H), 7.27 7.34 (m, 17H; Ar-H), 8.16 8.20 (m, 1H; Ar-H); 13C NMR
(75.469 MHz, CDCl3): d ˆ 17.06, 21.20, 27.12, 29.10, 44.15, 53.13, 53.43,
78.33, 103.77, 127.28 (2C), 127.68, 128.14, 128.19 (2C), 128.28 128.49
(6Aryl-C), 130.57 (d, J(C,P) ˆ 2.3 Hz), 131.86, 133.86 (d, J(C,P) ˆ 20.9 Hz,
2C), 134.00 (d, J(C,P) ˆ 21.4 Hz, 2C), 134.33 (C1', expected at d ꢀ 134 as a
doublet, is obscured by the signals at 133.86 134.33), 137.92 (d, J(C,P) ˆ
13.1 Hz), 138.22 (d, J(C,P) ˆ 11.7 Hz), 139.89, 140.83 (d, J(C,P) ˆ 28.1 Hz),
165.61 (d, J(C,P) ˆ 2.5 Hz); 31P NMR (81.015 MHz, CDCl3): d ˆ À3.5.
(3R*,4R*)-(Æ)-4-Hydroxy-3-isopropyl-4-phenylbuten-1-al dimethylacetal
(28): A saturated solution of ethanolic potassium hydroxide (15 mL) was
added to a solution of ester 27 (1.08 g, 2.0 mmol), and the resulting mixture
was heated at reflux for 3 h. Water (30 mL) and tert-butyl methyl ether
(20 mL) were added, and the phases were separated. The aqueous phase
was extracted with tert-butyl methyl ether (2 Â 15 mL) and the combined
organic phases were dried (Na2SO4). The solvent was removed in vacuo,
and the residue was purified by flash chromatography with petroleum
ether/tert-butyl methyl ether (4:1) to furnish alcohol 28 (0.58 g, 99%) as a
yellowish oil. 1H NMR (300.133 MHz, CDCl3): d ˆ 0.83 [d, 3J ˆ 6.8 Hz, 3H;
CH(CH3)2], 0.87 [d, 3J ˆ 6.9 Hz, 3H; CH(CH3)2], 1.35 1.53 (m, 2H; CH2),
1.69 1.74 [m, 1H; CH(CH3)2], 1.94 2.00 [m, 1H; CHCH(CH3)2], 2.36
(brs, 1H; OH), 3.05 (s, 3H; OCH3), 3.15 (s, 3H; OCH3), 3.69 [pseudo t, J ˆ
6.0 Hz, 1H; CH(OCH3)2], 4.60 (dd, J ˆ 7.0, 2.4 Hz, 1H; CHOH), 7.25 7.32
(m, 5H; Ar-H); 13C NMR (75.469 MHz, CDCl3): d ˆ 17.65, 21.57, 27.15,
29.08, 46.27, 52.87, 53.39, 75.99, 104.34, 126.75 (2C), 127.37, 128.24 (2C),
143.93.
The lactols 29 (103 mg, 0.5 mmol) were oxidized as described above for the
preparation of cis-30 to give a mixture of cis- and trans-lactone 30 (88 mg,
86%) in a diastereomer ratio cis/trans 64:36.
1
NMR data for trans-30: H NMR (300.133 MHz, CDCl3): d ˆ 0.82 (d, J ˆ
6.8 Hz, 3H; CH(CH3)2), 0.93 (d, J ˆ 6.8 Hz, 3H; CH(CH3)2), 1.72 1.83 (m,
1H; CH(CH3)), 2.31 2.64 (m, 3H; CHCH2), 5.14 (d, J ˆ 6.6 Hz, 1H;
