1,3-asymmetric induction in stereoselective rhodium-catalyzed hydroformylation of homomethallylic alcohols

Article abstract of DOI:10.1002/(SICI)1099-0690(199806)1998:6<1123::AID-EJOC1123>3.0.CO;2-Y

Introducing ortho-diphenylphosphanyl benzoyl as a substrate bound catalyst directing group (CDG) allows an efficient substrate-directed diastereoselective hydroformylation of acyclic homomethallylic alcohols 5, making use of 1,3-asymmetric induction. The corresponding anti-aldehydes 10 were obtained as the major diastereomer in all cases, with diastereomer ratios of ca. 91:9 (anti-syn). Supporting evidence could be obtained for the ability of the o-DPPB group to act as a catalyst-directing group (CDG) via a reversible catalyst coordination. Finally, a model has been devised that rationalizes the origin of the 1,3-asymmetric induction. This model is based on a conformational analysis (NMR studies, MACROMODEL/MM3 calculations) of the homomethallylic substrates and indicates a relationship between the preferred substrate conformation and the experimentally determined stereoselectivities. In agreement with this model was the predicted significant improvement in stereoselectivity upon hydroformylation of the anti-homomethallylic alcohol derivative 15 (→ 21).

Full text of DOI:10.1002/(SICI)1099-0690(199806)1998:6<1123::AID-EJOC1123>3.0.CO;2-Y

FULL PAPER
Substrate-Directed Diastereoselective Hydroformylations, 2[᭛]
1,3-Asymmetric Induction in Stereoselective Rhodium-Catalyzed
Hydroformylation of Homomethallylic AlcoholsȠ
Bernhard Breit
Fachbereich Chemie der Philipps-Universität Marburg,
Hans-Meerweinstraße, D-35043 Marburg, Germany
Fax: (internat.) ϩ49(0)6421/28 8917
E-mail: breit@ps1515.chemie.uni-marburg.de
Received December 17, 1997
Keywords: Asymmetric synthesis / Catalysis / Hydroformylations / Rhodium compounds / Catalyst-directing groups
Introducing ortho-diphenylphosphanyl benzoyl as
a
devised that rationalizes the origin of the 1,3-asymmetric
induction. This model is based on a conformational analysis
(NMR studies, MACROMODEL/MM3 calculations) of the
substrate bound catalyst directing group (CDG) allows an
efficient substrate-directed diastereoselective hydroform-
ylation of acyclic homomethallylic alcohols 5, making use of
1,3-asymmetric induction. The corresponding anti-aldehydes
10 were obtained as the major diastereomer in all cases, with
diastereomer ratios of ca. 91:9 (anti:syn). Supporting
homomethallylic substrates and indicates
a relationship
between the preferred substrate conformation and the
experimentally determined stereoselectivities. In agreement
with this model was the predicted significant improvement
evidence could be obtained for the ability of the o-DPPB in stereoselectivity upon hydroformylation of the anti-
group to act as a catalyst-directing group (CDG) via a
reversible catalyst coordination. Finally, a model has been
homomethallylic alcohol derivative 15 (Ǟ 21).
Introduction
as to whether the same concept and catalyst-directing group
would permit efficient 1,3-asymmetric induction. In this re-
gard, acyclic homomethallylic alcohol derivatives 3 ap-
peared to be a particularly interesting class of substrates,
since their hydroformylation would furnish an acyclic build-
ing block 4 with hydroxy- and methyl-substituted stereocen-
ters in a 1,3-relation. Such a structural motif is common
within the polyketide class of natural products[6], and its
stereoselective synthesis is therefore of considerable syn-
thetic importance.
Transition metal catalyzed reactions have evolved into
powerful tools for the construction of carbon backbones in
modern organic synthesis. This is particularly due to their
compatibility with multifunctionality in organic substrates,
and their reliable operation under usually mild conditions.
Some of these reactions also allow the generation of new
stereocenters, either with the aid of a chiral catalyst[1] or by
substrate-based asymmetric induction.[2] Despite its attract-
iveness with regard to synthetic efficiency, the latter ap-
proach is not frequently utilized in asymmetric synthesis.[3]
However, we recently found that the rational design of
such processes becomes possible with the aid of a
(CDG).[4] The role of such a group attached
to the substrate is to precoordinate the catalytically active
species, which should result in the selective positioning of
the catalyst at one of two diastereotopic, e.g. olefin faces of
a substrate. Based on this concept, the
-diphenylphos-
phanylbenzoate group ( -DPPB) has been developed as an
effective catalyst-directing group for the stereoselective
hydroformylation of methallylic alcohols 1.[4][5] In this pro-
cess, high 1,2-asymmetric induction gives rise to the forma-
tion of the
-aldehydes 2 as the major diastereomers with
In this paper, full details are given on the use of the
-
diastereoselectivities of up to 96:4. A question then arose
DPPB group as an efficient catalyst-directing group for the
stereoselective rhodium-catalyzed hydroformylation of
acyclic homomethallylic alcohols, giving rise to high 1,3-
[᭛]
For Part 1, see B. Breit,
1998, 1123Ϫ1134
1997, 1841Ϫ1851.
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1434Ϫ193X/98/0606Ϫ1123 $ 17.50ϩ.50/0
1123

B. Breit
FULL PAPER
asymmetric induction.[7] Additionally, a model is described
Interestingly, a significant dependence of the diastereose-
that rationalizes the stereochemical outcome of this reac- lectivity on the reaction temperature was observed. Thus,
tion on the basis of calculated and experimentally deter- reaction of -DPPB ester 10a at 90°C gave only a 70:30
mined preferred substrate conformations in solution.
ratio (Table 1, entry 3). Lowering the reaction temperature
to 70°C and even further to 50°C improved the dia-
Synthesis of Homomethallylic Alcohols and Exploratory
Investigations
stereomer ratio from 87:13 to 91:9 favoring the
hyde (Table 1, entries 1 and 2).
-alde-
Homomethallylic alcohols 5 were prepared according to
ImaiЈs protocol.[8] In exploratory studies, the potential of
the unprotected hydroxyl functionality of a homomethal-
lylic alcohol to act as a catalyst-directing group was evalu-
ated.
Table 1. Temperature dependence of the diastereoselectivity of hy-
droformylation of 9a
[b]
Entry
[°C]
[h]
Yield (%)[a]
:
1
2
3
50
70
90
72
93
99
99
91:9
Thus, alcohol 5a was subjected to hydroformylation con-
ditions (90°C, toluene, 0.35 mol% [Rh(CO)2acac], 7 mol%
PPh3) for 24 h. A low conversion (25%) to the correspond-
ing δ-lactols 6 was observed.[9] Subsequent oxidation of the
lactols 6 with pyridinium chlorochromate (PCC) furnished
24
24
87:13
70:30
[a]Isolated yield after column chromatographic purification. Ϫ
[b]Determined by NMR spectroscopic analysis of the crude pro-
duct.
the known lactones 7,[10] as a 62:38 mixture of
- and
The relative configuration of the hydroformylation prod-
uct 10a was determined by transformation to the literature-
known lactones 7.[10] Thus, the aldehyde functionality in
10a was protected as a dimethyl acetal (Ǟ 11). Subsequent
-7, indicating the inability of a hydroxyl group to act as
an efficient catalyst-directing group. This result parallels
observations made previously in the methallylic alcohol
series.[4]
alkaline hydrolysis furnished the alcohol 12 as well as
-
DPPBA (8). Deprotection of the acetal employing lithium
tetrafluoroborate in acetonitrile in the presence of 2% water
furnished the δ-lactols 6. Finally, PCC oxidation led to the
known -lactone 7.[10]
Synthesis and Hydroformylation of Homomethallylic o-
DPPB Esters
At this stage it was decided to explore the potential of
the recently developed catalyst-directing -DPPB group to
make more efficient use of the substrate-inherent chirality
information. For this purpose, homomethallylic -DPPB es-
ters 9 were prepared from homomethallylic alcohols 5 and
-diphenylphosphanylbenzoic acid [ -DPPBA (8)] em-
ploying the DCC/DMAP esterification protocol.[11]
Hydroformylation of 9a at 50°C furnished aldehyde 10a
in 91% yield with a diastereomer ratio of 91:9 (
:
).
1124
1998, 1123Ϫ1134

Substrate-Directed Diastereoselective Hydroformylations, 2
FULL PAPER
In order to study the influence of the -DPPB group on
the stereoselectivity of the hydroformylation reaction, the
benzoate 13 was prepared, which differs from the -DPPB
ester 9a in that the phosphane moiety is replaced by a CH
unit. Thus, derivative 13 should possess almost the same
structural features as the corresponding -DPPB ester 9a,
but should lack the ability to temporarily coordinate to the
catalytically active transition metal center.[12]
Table 2. Results of the substrate-directed diastereoselective hydro-
formylation of homomethallylic -DPPB esters 9
Hydroformylation of the homomethallylic esters 9a and
13 furnished the aldehydes 10a and 14, respectively. The -
DPPB derivative 9a could be reacted at 50°C, whereas the
phosphane-lacking derivative 13 required a slightly higher
reaction temperature of 70°C for hydroformylation to pro-
ceed smoothly, indicating a rate-enhancing effect of the in-
tramolecular phosphane ligand.[12] The stereochemical out-
come of these two reactions proved to be most interesting.
Thus, -DPPB ester 9a gave predominantly the
-alde-
hyde 10a in a diastereomer ratio of 91:9. In contrast, the
phosphane-free system 13 furnished preferentially the
-
aldehyde 14 in a diastereomer ratio of 71:29.[12] Evidently,
the homomethallylic substrate has an inherent ability (pre-
sumably of spatial origin) to direct the rhodium catalyst to
the diastereotopic olefin face, which leads to the formation
of the
-aldehyde. However, the -DPPB group is capable
of overruling these sterically-based directing effects, giving
rise to the formation of the opposite diastereomer. This re-
sult underlines the ability of the -DPPB group to act as a
(CDG) by reversible coordination
of the catalyst, and is in agreement with similar obser-
vations that have been made with the corresponding me-
thallylic alcohol -DPPB esters (1 Ǟ 2).[4]
To check whether the diastereoselectivity shows a depen-
dence on the nature of the substituent R at the stereogenic
center in the hydroformylation of the homomethallylic al-
cohol -DPPB esters 9, the hydroformylation of the homo-
methallylic -DPPB esters 9bϪf was investigated (see
Table 2).
[a]
[b]
Isolated yield after column chromatographic purification. Ϫ
Determined by NMR spectroscopic analysis of the crude product.
In all cases, hydroformylation was found to proceed
smoothly to give the corresponding aldehydes 10 in good and gave diastereomer ratios of ca. 91:9, irrespective of the
to excellent yields and with -diastereoselectivities of up nature of R. The hydroformylation of the alkenyl-substi-
to 91%. Surprisingly, and in striking contrast to previous tuted derivative 9f, in which a 1,1-disubstituted alkene
results on the hydroformylation of methallylic -DPPB es- could be reacted chemo-, regio- and stereoselectively in the
ters[4], no significant influence of the substituent R on the presence of a trisubstituted alkene to give the
-aldehyde
diastereoselectivity could be detected. Thus, primary and 10f in 85% yield and with a diastereomer ratio of 90:10
secondary alkyl as well as aryl substituents were tolerated (Table 2, entry 6) was particularly interesting.
1998, 1123Ϫ1134
1125

B. Breit
FULL PAPER
A comparison of the NMR spectra of the aldehydes 10 MACROMODEL/MM3 calculations, which indicate 9aA
revealed characteristic differences between the - (major) to be the preferred conformation at ambient temperature
and
- (minor) diastereomers. Thus, the 13C-NMR reso- (ca. 60% populated at 25°C, see Figure 1).[16]
nances of the methyl carbons at C3 are shifted to higher
field in
-10aϪf relative to those in
-10aϪf (see Table
3). The same holds for the 13C-NMR resonances of the oxy-
gen-substituted carbon atom C1. Accordingly, with the de-
rivatization of
point of reference, 13C-NMR spectroscopy allows the as-
signment of - and -configuration, respectively.
-10a to the known cis-δ-lactone 7 as a
Table 3. Comparison of selected NMR data of
- and
-alde-
hydes 10
Com-
pound
R
13C NMR resonan- 13C NMR resonances
ces of C1, δ of methyl-C at C 3, δ
10a
10b
10c
10d
10e
10f
21
-Pr
-Hex
77.07
77.15
76.69
73.40
75.44
69.79
79.10
79.90
19.32
19.49
19.65
19.84
19.74
19.76
19.98
20.60
20.53
20.54
20.27
20.54
19.92
20.12
76.57
73.20
75.03
69.74
-Hex
Ph
-MeOC6H4
( )-EtCHϭCMe 78.16
-Pr
79.60
Preferred Substrate Conformation and Model for
1,3-Asymmetric Induction
The aforementioned investigations showed that hydrofor-
mylation of the homomethallylic -DPPB esters 9 led to
diastereomer ratios of about 90:10, almost independently of
the nature of the substituent at the stereogenic center. From
this, it could be assumed that the observed 1,3-asymmetric
induction stemmed from a conformational preference in-
herent to the homomethallylic system itself. This hypothesis
was further supported by the strong temperature depen-
dence that was observed on hydroformylation of derivative
9a (see Table 1).
To gain further insight, the preferred conformation of the
homomethallylic -DPPB ester 9a in solution at 25°C was
examined by NMR techniques.[14] Thus, Ha showed a large
vicinal coupling constant of 9.0 Hz to Hc, whereas coupling
between Hb and Hc was smaller (4.1 Hz). Furthermore, the
2D-NOESY spectra revealed a strong NOE contact be-
tween the protons of the methyl substituent at the olefin
and Hc, as well as between one of the olefinic protons and
Ha. Both the vicinal coupling constants as well as the ob-
served NOE contacts showed compound 9a to have the pre-
ferred conformation 9aA.[15] This is in agreement with
1126
1998, 1123Ϫ1134

Substrate-Directed Diastereoselective Hydroformylations, 2
FULL PAPER
Figure 1. Energetically most stable conformation A of 9a according
to MACROMODEL/MM3 (ca. 60% populated at 25°C)
interaction reaches a maximum in conformation 9aB, but is
minimized in conformation 9aA. Thus, the observed reac-
tion pathway via conformer 9aA to the
stereomer -10a might be preferred due to the avoidance
of this repulsive -pentane interaction. On the other
major dia-
hand, if this assumption is correct, the incorporation of an
additional conformational destabilizing (repulsive) interac-
tion in conformer 9aB only, should disfavor conformer B
even more compared to conformer A, and ultimately,
should give rise to higher
-diastereoselectivity in the co-
urse of the hydroformylation reaction. Such an additional
repulsive interaction could result from incorporation of ad-
ditional 1,3-allylic strain[19] into conformer B. Such an in-
teraction would arise if Hb in 9a were to be replaced by a
methyl substituent.
Scheme 2. Preferred conformation A and minor conformation B of
15; the indicated coordination of the rhodium catalyst
is hypothetical
Scheme 1. Preferred conformation A and minor conformation B of
9a; the indicated coordination of the rhodium catalyst
is hypothetical
Substitution of Hb by a methyl group leads to the homo-
methallylic -DPPB ester 15, in which the oxygen-bearing-
and the methyl-substituted stereocenters have an
re-
If one assumes a coordination of the rhodium catalyst lation. MACROMODEL/MM3 force field calculations
between the catalyst-directing group and the alkene as indi- indicate that the homomethallylic -DPPB ester 15 has the
cated (Scheme 1), hydroformylation would furnish the
aldehyde
-
expected preferred conformation 15A, which is populated
-10a, as was indeed found experimentally. The to an extent of about 74% at 25°C (see Figure 2).[16] In
less preferred conformation 9aB (about 5 kJ/mol less stable other words, according to the force field calculations, com-
than conformation 9aA according to MACROMODEL/ pound 15, bearing the additional methyl substituent, shows
MM3) should give rise to complexation of the opposite dia- an enhanced conformational preference for conformer A.
stereotopic alkene face, i.e. to the formation of the corre-
On the other hand, substitution of Ha by a methyl group
sponding -aldehyde ( -10a). According to the MAC- (Ǟ16) should give rise to an additional repulsive 1,3-allylic
ROMODEL/MM3 calculations, conformation B is popu- strain in conformer 16A, which would destabilize this con-
lated to an extent of about 10% at 25°C. All other confor- formation (see Scheme 3). Consequently, for derivative 16,
mations of 9a (ca. 30% at 25°C) do not permit a chelating neither conformation A nor conformation B would be sig-
bidentate binding mode of the olefin and -DPPB group. nificantly favored. According to our model, this should re-
Assuming that the intramolecular process involving preco- sult in a low diastereoselectivity upon hydroformylation of
ordination of the catalyst by the -DPPB group is the only the
-homomethallylic -DPPB ester 16.
reaction pathway in operation,[17] other conformations can
This prediction was confirmed by a conformational
be neglected.
analysis (MACROMODEL/MM3) of derivative 16. Thus,
The question then arises as to the nature of the underly- neither conformer 16A nor conformer 16B was shown to
ing reasons for conformation 9aA being energetically fa- be significantly populated at ambient temperature (percent-
vored compared to conformation 9aB. The only argument age of both conformers of all populated conformations at
one could put forward is the avoidance of a repulsive
-
25°C amounts to < 10%).[16] Interestingly, the calculation
pentane interaction[18] between the oxygen substituent and indicated conformer 16C to be the major populated species
the methyl substituent at the olefin. Such a destabilizing at 25°C at a level of 87% (see Figure 3).
1998, 1123Ϫ1134
1127

B. Breit
FULL PAPER
Figure 2. Energetically most stable conformation A of 15 according
to MACROMODEL/MM3 (ca. 74% populated at 25°C)
The corresponding
-homomethallylic alcohol 17 (
Ն 95:5) was obtained according to Moise and Szymoniak
employing a Ti(III) allylating reagent prepared from
[Cp2TiCl2], isoprene and 2 equiv. of isopropylmagnesium
bromide.[20] Stereoselective synthesis of the
-dia-
stereomer 19 was achieved in moderate yield starting from
isovaleraldehyde and the known allylstannane 18[21]. Esteri-
fication with -DPPBA employing the DCC/DMAP proto-
col gave the corresponding -DPPB esters 15 and 16.
Scheme 3. Repulsive interactions in conformation A and conforma-
tion B of 16; the indicated coordination of the rhodium
catalyst is hypothetical
At first, it had to be verified that
-derivative 15 did
indeed populate conformation 15A in solution, as was pre-
dicted by our model and MACROMODEL/MM3 calcu-
lations. These investigations had to be performed on a suit-
able model system, which needed to be as similar as possible
to the situation in the course of the hydroformylation reac-
tion. The rhodium complex 20, which was obtained from
15 by reaction with [Rh(CO)2acac] in benzene at room tem-
perature, was considered to be an appropriate model sys-
tem.
Figure 3. Energetically most stable conformation C of 16 according
to MACROMODEL/MM3 (ca. 87% populated at 25°C)
The structure of 20 was confirmed by NMR spec-
troscopy. Thus, the 31P-NMR resonance of 20 is shifted
downfield by about 60 ppm (20: δ ϭ 54.8) compared to that
of the free phosphane 15 (δ ϭ Ϫ4.1), as expected for an η1-
coordination of a triarylphosphane to a rhodium(I) central
atom. Furthermore, the 31P-NMR resonance shows a
1
A conformer such as 16C does not permit the required characteristic splitting to a doublet, with a typical
P,Rh
chelating binding mode of the phosphane and olefin. There- coupling constant of 189.1 Hz, indicating that the P atom
fore, conformation 16C can be neglected as a reactive con- resides in the direct neighborhood of the rhodium nu-
formation for the -DPPB-directed hydroformylation reac- cleus.[22] The 13C-NMR spectrum reveals the presence of a
tion, even though it is significantly populated.
carbonyl ligand by a typical resonance at low field (δ ϭ
1128
1998, 1123Ϫ1134