CHPh), 7.18 7.33 (m, 5H; Ar-H); 13C NMR (75.469 MHz, CDCl3): d ˆ
18.85, 20.63, 29.61, 31.66, 49.97, 84.62, 126.05, 128.45, 128.63 (2C), 139.31,
176.25. The analytical data correspond to those reported for cis-30.
1-Bromomethyl-2-(diphenylphosphinoyl)benzene (32): Phosphorus tribro-
mide (0.7 mL, 7.4 mmol) was added dropwise at 08C to a suspension of [2-
(diphenylphosphinoyl)-phenyl] methanol[12] (2.87 g, 9.3 mmol) in THF
(50 mL) and CH2Cl2 (10 mL). After the mixture had been stirred for
15 min at this temperature and for a further 2 h at room temperature, TLC
showed complete consumption of starting material. Water (5 mL) was
added, and the reaction mixture was poured into water (100 mL). After
extraction with CH2Cl2 (3 Â 50 mL), the combined organic phases were
dried (MgSO4) and the solvent was evaporated. Purification of the residue
by flash chromatography with petroleum ether/ethyl acetate (1:1) furnished
bromide 32 (3.3 g, 96%) as a colorless waxy solid. 1H NMR (500.130 MHz,
CDCl3): d ˆ 4.93 (s, 2H; CH2), 6.97 (m, 1H; Ar-H), 7.17 (m, 1H; Ar-H),
7.39 7.51 (m, 8H; Ar-H), 7.56 7.60 (m, 4H; Ar-H); 13C NMR
(75.469 MHz, CDCl3): d ˆ 60.2, 127.4 (d, ˆ 12.4 Hz), 128.6 (d, J(C,P) ˆ
12.0 Hz, 4C), 130.7 (d, J(C,P) ˆ 72.3 Hz, 2C), 132.0 (d, J(C,P) ˆ 10.2 Hz,
4C), 132.1 (d, J(C,P) ˆ 2.9 Hz), 132.5 (d, J(C,P) ˆ 2.8 Hz), 133.0 (d,
J(C,P) ˆ 6.3 Hz), 133.2 (d, J(C,P) ˆ 11.9 Hz), 133.5 (s), 142.9 (d, J(C,P) ˆ
6.8 Hz, C1); 31P NMR (121.495 MHz, CDCl3): d ˆ 32.3 (s); analytical data
correspond to those reported previously.[12]
(Æ)-1-{[(2-Methyl-1-phenylallyl)oxy]methyl}-2-(diphenylphosphinoyl)-
benzene (33): Sodium hydride (84 mg, 2.1 mmol, 60% in oil, 1.1 equiv) was
washed oil-free with petroleum ether (2 mL) and suspended in THF
(4 mL). Subsequently, a solution of (Æ)-2-methyl-1-phenylprop-2-en-1-
ol[21] (311 mg, 2.1 mmol) in THF (2 mL) was added dropwise at 08C, and
the mixture was stirred for ca. 10 min at this temperature until gas
evolution had ceased. A solution of bromide 32 (650 mg, 1.75 mmol) in
THF (5 mL) was added over 5 min, and the resulting mixture was stirred at
room temperature for a further 2 h until TLC showed complete con-
sumption of the starting material. The reaction mixture was poured into
water (30 mL) and extracted with CH2Cl2 (5 Â 20 mL). The combined
organic phases were dried (MgSO4) and concentrated in vacuo. Flash
chromatography with petroleum ether/ethyl acetate (1:1) yielded allylic
ether 33 (660 mg, 86%) as a highly viscous, colorless oil. 1H NMR
(4R*,5R*)-(Æ)-4-Isopropyl-5-phenyl-tetrahydrofuran-2-one (cis-30):
A
solution of dimethylacetal 28 was dissolved in acetic acid (80%, 5 mL)
and stirred for 24 h at room temperature. The reaction was terminated by
cautious addition of aqueous saturated NaHCO3 solution (30 mL). The
reaction mixture was extracted with tert-butyl methyl ether (3 Â 15 mL)
and the combined organic phases were dried (MgSO4). Evaporation of the
solvent in vacuo and purification of the residue by flash chromatography
with petroleum ether/tert-butyl methyl ether (4:1) furnished the lactols 29
(44 mg, 75%) as a colorless oil. Lactols 29 were used directly in the
subsequent oxidation.
PCC on Al2O3 (1 mmolgÀ1, 630 mg) was added to a magnetically stirred
solution of lactols 29 (44 mg, 0.21 mmol) in CH2Cl2 (8 mL), and the
resulting suspension was stirred for 26 h at room temperature. The reaction
mixture was filtered through silica gel with CH2Cl2 (200 mL). After
evaporation of the solvent in vacuo, the residue was purified by flash
chromatography with petroleum ether/tert-butyl methyl ether (4:1) to
ˆ
(300.130 MHz, CDCl3): d ˆ 1.42 (s, 3H; CH3), 4.65 (s, 1H; CH2 ), 4.74 (d,
J ˆ 14.1 Hz, 1H; CH2), 4.83 (s, 1H), 4.85 (d, J ˆ 14.0 Hz, 1H; CH2), 4.99 (s,
ˆ
2H; CH2 ), 7.05 (m, 1H; Ar-H), 7.20 7.39 (m, 6H; Ar-H), 7.45 7.67 (m,
11H; Ar-H), 7.91 (m, 1H; Ar-H); 13C NMR (75.469 MHz, CDCl3): d ˆ 17.2,
67.9 (d, J(C,P) ˆ 4.8 Hz, CH2), 85.0, 113.1, 126.2 128.5 (15 Ar-C), 129.5,
131.8 133.9 (7 Ar-C), 140.3, 144.3 (d, J(C,P) ˆ 3.8 Hz); 31P NMR
furnish lactone cis-30 (39 mg, 91%) as
a
colorless oil. 1H NMR
(300.133 MHz, CDCl3): d ˆ 0.70 [d, J ˆ 6.7 Hz, 3H; CH(CH3)2], 0.78 [d,
J ˆ 6.5 Hz, 3H; CH(CH3)2], 1.20 1.27 [m, 1H; CH(CH3)], 2.47 2.58 (m,
Chem. Eur. J. 2003, 9, No. 2
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0433 $ 20.00+.50/0
433

B. Breit et al.
FULL PAPER
‡
(121.495 MHz, CDCl3): d ˆ 31.4 (s); found 439.1792 [M‡H] (HRMS,
[1] B. Breit, W. Seiche, Synthesis 2001, 1 36.
FAB); C29H28O2P calcd 439.1827.
[2] C. D. Frohning, C. W. Kohlpaintner, in Applied Homogeneous Catal-
ysis with Organometallic Compounds (Eds.: B. Cornils, W. A. Herr-
mann), Wiley-VCH, Weinheim, 2000, pp. 29 104.
(Æ)-2-{[(2-Methyl-1-phenylallyl)oxy]methyl}phenyl-(diphenyl)-phosphane
(34): Triethylamine (9.92 g, 98 mmol) was added to a solution of phosphane
oxide 33 (2.07 g, 4.9 mmol) in toluene (15 mL). The solution was cooled to
08C, and trichlorosilane (3.32 g, 24.5 mmol) was added. The reaction
mixture was heated at 1408C in a sealed tube for 45 h. After cooling to
room temperature, the reaction mixture was poured into aqueous sodium
hydroxide solution (5%, 100 mL). After addition of brine (30 mL), the
mixture was extracted with ether (4 Â 100 mL). The combined organic
phases were dried (MgSO4) and concentrated. The residue was purified by
flash chromatography with petroleum ether/tert-butyl methyl ether (98:2)
[3] B. Breit, Chem. Eur. J. 2000, 6, 1519 1524.
[4] S. D. Burke, J. E. Cobb, Tetrahedron Lett. 1986, 27, 4237 4240; R. W.
Jackson, P. Perlmutter, E. E. Tasdelen, J. Chem. Soc. Chem. Commun.
1990, 763 764; I. J. Krauss, C. C.-Y. Wang, J. L. Leighton, J. Am.
Chem. Soc. 2001, 123, 11514 11515.
[5] B. Breit, Angew. Chem. 1996, 108, 3021 3023; Angew. Chem. Int. Ed.
Engl. 1996, 35, 2835 2837; B. Breit, Liebigs Ann. 1997, 1841 1851.