Substrate-Directed Diastereoselective Hydroformylations, 2
FULL PAPER
Darmstadt, 40Ϫ63 µm. Ϫ Hydroformylation reactions were per-
formed in 100 or 200 ml stainless steel autoclaves equipped with
magnetic stirrers. Ϫ Gases: Carbon monoxide 2.0 (Messer-Gries-
heim), hydrogen 3.0 (Messer-Griesheim).
189.6) and a characteristic splitting to a doublet of doublets
1
2
(
ϭ 76.1 Hz,
ϭ 25.4 Hz).[23]
C,P
C,Rh
To determine the preferred conformation of the rhodium
complex 20 in solution, 2D NOESY and selective proton-
decoupling experiments were performed. Thus, NOEЈs and
vicinal coupling constants were observed for the rhodium
complex 20 similar to those found in the case of 9a (see
Scheme 2). Consequently, compound 20 can be assigned the
same preferred conformation A in solution at room tem-
perature as 9a, although it is impossible to determine the
differences in conformer populations between 9a and 15/20
on the basis of these experimental data.
However, of major interest was the influence of the ad-
ditional methyl substituents attached to the -DPPB esters
15 and 16 on the stereochemical outcome of the hydrofor-
mylation reaction. When 15 was hydroformylated at 50°C,
5:[8] To a magnetically stirred solution of the appropriate
aldehyde (50 mmol) and 4.95 g (55 mmol) of methallyl chloride in
100 ml of DMF, was added 16.9 g (75 mmol) of tin dichloride
dihydrate followed by 11.24 g (75 mmol) of sodium iodide. The
reaction mixture was stirred for 18 h at room temp., and then a
mixture of 50 ml of 25% aq. NH4F solution and 100 ml of
-
butyl methyl ether was added. The mixture was stirred vigorously
and then the layers were allowed to separate. The aqueous phase
was reextracted with two 80 ml portions of
-butyl methyl ether.
The combined organic phases were dried (Na2SO4), the solvent was
evaporated, and the residue was purified by distillation to give the
alcohol 5 as a colorless liquid.
(5a): From 3.6 g (50 mmol) isobutyral-
dehyde was obtained 5.28 g (85%) of 5a (b.p. 60°C/40 mbar). Spec-
troscopic and analytical data were identical to those reported pre-
viously.[24]
the
-aldehyde 21 was obtained in 91% yield with a dia-
stereomer ratio of 96:4. This is a significantly higher dia-
stereoselectivity than was found in the absence of the ad-
ditional methyl substituent (compare 9a Ǟ 10a, 91:9) and
is therefore fully consistent with the proposed model. Ac-
cordingly, hydroformylation of the
ceeded in a stereorandom fashion, also as predicted by
this model.
(5b): From 5.61 g (50 mmol)
cyclohexane carbaldehyde was obtained 6.12 g (73%) of 5b (b.p.
125°C/50 mbar). Analytical data were identical to those reported
previously.[25] Ϫ 1H NMR (200 MHz, CDCl3): δ ϭ 1.1Ϫ2.0 (m, 11
H, -Hex), 1.8 (s, 3 H, CH3), 2.1 (dd, ϭ 12.5 Hz, 10.0 Hz, 1 H,
CH2), 2.3 (d, ϭ 12.5 Hz, 1 H, CH2), 3.5 (m, 1 H, OCH), 4.85
(br s, 1 H, OH), 4.93 (s, 1 H, ϭCH2), 5.02 (s, 1 H, ϭCH2). Ϫ 13C
NMR (50.329 MHz, CDCl3): δ ϭ 22.1, 26.1, 26.2, 26.6, 28.0, 29.0,
42.9, 43.3, 72.4, 113.2, 143.2.
-derivative 16 pro-
(5c): From 5.7 g (50 mmol) 1-heptanal was
obtained 6.9 g (81%) of 5c. Spectroscopic and analytical data were
identical to those reported previously.[26]
(5d): From 5.3 g (50 mmol) benz-
aldehyde was obtained 5.76 g (71%) of 5d (b.p. 118°C/12 mbar).
Spectroscopic and analytical data were identical to those reported
previously.[8]
(5e): From 6.8 g (50 mmol) 2-
methoxybenzaldehyde was obtained 4.45 g (47%) of 5e (b.p.
129Ϫ134°C/8 mbar). Spectroscopic and analytical data were ident-
ical to those reported previously.[27]
This work was supported by the
the , and by the
(SFB 260). The author thanks Prof.
for his continuous support, the companies
,
(5f): From 4.9 g (50 mmol) 2-
methyl-2-pentenal was obtained 7.37 g (95%) of 5f (b.p. 65°C/40
mbar). Ϫ H NMR (200 MHz, CDCl3): δ ϭ 0.93 (t, ϭ 7.6 Hz,
3 H, CH2C 3), 1.55 (s, 3 H, CH3), 1.67 (s, 3 H, CH3), 1.8 (s, 1 H,
OH), 1.96 (m, 2 H), 2.17 (d, ϭ 7.0 Hz, 2 H, H at C3), 4.05 (t,
ϭ 6.6 Hz, 1 H, OCH), 4.73 (d, ϭ 1.0 Hz, 1 H, ϭCH2), 5.36 (t,
ϭ 7.2 Hz, 1 H, ϭCH). Ϫ 13C NMR (50.329 MHz, CDCl3): δ ϭ
11.3, 13.89, 20.74, 22.18, 44.24, 74.58, 113.24, 128.14, 135.84,
142.64. Ϫ C10H18O (154.3): calcd. C 77.87, H 11.76; found C 77.78,
H 11.74.
1
and
for
their gifts of chemicals, and
tal assistance.
for his skillful experimen-
Experimental Section
Reactions were performed in flame-dried glassware
either under argon (purity > 99.998%) or under nitrogen. The sol-
vents were dried by standard procedures, distilled, and stored under
nitrogen. All temperatures quoted are uncorrected. Ϫ 1H-, 13C-
±
[(±)-
5a]: To a solution of 2 mg (7.7 ϫ 10Ϫ3 mmol) of [Rh(CO)2(acac)]
NMR spectra: Bruker ARX-200, Bruker AC-300, Bruker WH-400, in 2 ml of toluene at 20°C (exclusion of air and moisture), 40.5 mg
Bruker AMX-500 with tetramethylsilane (TMS), chloroform
(1.55 ϫ 10Ϫ1 mmol) PPh3 was added and the mixture was stirred
(CHCl3) or benzene (C6H6) as internal standards. Ϫ 31P-NMR for 15 min. at 20°C. Then, 278 mg (2.17 mmol) (±)-5a was added
spectra: Bruker WH 400 (161.978 MHz) with 85% H3PO4 as exter-
nal standard. Ϫ Melting points: Melting point apparatus by Dr.
and the resulting solution was cannulated into a stainless steel
autoclave, which had been repeatedly evacuated and refilled with
Tottoli (Büchi). Ϫ Elemental analyses: CHN rapid analyzer argon. The flask and the cannula were rinsed with an additional 1
(Heraeus). Ϫ Flash chromatography: Silica gel Si 60, E. Merck AG,
1998, 1123Ϫ1134
ml of toluene. The autoclave was heated to 90°C, then pressurized
1129

B. Breit
FULL PAPER
1
successively with carbon monoxide (10 bar) and hydrogen (10 bar),
and the reaction solution was stirred under these conditions for
23 h. Thereafter, the autoclave was cooled rapidly to 20°C and
viscous oil. Ϫ H NMR (300 MHz, CDCl3): δ ϭ 0.8Ϫ1.9 (m, 11
H), 1.6 (s, 3 H, CH3), 2.17 (m, 2 H, H at C2), 4.6 (s, 2 H, ϭCH2),
4.97 (d pseudo t, ϭ 8.1, 5.3 Hz, 1 H, OCH), 6.81 (m, 1 H, ArH),
depressurized. The degree of conversion was determined to be 25% 7.17Ϫ7.3 (m, 12 H, ArH), 7.59 (m, 1 H, ArH). Ϫ 13C NMR (75.469
by GC. The solvent was evaporated and the residue was subjected
to PCC oxidation conditions without further purification.
MHz, CDCl3): δ ϭ 22.24, 24.98, 26.04, 26.27, 27.76, 29.07, 39.9,
41.3, 76.54, 113.13, 127.94, 128.3 (d, ϭ 6.5 Hz, 4 C), 128.44
(2 C), 130.48 (d, ϭ 1.96 Hz), 131.59, 133.86 (d,
C,P
ϭ 20.83
C,P
C,P
±
Hz, 2 C), 134.02 (d, C,P ϭ 20.9 Hz, 2 C), 134.05, 134.37 (d, C,P ϭ
17.7 Hz), 138.13 (d, ϭ 9.9 Hz), 138.28 (d, ϭ 12.5 Hz),
140.76 (d,
ϭ 27.9 Hz), 141.91, 166.08. Ϫ 31P NMR (161.978
(7): To a solution of the aforementioned residue in 10 ml
CH2Cl2, 4 g of PCC on Al2O3 (1 mmol/g) was added and the sus-
pension was stirred for 12 h at room temp. The reaction mixture
was then filtered through silica gel with a further 50 ml of CH2Cl2,
the solvent was evaporated, and the residual product was analyzed
C,P
C,P
C,P
MHz, CDCl3): δ ϭ Ϫ3.8. Ϫ C30H27O2P (456.6): calcd. C 78.92, H
7.29; found C 78.64, H 7.17.
±
by NMR to determine the diastereomer ratio [62:38 (
:
)].
(9c): From 0.851 g (10 mmol) of homomethallylic alcohol (±)-
5c, 1.834 g (80%) of (±)-9c was obtained as a colorless, viscous oil.
Yield of lactones after purification by flash chromatography with
petroleum ether/ -butyl methyl ether (4:1): 56 mg (66% of theo-
retical). Spectroscopic and analytical data matched those reported
previously.[10]
Ϫ
1H NMR (300 MHz, CDCl3): δ ϭ 0.86 (t, ϭ 6.5 Hz, 3 H,
CH3), 1.17Ϫ1.29 (m, 8 H), 1.47 (m, 2 H), 1.65 (s, 3 H, CH3), 2.13
(dd, ϭ 14.4, 5.7 Hz, 1 H, H at C2), 2.25 (dd, ϭ 14.2, 7.3 Hz,
1 H, H at C2), 4.65 (d, ϭ 1.0 Hz, 1 H, ϭCH2), 4.69 (pseudo t,
ϭ 1.7 Hz, 1 H, ϭCH2), 5.15 (m, 1 H, OCH), 6.88 (m, 1 H, ArH),
7.2Ϫ7.4 (m, 12 H, ArH), 8.05 (m, 1 H, ArH). Ϫ 13C NMR (75.469
MHz, CDCl3): δ ϭ 14.18, 22.57, 22.69, 25.4, 29.25, 31.79, 34.05,
To a solution
of 1 equiv. of the appropriate homomethallylic alcohol in CH2Cl2
(0.5 ), 1 equiv. of -DPPBA 8,[28] 0.1 equiv. of DMAP and 1.1
equiv. of DCC were successively added, and the resulting mixture
was stirred at room temperature until TLC analysis indicated com-
plete consumption of the starting material. Then, the reaction mix-
ture was filtered through a plug of CH2Cl2-wetted Celite, which
was subsequently washed with further CH2Cl2. An appropriate
amount of silica gel was added to the filtrate, and the solvent was
removed. Flash chromatography with petroleum ether/ -butyl
methyl ether (9:1) furnished the -DPPB esters 9, usually as slightly
yellow to colorless, highly viscous oils.
42.88, 73.56, 113.34, 128.2, 128.52 (d,
ϭ 5.2 Hz, 4 C), 128.62
C,P
(2 C), 130.65 (d, C,P ϭ 2.6 Hz), 131.78, 134.06 (d,
ϭ 20.8 Hz,
C,P
2 C), 134.11 (d,
18.8 Hz), 138.38 (d,
140.47 (d,
ϭ 27.4 Hz), 141.88, 166.47. Ϫ 31P NMR (161.978
ϭ 20.8 Hz, 2 C), 134.33, 134.99 (d,
ϭ
C,P
C,P
ϭ 12.0 Hz), 138.44 (d,
ϭ 12.3 Hz),
C,P
C,P
C,P
MHz, CDCl3): δ ϭ Ϫ4.1. Ϫ C30H35O2P (458.6): calcd. C 78.58, H
7.69; found C 78.36, H 7.73.
±
±
(9d): From 1.62 g (10 mmol) of homomethallylic al-
(9a): From 1.28 g (10 mmol) of homomethallylic
alcohol (±)-5a, 3.218 g (77%) of (±)-9a was obtained as a colorless,
cohol (±)-5d, 4.22 g (94%) of (±)-9d was obtained as a colorless,
viscous oil. Ϫ 1H NMR (300 MHz, CDCl3): δ ϭ 1.69 (s, 3 H,
CH3), 2.5 (dd, ϭ 14.0, 6.1 Hz, 1 H, CH2), 2.72 (dd, ϭ 14.0, 8.0
Hz, 1 H, CH2), 4.69 (s, 1 H, ϭCH2), 4.75 (s, 1 H, ϭCH2), 6.15
(dd, ϭ 8.0, 6.2 Hz, 1 H, OCH), 6.96 (m, 1 H, ArH), 7.29Ϫ7.44
(m, 17 H, ArH), 8.16 (m, 1 H, ArH). Ϫ 13C NMR (75.469 MHz,
CDCl3): δ ϭ 22.49, 44.43, 74.88, 113.8, 126.72, 127.75 (2 C), 128.08
(2 C), 128.33, 128.35 (d, C,P ϭ 7.1 Hz, 4 C), 128.42, 128.48, 130.61
1
viscous oil. Ϫ H NMR (500 MHz, C7D8): δ ϭ 0.80 (d, ϭ 6.9
Hz, 3 H, CH3), 0.84 (d, ϭ 6.8 Hz, 3 H, CH3), 1.61 (br s, 3 H,
CH3), 1.69 (m, 1 H, CH), 2.06 (dd, ϭ 14.1, 4.1 Hz, 1 H, CH2),
2.21 (dd, ϭ 14.1, 9.0 Hz, 1 H, CH2), 4.68 (s, 1 H, ϭCH2), 4.7 (s,
1 H, ϭCH2), 5.17 (d pseudo t, ϭ 9.2, 4.5 Hz, 1 H, OCH), 6.96
(m, 1 H, Ar H), 7.05Ϫ7.1 (m, 8 H, ArH), 7.32 (m, 4 H, ArH), 8.14
(m, 1 H, ArH).
(d,
ϭ 2.5 Hz), 131.77, 133.87 (d,
ϭ 20.6 Hz, 2 C), 133.94
C,P
C,P
(d, C,P ϭ 20.8 Hz, 2 C), 134.24, 134.56 (d, C,P ϭ 18.9 Hz), 138.02
(d, ϭ 12.5 Hz), 138.18 (d, ϭ 11.8 Hz), 140.06, 140.53 (d,
Table 4. Temperature dependence of selected vicinal proton-proton
coupling constants of 9a in [D8]toluene (400 MHz)
C,P
C,P
ϭ 27.7 Hz), 140.83, 165.73. Ϫ 31P NMR (161.978 MHz,
C,P
CDCl3): δ ϭ Ϫ4.1. Ϫ C30H27O2P (450.5): calcd. C 79.98, H 6.04;
found C 79.85, H 5.91.
Temperature
[K]
Vicinal coupling constants [Hz] of the
sp3 CH2 unit of 9a
±
298
303
323
343
363
9.0/4.1
9.0/4.1
8.7/4.3
8.5/4.5
8.4/4.6
(9e): From 0.961 g (5 mmol) of homo-
methallylic alcohol (±)-5e, 1.78 g (74%) of (±)-9e was obtained as
colorless crystals, m.p. 130°C. Ϫ 1H NMR (400 MHz, CDCl3): δ ϭ
1.6 (s, 3 H, CH3), 2.46 (dd, ϭ 14.0, 8.6 Hz, 1 H, CH2), 2.51 (dd,
ϭ 14.0, 4.8 Hz, 1 H, CH2), 3.78 (s, 3 H, OCH3), 4.59 (s, 1 H, ϭ
CH2), 6.49 (dd, ϭ 8.3, 5.3 Hz, 1 H, OCH), 6.8Ϫ6.9 (m, 3 H,
ArH), 7.17Ϫ7.45 (m, 14 H, ArH), 8.15 (m, 1 H, ArH). Ϫ 13C NMR
(75.469 MHz, CDCl3): δ ϭ 22.57, 44.15, 55.53, 69.93, 110.62,
113.35, 120.63, 126.6, 128.33, 128.52 (d, C,P ϭ 7.1 Hz, 4 C), 128.56
(2 C), 128.63, 129.25, 130.79 (d,
C,P ϭ 20.5 Hz, 2 C), 134.16 (d, C,P ϭ 20.8 Hz, 2 C), 134.48, 134.9
(d, ϭ 18.5 Hz), 138.34 (d, ϭ 12.9 Hz), 138.44 (d,
11.6 Hz), 140.74 (d,
NMR (161.978 MHz, CDCl3): δ ϭ Ϫ4.4. Ϫ C31H29O3P (480.5):
13C NMR (75.469 MHz, CDCl3): δ ϭ 17.48, 18.68, 22.28, 31.39,
39.81, 76.86, 113.18, 128.03, 128.35 (d,
ϭ 6.9 Hz, 2 C), 128.48 (2 C), 130.4 (d,
131.65, 133.9 (d,
ϭ 7.8 Hz, 2 C), 128.38
C,P
(d,
ϭ 2.3 Hz),
ϭ 20.8 Hz,
C,P
C,P
ϭ 20.8 Hz, 2 C), 134.05 (d,
C,P
C,P
ϭ 2.2 Hz), 131.91, 133.96 (d,
C,P
2 C), 134.15, 134.56 (d,
ϭ 17.8 Hz), 138.23 (d,
ϭ 11.7
C,P
C,P
Hz), 138.33 (d, C,P ϭ 12.5 Hz), 140.82 (d, C,P ϭ 27.6 Hz), 141.86,
166.09. Ϫ 31P NMR (161.978 MHz, CDCl3): δ ϭ Ϫ3.9. Ϫ
C27H29O2P (416.5): calcd. C 77.86, H 7.02; found C 77.75, H 7.15.
ϭ
C,P
C,P
C,P
ϭ 27.7 Hz), 141.77, 156.31, 165.7. Ϫ 31P
C,P
calcd. C 77.48, H 6.08; found C 77.46, H 6.33.
±
(9b): From 0.841 g (5 mmol) of homomethallylic
alcohol (±)-5b, 1.436 g (63%) of (±)-9b was obtained as a colorless,
±
(9f): From 0.771 g (5 mmol) of homomethal-
1130
1998, 1123Ϫ1134