[6] B. Breit, M. Dauber, K. Harms, Chem. Eur. J. 1999, 5, 1819 2827; B.
Breit, S. K. Zahn, J. Org. Chem. 2001, 66, 4870 4877.
1
to furnish phosphine 34 (1.56 g, 77%) as a viscous, colorless oil. H NMR
[7] B. Breit, S. K. Zahn, Angew. Chem. 1999, 111, 1022 1024; Angew.
Chem. Int. Ed. 1999, 38, 969 971; B. Breit, S. K. Zahn, Angew. Chem.
2001, 113, 1964 1967; Angew. Chem. Int. Ed. 2001, 40, 1910 1913.
[8] G. Hˆfle, W. Steglich, H. Vorbr¸ggen, Angew. Chem. 1978, 90, 602
615; Angew. Chem. Int. Ed. Engl. 1978, 17, 569 583.
[9] P. W. N. M. van Leeuwen, C. P. Casey, G. T. Whiteker, in Rhodium
Catalyzed Hydroformylation (Eds.: P. W. N. M. van Leeuwen, C.
Claver), Dordrecht, 2000, Chapter 4, pp. 63 105.
(500.130 MHz, CDCl3): d ˆ 1.49 (s, 3H; CH3), 4.66 (d, J ˆ 12.6 Hz, 1H),
4.74 4.77 (m, 2H), 4.91 (s, 1H), 5.06 (s, 1H), 6.88 (m, 1H; Ar-H), 7.16 7.30
(m, 16H; Ar-H), 7.35 (m, 1H; Ar-H), 7.63 (m, 1H; Ar-H); 13C NMR
(125.758 MHz, CDCl3): d ˆ 17.4 (CH3), 68.5 (d, J(C,P) ˆ 25.0 Hz), 85.0,
ˆ
113.3 (CH2 ), 126.5 (s, 2C), 127.1 (s), 127.5, 127.8 (d, J(C,P) ˆ 5.7 Hz), 128.0
(s, 2C), 128.5 (d, J(C,P) ˆ 7.0 Hz, 4C), 128.6 (s, 2C), 128.9 (s), 133.4 (s),
133.8 (d, J(C,P) ˆ 19.6 Hz, 4C), 135.1 (d, J(C,P) ˆ 15.3 Hz), 136.6 (d,
J(C,P) ˆ 10.2 Hz, 2C), 140.5 (s), 143.1 (d, J(C,P) ˆ 23.6 Hz), 144.9; 31P
NMR (121.495 MHz, CDCl3): d ˆ À16.1 (s); elemental analysis calcd (%)
for C29H27PO (422.5): C 82.44, H 6.44; found C 82.16, H 6.56.
[10] D. Gleich, R. Schmid, W. A. Herrmann, Organometallics 1998, 17,
4828 4834.
[11] G. S. Hammond, J. Am. Chem. Soc. 1955, 77, 334 338.
[12] T. Kazuhide, M. Yabuta, S. Nakamura, T. Yamagata, J. Chem. Soc.
Dalton Trans. 1993, 2781 2789.
[13] W. H. Baddley, M. S. Fraser, J. Am. Chem. Soc. 1969, 91, 3361 3663;
W. H. Baddley, M. S. Fraser, J. Organomet. Chem. 1972, 36, 337 387.
[14] C. P. Casey, G. T. Whiteker, Isr. J. Chem. 1990, 30, 299 304; for a
reviewsee P. W. N. M. van Leeuwen, P. C. J. Kamer, J. N. H. Reek, P.
Dierkes, Chem. Rev. 2000, 100, 2741 2769.
(3R*,4S*)-4-[2-(Diphenylphosphanyl)-benzyloxy]-3-methyl-4-phenylbu-
tyraldehyde (syn-35) and (3R*,4R*)-4-[2-(diphenylphosphanyl)-benzy-
loxy]-3-methyl-4-phenyl-butyraldehyde (anti-35): This compound was
obtained from allylic ether 34 (211 mg, 0.5 mmol) after 24 h at 908C and
20 bar syngas pressure, by the procedure described for the hydroformyla-
tion of o-DPPB esters 1a d and 17 21, a quantitative conversion of
starting material according to NMR analysis of the crude reaction product.