Substrate-Directed Diastereoselective Hydroformylations, 2
FULL PAPER
lylic alcohol (±)-5f, 1.37 g (62%) of (±)-9f was obtained as a color-
(300 MHz, CDCl3): δ ϭ 0.82Ϫ1.76 (m, 13 H), 0.88 (d, ϭ 6.5 Hz,
3 H, CH3), 1.87 (m, 1 H), 2.13 (ddd, ϭ 11.5, 6.5, 2.2 Hz, 2 H, H
1
less, viscous oil. Ϫ H NMR (300 MHz, CDCl3): δ ϭ 0.9 (t,
ϭ
7.5 Hz, 3 H, CH2C 3), 1.58 (s, 3 H, CH3), 1.65 (s, 3 H, CH3), 1.97 at C4), 5.02 (ddd, ϭ 10.1, 5.9, 2.5 Hz, H at C1), 6.91 (m, 1 H,
(m, 2 H, C 2CH3), 2.25 (dd, ϭ 14.0, 6.3 Hz, 1 H, CH2), 2.39 ArH), 7.27Ϫ7.46 (m, 12 H, ArH), 7.95 (m, 1 H, ArH), 9.68 (pseudo
(dd, ϭ 14.0, 8.0 Hz, 1 H, CH2), 4.65 (s, 1 H, ϭCH2), 4.68 (s, 1 t, ϭ 2.3 Hz, 1 H, CHO- ), [9.48 (dd, ϭ 3.2, 1.1 Hz, CHO-
H, ϭCH2), 5.47 (m, 2 H, OCH and ϭCH), 6.9 (m, 1 H, ArH),
)]. Ϫ 13C NMR (75.469 MHz, CDCl3): δ ϭ 19.49, 24.96, 26.08,
7.26Ϫ7.39 (m, 12 H, ArH), 8.06 (m, 1 H, ArH). Ϫ 13C NMR 26.14, 26.41, 28.28, 28.97, 38.66, 42.18, 51.58, 76.61, 128.31, 128.53
(75.469 MHz, CDCl3): δ ϭ 11.9, 13.94, 20.98, 22.93, 41.49, 78.49, (d, ϭ 7.3 Hz, 2 C), 128.56 (d, ϭ 7.0 Hz, 2 C), 128.73 (2
C,P
C,P
113.34, 128.25, 128.52 (d, C,P ϭ 7.25 Hz, 4 C), 128.64, (2 C) 130.68
C), 130.91 (d,
ϭ 2.3 Hz), 132.02, 133.95 (d,
ϭ 20.7 Hz, 2 C), 134.38 (d,
ϭ 20.8 Hz, 2
ϭ 18.0 Hz),
ϭ 12.8 Hz),
), [202.86
C,P
C,P
C,P
C,P
(d, ϭ 2.3 Hz), 131.12, 131.82, 132.11, 134.05 (d, ϭ 20.7 C), 134.27 (d,
C,P
C,P
C,P
Hz, 2 C), 134.14 (d, C,P ϭ 20.8 Hz, 2 C), 134.36, 135.03 (d, C,P ϭ 134.41, 138.28 (d,
18.6 Hz), 138.37 (d, ϭ 13.7 Hz), 138.34 (d, ϭ 11.5 Hz), 140.77 (d,
140.57 (d,
ϭ 12.8 Hz), 138.37 (d,
C,P
ϭ 27.7 Hz), 166.68, 202.67 (CHO-
C,P
C,P
C,P
ϭ 27.3 Hz), 141.6, 165.9. Ϫ 31P NMR (161.978 (CHO- )]. Ϫ 31P NMR (161.978 MHz, CDCl3): δ ϭ Ϫ3.7. Ϫ
C,P
MHz, CDCl3): δ ϭ Ϫ4.2. Ϫ C29H31O2P (442.5): calcd. C 78.71, H C31H35O3P (486.6): calcd. C 76.52, H 7.00; found C 76.37, H 7.28.
7.06; found C 78.50, H 6.82.
±
(9). Ϫ
[(±)-
-10c was obtained. Reaction temperature
). Ϫ
ϭ 7.2 Hz, 3 H,
resulting mixture was stirred at this temperature for 15 min. Sub- CH2C 3), 0.9 (d, ϭ 6.9 Hz, 3 H, CHC 3), 1.15Ϫ1.6 (m, 12 H),
-10c]: From 229 mg (0.5 mmol) 9c,
To a solution of 0.9 mg (3.5 ϫ 10Ϫ3 mmol) 198 mg (81%) of
[Rh(CO)2(acac)] in 3 ml toluene at 20°C (exclusion of air and 30°C. Reaction time 168 h. Diastereomer ratio 90:10 (
moisture), 4.5 mg (1.4 ϫ 10Ϫ2 mmol) P(OPh)3 was added and the 1H NMR (300 MHz, CDCl3): δ ϭ 0.86 (t,
sequently, -DPPB ester 9 (0.5 mmol) was added and the resulting
solution was cannulated into a stainless steel autoclave, which had
been repeatedly evacuated and refilled with argon. The flask and
cannula were rinsed with a further 2 ml of toluene. The autoclave
was heated to the appropriate temperature, then pressurized suc-
cessively with carbon monoxide (10 bar) and hydrogen (10 bar),
and the reaction mixture was stirred under these conditions until
TLC indicated complete consumption of the starting material. The
2.0 (m, 1 H), 2.17 (ddd, ϭ 16.1, 7.4, 2.1 Hz, 1 H, H at C4), 2.26
(ddd, ϭ 16.0, 6.2, 2.0 Hz, 1 H, H at C4), 5.12 (m, 1 H, H at C1),
6.89 (m, 1 H, ArH), 7.24Ϫ7.5 (m, 12 H, ArH), 8.08 (m, 1 H, ArH),
9.66 (pseudo t, ϭ 2.34 Hz, 1 H, CHO-
13C NMR (75.469 MHz, CDCl3): δ ϭ 14.18, 19.65, 22.6, 24.89,
25.4, 29.22, 31.76, 35.09, 41.55, 51.38, 77.32, 128.32, 128.55 (d,
ϭ 7.2 Hz, 2 C), 128.59 (d, ϭ 7.5 Hz, 2 C), 128.72 (2 C),
), [9.53 (m, CHO- )].
Ϫ
C,P
C,P
130.87 (d,
ϭ 2.5 Hz), 132.0, 133.95 (d,
ϭ 20.6 Hz, 2 C),
C,P
C,P
autoclave was then cooled rapidly to 20°C and the contents were 134.15 (d, C,P ϭ 20.8 Hz, 2 C), 134.44, 134.79 (d, C,P ϭ 19.0 Hz),
filtered through a small plug of silica with 30 ml of
-butyl
138.33 (d,
ϭ 12.5 Hz), 138.45 (d,
ϭ 11.5 Hz), 140.37 (d,
), [202.5 (CHO- )]. Ϫ
C,P
C,P
methyl ether. After evaporation of the solvent in vacuo, the crude
product was analyzed by NMR to determine the degree of conver-
sion and the diastereomer ratio. Subsequent flash chromatography
with petroleum ether/ -butyl methyl ether (9:1) furnished the al-
dehydes 10 as highly viscous oils.
ϭ 27.9 Hz), 166.74, 202.42 (CHO-
C,P
31P NMR (161.978 MHz, CDCl3): δ ϭ Ϫ3.9. Ϫ C31H37O3P (488.6):
calcd. C 76.21, H 7.63; found C 75.91, H 7.75.
±
[(±)-
-10d was obtained. Reaction temperature
30°C. Reaction time 120 h. Diastereomer ratio 90:10 ( ). Ϫ
-10a was obtained. Reaction temperature 1H NMR (300 MHz, CDCl3): δ ϭ 1.05 (d, ϭ 6.3 Hz, 3 H, CH3),
-10d]: From 225 mg (0.5 mmol) 9d,
±
173 mg (72%) of
[(±)-
-10a]: From 208 mg (0.5 mmol) 9a,
208 mg (93%) of
50°C. Reaction time 72 h. Diastereomer ratio 91:9 (
NMR (300 MHz, CDCl3): δ ϭ 0.78 [d, ϭ 6.8 Hz, 6 H, 14.0, 9.5, 4.5 Hz, 1 H, H at C2), 2.2Ϫ2.4 (m, 3 H), 6.16 (dd,
). Ϫ 1H 1.67 (ddd, ϭ 14.1, 8.5, 4.6 Hz, 1 H, H at C2), 2.11 (ddd,
ϭ
ϭ
CH(C 3)2], 0.8 (d, ϭ 6.5 Hz, 3 H, CH3), 1.24 (ddd, ϭ 14.0, 9.5, 4.5 Hz, 1 H, H at C1), 7.07 (m, 1 H, ArH), 7.22Ϫ7.5 (m, 17
10.1, 2.6 Hz, 1 H, H at C2), 1.51 (ddd, ϭ 14.0, 10.2, 3.7 Hz, 1 H, ArH), 8.25 (m, 1 H, ArH), 9.74 (pseudo t, ϭ 1.9 Hz, 1 H,
H, H at C2), 1.72 [m, 1 H, C (CH3)2], 1.88 (m, 1 H, H at C3), CHO-
), [9.62 (dd, ϭ 2.6, 1.5 Hz, CHO- )]. Ϫ 13C NMR
2.16 (ddd, ϭ 16.3, 7.3, 2.4 Hz, 1 H, H at C4), 2.18 (ddd, (75.469 MHz, CDCl3): δ ϭ 19.84, 25.21, 43.77, 51.16, 75.04,
125.58, 126.82 (2 C), 128.18 (2 C), 128.48, 128.77 (d, ϭ 7.2
Hz, 2 C), 128.86 (d, ϭ 7.2 Hz, 2 C), 129.27 (2 C), 131.14 (d,
ϭ
16.3, 6.4, 2.3 Hz, 1 H, H at C4), 4.96 (ddd, ϭ 10.1, 5.5, 2.4 Hz,
1 H, H at C1), 6.87 (m, 1 H, ArH), 7.2Ϫ7.34 (m, 12 H, ArH), 8.07
C,P
C,P
(m, 1 H, ArH), 9.59 (pseudo t, ϭ 2.2 Hz, 1 H, CHO-
(dd,
ϭ 3.1, 1.1 Hz, CHO- )]. Ϫ 13C NMR (75.469 MHz,
CDCl3): δ ϭ 17.8, 18.29, 19.32, 24.71, 32.09, 38.11, 51.31, 76.84,
128.14, 128.35 (d, ϭ 7.2 Hz, 2 C), 128.39 (d,
ϭ 7.4 Hz, 2 28.9 Hz), 140.7, 166.43, 202.09. Ϫ 31P NMR (161.978 MHz,
), [9.42
ϭ 2.6 Hz), 132.25, 134.12 (d,
ϭ 20.5 Hz, 2 C), 134.32 (d,
C,P
C,P
ϭ 20.8 Hz, 2 C), 134.69, 134.82 (d,
ϭ 18.9 Hz), 138.14
C,P
C,P
(d,
ϭ 12.4 Hz), 138.3 (d, ϭ 11.5 Hz), 140.51 (d,
ϭ
C,P
C,P
C,P
C,P
C,P
C), 128.54 (2 C), 130.61 (d,
ϭ 2.3 Hz), 131.84, 133.74 (d, CDCl3): δ ϭ Ϫ4.3. Ϫ C31H29O3P (480.5): calcd. C 77.48, H 6.08;
C,P
C,P ϭ 21.1 Hz, 2 C), 134.02 (d, C,P ϭ 21.5 Hz, 2 C), 134.19 (signal found C 77.16, H 6.18.
for C1Ј expected as a doublet at ca. 134.1 is obscured by the signals
±
at 134.02 and 134.19), 138.04 (d, C,P ϭ 12.8 Hz), 138.14 (d, C,P ϭ
[(±)-
-10e]: From 240 mg (0.5
11.2 Hz), 140.55 (d,
202.29 (CHO-
MHz, CDCl3): δ ϭ Ϫ3.9. Ϫ C28H31O3P (446.5): calcd. C 75.32, H
7.00; found C 74.99, H 7.21.
ϭ 27.7 Hz), 166.5 (d,
ϭ 2.6 Hz),
C,P
C,P
mmol) 9e, 199 mg (78%) of
-10e was obtained. Reaction tem-
perature 30°C. Reaction time 240 h. Diastereomer ratio 90:10 (
), [202.52 (CHO- )]. Ϫ 31P NMR (161.978
1
). Ϫ H NMR (300 MHz, CDCl3): δ ϭ 0.94 (d, ϭ 6.5 Hz, 3
H, CH3), 1.58 (ddd, ϭ 14.0, 8.4, 4.4 Hz, 1 H, H at C2), 1.84
(ddd, ϭ 14.0, 8.8, 5.1 Hz, 1 H, H at C2), 2.0Ϫ2.22 (m, 2 H), 2.38
(m, 1 H), 3.74 (s, 3 H, OCH3), 6.45 (dd, ϭ 8.7, 4.4 Hz, 1 H,
±
[(±)-
-9b]: From 228 mg (0.5 mmol) 9b, 219
mg (90%) of
Reaction time 72 h. Diastereomer ratio 91:9 (
-10b was obtained. Reaction temperature 50°C. HCO), 6.79 (d, ϭ 7.9 Hz, 1 H, ArH), 6.84 (t, ϭ 7.7 Hz, 1 H,
). Ϫ 1H NMR ArH), 6.94 (m, 1 H, ArH), 7.1Ϫ7.43 (m, 14 H, ArH), 8.15 (m, 1
1998, 1123Ϫ1134
1131

B. Breit
FULL PAPER
H, ArH), 9.64 (pseudo t, ϭ 2.1 Hz, 1 H, CHO-
), [9.54 (dd, 14.0, 9.5, 2.7 Hz, 1 H, H at C5), 1.42 (ddd, ϭ 14.0, 10.0, 4.1 Hz,
ϭ 3.0, 1.5 Hz, CHO- )]. Ϫ 13C NMR (75.469 MHz, CDCl3): 1 H, H at C5), 1.46Ϫ1.8 (m, 5 H, H at C2, C3, C7 and OH), 3.31
δ ϭ 19.71, 25.13, 42.98, 51.0, 55.48, 69.71, 110.53, 120.74, 126.33,
(s, 6 H, 2 OCH3), 3.44 (m, 1 H, H at C6), 4.48 [pseudo t, ϭ 6.0
128.41, 128.54 (d,
ϭ 7.1 Hz, 4 C), 128.6 (2 C), 128.64, 129.42, Hz, 1 H, C (OCH3)2]. Ϫ 13C NMR (75.469 MHz, CDCl3): δ ϭ
C,P
130.93 (d,
ϭ 2.8 Hz), 131.99, 133.91 (d,
ϭ 20.5 Hz, 2 C), 17.16, 18.59, 19.63, 26.05, 34.1, 40.26, 41.45, 52.37, 52.45, 74.25,
C,P
C,P
134.11 (d, C,P ϭ 20.7 Hz, 2 C), 134.54, 134.92 (d, C,P ϭ 19.6 Hz), 103.02. Ϫ C11H24O3 (204.3): calcd. C 64.67, H 11.84; found C
138.1 (d,
ϭ 12.8 Hz), 138.3 (d,
ϭ 11.4 Hz), 140.6 (d,
64.90, H 11.61.
C,P
C,P
ϭ 27.1 Hz), 155.94, 166.12 (d,
ϭ 2.2 Hz), 202.6 (CHO-
C,P
C,P
±
), [202.83 (CHO- )]. Ϫ 31P NMR (161.978 MHz, CDCl3):
(
-7): To a solution of 175 mg (0.86 mmol)
-12 in 2.5 ml of
δ ϭ Ϫ4.6. Ϫ C32H31O4P (510.6): calcd. C 75.28, H 6.12; found C
74.94, H 6.10.
dry acetonitrile and 50 µl water, 84 mg (0.9 mmol) of lithium tetra-
fluoroborate was added at room temp. After magnetic stirring for
18 h, the reaction mixture was diluted with 30 ml of diethyl ether,
washed with 10 ml of satd. aq. sodium bicarbonate solution, and
dried (Na2SO4) to give 162 mg of lactols 6 as a pale-yellow oil.
Lactols 6 were used directly in the subsequent oxidation step with-
out further purification.
±
[(±)-
-10f]: From 221 mg (0.5
mmol) 9f, 201 mg (85%) of
-10f was obtained. Reaction tem-
perature 30°C. Reaction time 168 h. Diastereomer ratio 90:10 (
1
). Ϫ H NMR (300 MHz, CDCl3): δ ϭ 0.9 (t, ϭ 7.3 Hz, 3
H, CH2C 3), 1.4 (m, 2 H), 1.7 (m, 1 H), 1.95 (m, 2 H), 2.16 (ddd,
ϭ 16.3, 8.2, 2.6 Hz, 1 H, H at C4), 2.34 (ddd, ϭ 16.2, 5.3, 3.4
Hz, 1 H, H at C4), 5.41 (m, 2 H, ϭCH and H at C1), 6.91 (m, 1
H, ArH), 7.24Ϫ7.43 (m, 12 H, ArH), 8.07 (m, 1 H, ArH), 9.67
To a magnetically stirred solution of lactols 6 in 5 ml CH2Cl2 at
room temp., 1.72 g (1.72 mmol) of PCC on Al2O3 (1 mmol/g) was
added and the suspension was stirred for 24 h. The reaction mix-
ture was then filtered through silica gel with a further 50 ml of
CH2Cl2, the solvent was evaporated, and the residue was purified
by flash chromatography with petroleum ether/ -butyl methyl
ether (4:1) to afford 131 mg (98% both steps) of the lactone -7
as a colorless oil. Spectroscopic and analytical data matched those
reported previously.[10]
(pseudo t, ϭ 2.3 Hz, 1 H, CHO-
CHO- )]. Ϫ 13C NMR (75.469 MHz, CDCl3): δ ϭ 11.81, 13.76,
19.76, 20.82, 24.9, 39.74, 50.8, 78.16, 128.21, 128.43 (d, ϭ 7.1
), [9.6 (dd, ϭ 2.6, 1.7 Hz,
C,P
Hz, 4 C), 128.54 (2 C), 130.65 (d,
ϭ 2.6 Hz), 130.9, 131.8,
C,P
132.14, 133.87 (d,
ϭ 20.5 Hz, 2 C), 134.03 (d,
ϭ 20.8 Hz,
C,P
C,P
2 C), 134.3, 134.7 (d,
ϭ 19.0 Hz), 138.08 (d,
ϭ 12.8 Hz),
C,P
C,P
138.2 (d,
Ϫ
ϭ 11.8 Hz), 140.4 (d,
ϭ 27.3 Hz), 166.09, 202.3.
C,P
C,P
±
31P NMR (161.978 MHz, CDCl3): δ ϭ Ϫ4.2. Ϫ C30H33O3P
[(±)-13]: According to the General Procedure described for
the preparation of the homomethallylic -DPPB esters 9, reaction
of 256 mg (2 mmol) of homomethallylic alcohol (±)-5a, 577 mg (2
mmol) 2-diphenylmethylbenzoic acid[29], 24 mg (0.2 mmol) DMAP
and 433 mg (2.1 mmol) DCC in 10 ml CH2Cl2 afforded 476 mg
(472.6): calcd. C 76.41, H 6.84; found C 76.26, H 6.90.
-10a. Ϫ
±
(
-11): To a solution of 515 mg (1.15 mmol) of
1
(50%) of (±)-13 as a colorless, viscous oil. Ϫ H NMR (300 MHz,
10a in 5 ml dry methanol were added 150 mg of Bayer Lewatit
SPC 108 and 200 mg of MgSO4. After magnetic stirring at room
temp. for 24 h, the reaction mixture was filtered through basic alu-
mina. Evaporation of the solvent afforded 537 mg (95%) of
11 as a colorless, viscous oil. Diastereomer ratio 91:9 (
CDCl3): δ ϭ 0.77 [d, ϭ 6.8 Hz, 6 H, CH(C 3)2], 1.68 (s, 3 H,
CH3), 1.70 [m, 1 H, C (CH3)2], 2.19 (m, 2 H, CH2), 4.66 (s, 2
H, ϭCH2), 5.05 (d pseudo t, ϭ 7.8, 5.3 Hz, 1 H, HCO), 6.64 (s,
1 H, C Ph2), 6.99Ϫ7.38 (m, 13 H, ArH), 7.8 (dd, ϭ 7.6, 1.2 Hz,
1 H, ArH). Ϫ 13C NMR (75.469 MHz, CDCl3): δ ϭ 17.30, 18.55,
22.21, 31.43, 39.82, 76.27, 113.09, 126.06, 126.13, 128.10, 129.67,
130.08, 130.89, 131.02, 141.9, 143.87, 143.91, 144.37, 167.34. Ϫ
C28H30O2 (398.5): calcd. C 84.38, H 7.59; found C 84.44, H 7.46.
-
). Ϫ
1H NMR (300 MHz, CDCl3): δ ϭ 0.74 (d, ϭ 6.2 Hz, 9 H, 3
CH3), 1.19Ϫ1.63 [m, 6 H, H at C2, C3, C4 and C (CH3)2], 3.2 (s,
6 H, OCH3), 4.35 (pseudo t, ϭ 5.5 Hz, 1 H, H at C5), 4.92 (m,
1 H, H at C1), 6.83 (m, 1 H, ArH), 7.2Ϫ7.32 (m, 12 H, ArH), 8.0
(m, 1 H, ArH). Ϫ 13C NMR (75.469 MHz, CDCl3): δ ϭ 17.93,
18.54, 19.57, 25.87, 32.08, 38.43, 40.36, 52.48, 52.68, 77.22, 103.24,
±
[(±)- -14]
±
128.15, 128.48 (d,
ϭ 7.3 Hz, 4 C), 128.6 (2 C), 130.57 (d,
C,P
[(±)-
-14]: Ac-
ϭ 2.1 Hz), 131.79, 133.94 (d,
ϭ 20.7 Hz, 2 C), 134.15 (d,
C,P
C,P
cording to the General Procedure described for the hydrofor-
mylation of -DPPB esters 9, reaction of 199 mg (0.5 mmol) 13
ϭ 20.8 Hz, 2 C), 134.29, 134.75 (d,
ϭ 18.3 Hz), 138.34
C,P
C,P
(d,
ϭ 11.8 Hz), 138.4 (d,
ϭ 12.7 Hz), 140.83 (d,
ϭ
C,P
C,P
C,P
afforded 118 mg (55%) of
-14 and
-14. Diastereomer ratio
27.7 Hz), 166.51. Ϫ 31P NMR (161.978 MHz, CDCl3): δ ϭ Ϫ4.2.
71:29 ( ). Spectroscopic data of the mixture of diastereomers
are given, with signals of the minor diastereomer in []. Ϫ 1H NMR
(300 MHz, CDCl3): δ ϭ 0.87 (m, 9 H, CH3), 1.5Ϫ2.5 (m, 6 H, CH,
CH2), 4.98 (m, 1 H, OCH), 6.98Ϫ7.4 (m, 13 H, ArH), 7.88 (m, 1
H, ArH), 9.53 (m, 1 H, CHO), 6.78, 6.82 (each s, together 1 H,
CHPh2). Ϫ 13C NMR (75.469 MHz, CDCl3): δ ϭ 17.4 [17.73],
18.59 [18.44], 20.73 [19.59], 24.99 [24.72], 31.82 [32.24], 37.98
[38.16], 51.93 [51.87], 76.44 [76.16], 120.13 [120.20], 126.26, 128.3,
129.8, 130.3, 132.0, 132.5, 143.93, 144.0, 144.75, 167.5, 202.34. Ϫ
C29H32O3 (428.6): calcd. C 81.27, H 7.59; found C 80.81, H 7.57.
Ϫ C30H37O4P (492.6).
±
(
-12): To a solution of 401 mg (0.81 mmol)
-11 in 5 ml
of ethanol, 3 ml of a saturated ethanolic potassium hydroxide solu-
tion was added at room temp. The resulting mixture was heated to
reflux for 3 h, cooled to room temp., and concentrated. The residue
was transferred to a separatory funnel with 50 ml of water and
extracted with three 30 ml portions of
-butyl methyl ether. The
combined organic phases were dried (Na2SO4), the solvent was eva-
porated, and the residue was purified by flash chromatography with
petroleum ether/ -butyl methyl ether (4:1) to furnish 160 mg
±
[(±)- -11b]
±
(97%) of
Ϫ
-12 as a yellow oil. Diastereomer ratio 91:9 (
).
1H NMR (300 MHz, CDCl3): δ ϭ 0.9 [d, ϭ 6.8 Hz, 6 H,
[(±)-
-11b]: To a solution of 200 mg (0.467 mmol) of 14 in 2 ml
CH(C 3)2], 0.95 [d, ϭ 6.6 Hz, 3 H, CH(C 3)2], 1.22 (ddd,
ϭ
of dry methanol, 50 mg of Bayer Lewatit SPC 108 and 100 mg
1132
1998, 1123Ϫ1134