The diastereomer ratio was syn/anti 68:32. Flash chromatography of the
crude product with petroleum ether/tert-butyl methyl ether (9:1) gave
aldehydes 35 (154 mg, 64%) as a highly viscous colorless oil. Signals of the
minor (anti) diastereomer in square brackets. 1H NMR (500.130 MHz,
CDCl3): d ˆ 0.76 (d, J ˆ 6.9 Hz, 3H; CH3), [0.69 (d, J ˆ 6.8 Hz, 3H; CH3],
1.96 (ddd, J ˆ 18.4, 8.0, 1.8 Hz, 1H), [2.07 (m, 1H)], 2.26 (m, 1H)], 2.34 (dd,
J ˆ 16.6 Hz, 5.4 Hz, 1H), [2.51 (dd, J ˆ 16.5 Hz, 5.1 Hz, 1H)], 4.25 (d, J ˆ
5.1 Hz, 1H) [4.01 (d, J ˆ 7.0 Hz, 1H)], 4.47 4.64 (m, 2H), 6.91 (m, 1H; Ar-
H), 7.15 7.36 (m, 17 Ar-H), 7.48 (m, 1H; Ar-H), 9.47 (s, 1H; CHO), [9.54
(s, 1H; CHO)]; 13C NMR (125.758 MHz, CDCl3): d ˆ 15.6 [16.9], 34.4
[35.4], 46.8 [47.5], 69.4 [69.2], 84.7 [86.2], 127.3 [127.4] (2C), 127.8 [127.7],
128.1 [128.2] (2C), 128.4 128.7 (4 Ar-C), 129.0, 133.7 (d, J(C,P) ˆ 19.6 Hz,
4C), 135.4 [135.2], 135.5, 136.6 (d, J(C,P) ˆ 10.1 Hz, 2C), 136.7 (d, J(C,P) ˆ
10.3 Hz, 2C), 139.2 [140.1], 142.8 [142.5], 142.9, 202.4 [202.6]; 31P NMR
(202.457 MHz, CDCl3): d ˆ À18.7 (s); elemental analysis calcd (%) for
C30H29PO2 (452.3): C 79.62, H 6.46; found C 79.73, H 6.51.
[15] The CAChe Molecular Modeling package was used for these
calculations, employing the MM2 parameter set. For the rhodium
atom
a divalent rhodium (unconfigured) was selected with the
modified parameters given in Table 4. The olefin is bound through a
2
À
coordinative bond to the rhodium atom with fixed Rh C(sp ) bond
lengths of 200 pm.
[16] C. K. Cheung, R. S. Wedinger, W. J. le Noble, J. Org. Chem. 1989, 54,
570 573.
[17] C. S. Marvel, R. L. Myers, J. H. Saunders, J. Am. Chem. Soc. 1948, 70,
1694 1699; H. Nakahira, I. Ryu, M. Ikebe, Y. Oku, A. Ogawa, J. Org.
Chem. 1992, 57, 17 28.
[18] M. B. Green, W. J. Hickinbottom, J. Chem. Soc. 1957, 3262 3268.
[19] G. M. Sheldrick, SHELXS-97, Program for Solving Crystal Structures,
University of Gˆttingen, 1997.
[20] G. M. Sheldrick, SHELXL-97, Program for Refining Crystal Struc-
tures, University of Gˆttingen, 1997.
[21] P. G. Stevens, O. C. W. Allenby, A. S. DuBois, J. Am. Chem. Soc. 1940,
62, 1424 1428.
Acknowledgement
This work was supported by the Fonds der Chemischen Industrie, the
Deutsche Forschungsgemeinschaft, and the Alfried-Krupp Award for
young university teachers. We thank Dr. M. Keller, Universit‰t Freiburg,
for the X-ray crystal structure analysis of compound 2d.
Received: August 6, 2002
434
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0434 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 2