Substrate-Directed Diastereoselective Hydroformylations, 2
FULL PAPER
of MgSO4 were added at room temp. After magnetic stirring at
then 10 ml of water was added. The organic phase was separated
room temp. for 24 h, the reaction mixture was filtered through basic and the aqueous phase was extracted with two 20 ml portions of
alumina. Evaporation of the solvent afforded 220 mg (99%) of 11b
as a colorless, viscous oil. Diastereomer ratio 71:29 ( ). Ϫ
CH2Cl2. The combined organic phases were dried (Na2SO4), the
solvent was removed in vacuo, and the residue was purified by fil-
1H NMR (300 MHz, CDCl3): δ ϭ 0.8Ϫ0.9 (m, 9 H, CH3), tration through a plug of silica (5 ϫ 3 cm) with 200 ml petroleum
1.35Ϫ1.8 (m, 6 H, CH, CH2), 3.25, 3.26, 3.30, 3.32 (each s, together ether/ -butyl methyl ether (9:1). Flash chromatography of the re-
6 H, OCH3), 4.45 (m, 1 H, OCH), 5.06 [m, 1 H, C(OMe)2], 6.77, sulting crude product with petroleum ether/ -butyl methyl ether
6.82 (each s, together 1 H, C Ph2), 7.07Ϫ7.31 (m, 13 H, ArH), 7.9
(19:1) furnished 350 mg (28%) of 19 as a colorless oil. Diastereomer
(m, 1 H, ArH). Ϫ 13C NMR (75.469 MHz, CDCl3): δ ϭ 17.0 ratio according to NMR analysis Ն 95:5 (
). Ϫ 1H NMR
[17.26], 18.39 [18.23], 20.44 [19.51], 26.14 [25.74], 31.25 [31.89],
38.12 [38.22], 51.82 [51.75], 52.57 [52.27], 76.79 [76.26], 102.79
[103.0], 126.05, 128.09, 129.63, 130.0, 130.9, 131.03, 143.84, 143.9,
(300 MHz, CDCl3): δ ϭ 0.84 (d, ϭ 6.8 Hz, 3 H, CH3), 0.88 (d,
ϭ 6.6 Hz, 3 H, CH3), 0.96 (d, ϭ 6.4 Hz, 3 H, CH3), 1.63 [m,
2 H, C (CH3)2 and OH], 1.64 (s, 3 H, CH3), 2.24 (m, 1 H, H at
144.39, 144.56 [144.64], 167.31 [167.39]. Ϫ C31H38O4 (474.6): calcd. C3), 3.15 (m, 1 H, H at C4), 4.7 (s, 1 H, ϭCH2), 4.77 (m, 1 H, ϭ
C 78.45, H 8.06; found C 78.40, H 7.98.
CH2). Ϫ 13C NMR (75.469 MHz, CDCl3): δ ϭ 12.96, 16.59, 19.41,
20.63, 30.04, 43.4, 76.93, 111.06, 148.17. Ϫ C9H18O (142.2). Ana-
lytical data were identical to those reported previously.[30]
±
[(±)- -12]
±
[(±)-
-12]: To a solution of 190 mg
±
(0.81 mmol) 11b in 2 ml of ethanol, 2 ml of a saturated ethanolic
sodium hydroxide solution was added at room temp. The resulting
mixture was heated to reflux for 3 h, cooled to room temp., and
concentrated. The residue was transferred to a separatory funnel
(15): According to the General Procedure for the
preparation of the homomethallylic -DPPB esters 9 (exception: 1
equiv. of DMAP was used), the reaction of 318 mg (2.236 mmol)
of homomethallylic alcohol 17 afforded 670 mg (70%) 15 as a col-
with 20 ml of water and was extracted with three 30 ml portions orless, viscous oil. Ϫ 1H NMR (300 MHz, CDCl3): δ ϭ 0.76 (d,
of
-butyl methyl ether. The combined organic phases were dried
ϭ 6.8 Hz, 3 H, CH3), 0.83 (d, ϭ 7.0 Hz, 3 H, CH3), 0.88 (d,
ϭ 7.1 Hz, 3 H, CH3), 1.51 (s, 3 H, ϭCϪCH3), 1.88 [m, 1 H,
(CH3)2], 2.5 (pseudo t, ϭ 7.1 Hz, 1 H, H at C3), 4.55 (s, 1
(Na2SO4), the solvent was evaporated, and the residue was purified
by flash chromatography with petroleum ether/ -butyl methyl
C
ether (4:1) to furnish 80 mg (98%) of 12 as a 71:29 (
:
) mix-
H, ϭCH2), 4.64 (s, 1 H, ϭCH2), 4.89 (dd, ϭ 8.1, 4.8 Hz, 1 H,
ture of diastereomers in the form of a yellow oil. Spectroscopic OCH), 6.86 (m, 1 H, ArH), 7.2Ϫ7.34 (m, 12 H, ArH), 8.04 (m, 1
1
data of
0.8Ϫ0.96 (m, 9 H, 3 CH3), 1.24Ϫ1.8 (m, 6 H, CH, CH2), 2.0 (br 19.0, 19.97, 29.38, 43.46, 79.86, 112.66, 128.04, 128.4 (d, C,P ϭ 6.8
m, 1 H, OH), 3.25 (s, 3 H, OCH3), 3.26 (s, 3 H, OCH3), 3.4 (d Hz, 2 C), 128.46 (d, ϭ 6.9 Hz, 2 C), 128.55 (2 C), 130.5 (d,
pseudo t, ϭ 8.2, 4.8 Hz, 1 H, HCO), 4.42 [m, 1 H, C(OMe)2]. ϭ 1.7 Hz), 131.75, 133.96 (d,
13C NMR (75.469 MHz, CDCl3): δ ϭ 16.87, 18.72, 21.18, 25.96, 134.25 (d,
-9 are as follows. Ϫ H NMR (300 MHz, CDCl3): δ ϭ H, ArH). Ϫ 13C NMR (75.469 MHz, CDCl3): δ ϭ 16.15, 16.33,
C,P
ϭ 21.1 Hz, 2 C), 134.17,
C,P
C,P
Ϫ
ϭ 21.7 Hz, 2 C) (signal for C1Ј expected at ca. 134
C,P
33.60, 38.60, 41.59, 52.47, 52.58, 73.77, 103.36. Ϫ Analytical data is obscured by the signals at 134.17 and 134.25), 138.4 (d,
ϭ
C,P
were identical to those of
-12.
11.5 Hz), 138.47 (d,
146.84, 165.87 (d,
CDCl3): δ ϭ Ϫ4.1. Ϫ C28H31O2P (430.5): calcd. C 78.12, H 7.26;
ϭ 13.1 Hz), 141.57 (d,
ϭ 28.2 Hz),
C,P
C,P
ϭ 3.0 Hz). Ϫ 31P NMR (161.978 MHz,
C,P
(±)
(17)[20]: To a sus-
pension of 1.0 g (4.03 mmol) [Cp2TiCl2] in 20 ml of THF, 4.48 ml
(8.06 mmol) of a 1.8  solution of -PrMgBr in Et2O and 0.5 ml
(5 mmol) isoprene were simultaneously added dropwise by means
of syringes. After stirring the resulting mixture for 30 min. at room
temp., 361 mg (5 mmol) of isobutyraldehyde was added dropwise.
The reaction mixture was allowed to stir for a further 60 min. at
found C 78.25, H 7.42.
±
(16): According to the General Procedure for the
preparation of the homomethallylic -DPPB esters 9 (exception: 1
equiv. of DMAP was used), the reaction of 495 mg (3.48 mmol) of
room temp. and then 20 ml of satd. aq. NaHCO3 solution was homomethallylic alcohol 19 afforded 703 mg (47%) of 16 as a col-
added. The organic phase was separated and the aqueous phase
was extracted with two 20 ml portions of -butyl methyl ether.
The combined organic phases were dried (Na2SO4), and filtered
through a plug of Celite with a further 50 ml of -butyl methyl
orless, viscous oil. Ϫ 1H NMR (300 MHz, CDCl3): δ ϭ 0.77 (d,
ϭ 6.9 Hz, 3 H, CH3), 0.84 (d, ϭ 6.7 Hz, 3 H, CH3), 0.91 (d,
ϭ 6.9 Hz, 3 H, CH3), 1.82 [m, 1 H, C (CH3)2], 2.47 (m, 1 H,
C
CH3), 4.66 (m, 1 H, ϭCH2), 4.71 (s, 1 H, ϭCH2), 4.99 (dd,
ϭ
ether. The solvent was evaporated and the residue was subjected to 8.2, 4.3 Hz, 1 H, H at C1), 6.9 (m, 1 H, ArH), 7.23Ϫ7.36 (m, 12
flash chromatography with petroleum ether/ -butyl methyl ether
(9:1) to furnish 238 mg (42%) of 17 as a colorless oil. Diastereomer
H, ArH), 8.1 (m, 1 H, ArH). Ϫ 13C NMR (75.469 MHz, CDCl3):
δ ϭ 15.45, 16.39, 19.7, 19.9, 29.71, 43.4, 80.11, 111.94, 128.08,
ratio according to NMR analysis Ն 95:5 (
(200 MHz, CDCl3): δ ϭ 0.8 (d, ϭ 6.8 Hz, 3 H, CH3), 0.9 (d,
7.0 Hz, 3 H, CH3), 0.97 (d, ϭ 7.0 Hz, 3 H, CH3), 1.5 (d, ϭ 2.6 ca. 134 is obscured by the signals at 134.07 and 134.26), 138.26 (d,
Hz, 1 H, OH), 1.64 (s, 3 H, CH3), 1.7 [m, 1 H, C (CH3)2], 2.2 (m, C,P ϭ 12.1 Hz, 2 C), 141.26 (d, C,P ϭ 28.2 Hz), 147.05, 166.31. Ϫ
). Ϫ 1H NMR 128.42 (d, C,P ϭ 7.4 Hz, 4 C), 128.47 (2 C), 130.53, 131.79, 134.04,
ϭ
134.07 (d,
ϭ 20.9 Hz, 4 C), 134.26 (signal for C1Ј expected at
C,P
1 H, H at C3), 3.2 (m, 1 H, H at C4), 4.77 (s, 1 H, ϭCH2), 4.88 31P NMR (161.978 MHz, CDCl3): δ ϭ Ϫ4.1. Ϫ C28H31O2P (430.5):
(m, 1 H, ϭCH2). Ϫ 13C NMR (75.469 MHz, CDCl3): δ ϭ 14.0,
15.6, 18.4, 20.7, 29.0, 45.4, 76.2, 112.9, 148.0. Ϫ C9H18O (142.2).
Analytical data were identical to those reported previously.[30]
calcd. C 78.12, H 7.26; found C 77.86, H 7.10.
η
±
±
(19): To a solution
(20): To a solution of 47 mg (0.183 mmol) of [Rh(CO)2acac] in 1
of 634 mg (8.79 mmol) isobutyraldehyde in 15 ml of CH2Cl2 at ml C6D6, 79 mg (0.183 mmol) of 15 was added at room temp. Gas
Ϫ78°C, was added 2.49 g (17.58 mmol) of borontrifluorideϪether
evolution was observed and the solution turned intensely yellow.
followed by 3.16 g of the methallylstannane 18.[21] The reaction After stirring for 1 h at room temp., completion of the reaction
mixture was allowed to warm to Ϫ10°C over a period of 2 h and was checked by 31P-NMR spectroscopy. The solvent was then eva-
1998, 1123Ϫ1134
1133

B. Breit
1993,
FULL PAPER
[2]
[3]
A. H. Hoveyda, D. A. Evans, G. C. Fu,
,
porated in vacuo to furnish 120 mg (99%) of the rhodium complex
1307Ϫ1370.
1
20 as a yellow glass. Ϫ H NMR (300 MHz, CDCl3): δ ϭ 0.64 (d,
For successful exceptions see: [3a] A. H. Hoveyda, Z. Xu,
[3b]
ϭ 6.8 Hz, 3 H, CH3), 0.76 (d, ϭ 6.7 Hz, 3 H, CH3), 0.81 (d,
ϭ 7.0 Hz, 3 H, CH3), 1.34 (s, 3 H, CH3), 1.43 (s, 3 H, CH3), 1.77
(m, 1 H, H at C5), 1.97 (s, 3 H, CH3), 2.47 (m, 1 H, H at C3), 4.55
(s, 1 H, ϭCH2), 4.61 (s, 1 H, ϭCH2), 4.82 (dd, ϭ 7.9, 5.0 Hz, 1
H, OCH), 5.26 (s, 1 H, CHacac), 6.89 (ddd, ϭ 11.7, 7.7, 1 Hz, 1
H, ArH), 7.22Ϫ7.8 (m, 12 H, ArH), 8.18 (m, 1 H, ArH). Ϫ 13C
NMR (75.469 MHz, CDCl3): δ ϭ 16.08, 16.41, 18.87, 19.6, 26.33,
27.43, 29.24, 43.28, 79.6, 100.15, 112.39, 127.8Ϫ136.0 (all aryl-C),
1991,
, 5079Ϫ5080. Ϫ
A. H. Hoveyda, Z.
Xu, J.-P. Morken, A. F. Houri,
1991,
, 8950Ϫ8952. Ϫ
[3c]
A. F. Houri, M. T. Didiuk, Z. Xu, N. R. Horan, A. H.
Hoveyda,
1993,
, 6614Ϫ6624. See also refs.[4][5][7]. For
substrate directed transition metal catalysed reduction processes
see: J. M. Brown,
1987, , 169Ϫ182;
1987, , 190Ϫ203; D. A. Evans, G. C. Fu,
A. H. Hoveyda,
1992,
, 6671Ϫ6679; K.
Burgess, M. J. Ohlmeyer,
1991, , 1179Ϫ1191.
[4] [4a]
B. Breit,
1996,
, 3021Ϫ3023;
146.82, 163.9, 184.64, 187.03, 189.56 (dd,
ϭ 76.1 Hz,
ϭ
[4b]
C,Rh
C,P
1996, , 2835Ϫ2837. Ϫ
B. Breit,
25.4 Hz, RhCO). Ϫ 31P NMR (161.978 MHz, CDCl3): δ ϭ 54.8
[4c]
1997, 1841Ϫ1851. Ϫ
B. Breit in
(Eds.: G. Helmchen, J. Dibo, D. Flubacher,
(d,
ϭ 189.1 Hz). Ϫ C34H38O5PRh (660.6).
P,Rh
B. Wiese), Vieweg, Braunschweig, 1997, 139Ϫ146.
For a review on stereoselective hydroformylations, see:
Eilbracht in
[5]
[5a]
P.
±
(21): According to the General Pro-
(Eds.: G. Helmchen, R. W. Hoffmann,
J. Mulzer, E. Schaumann), Thieme, Stuttgart, 1995, p.
2488Ϫ2557. Ϫ For substrate-directed diastereoselective hydro-
cedure for the hydroformylation of the homomethallylic -DPPB
esters 9, the reaction of 215 mg (0.5 mmol) of 15 afforded 210
mg (91%) of 21 as a colorless, viscous oil. Diastereomer ratio 96:4
formylation of cyclic systems, see: Ϫ [5b] S. D. Burke, J. E. Cobb,
[5c]
1986, , 4237Ϫ4240. Ϫ
W. R. Jackson,
(
). Relative configuration of 21 was assigned on the basis
P. Perlmutter, E. E. Tasdelen,
of characteristic carbon-13 NMR chemical shifts (see Table 3). Ϫ
1H NMR (300 MHz, CDCl3): δ ϭ 0.62 (d, ϭ 6.6 Hz, 3 H, CH3),
0.65 (d, ϭ 7.0 Hz, 3 H, CH3), 0.74 (d, ϭ 6.8 Hz, 3 H, CH3),
0.78 (d, ϭ 6.7 Hz, 3 H, CH3), 1.67 (m, 2 H), 1.86 (m, 1 H), 2.06
(ddd, ϭ 13.0, 6.2, 2.2 Hz, 1 H, H at C4), 2.17 (ddd, ϭ 12.7,
7.0, 2.4 Hz, 1 H, H at C4), 4.89 (dd, ϭ 9.5, 3.3 Hz, 1 H, H at
C1), 6.85 (m, 1 H, ArH), 7.18Ϫ7.5 (m, 12 H, ArH), 8.07 (m, 1 H,
ArH), 9.56 (pseudo t, ϭ 2.3 Hz, 1 H, CHO). Ϫ 13C NMR (75.469
MHz, CDCl3): δ ϭ 9.97, 13.84, 15.32, 20.08, 27.27, 29.29, 38.15,
49.77, 79.77, 128.27, 128.5 (d, C,P ϭ 7.3 Hz, 2 C), 128.53 (d, C,P ϭ
1990, 763Ϫ764.
See for example:
[6]
[6a]
: H. Tanaka, A. Kuroda, H. Maru-
sawa, H. Hatanaka, T. Kino, T. Goto, M. Hashimoto, T. Taga,
1987,
tanaka, M. Hashimoto, M. Nishiyama, T. Goto, M. Okuhara,
M. Kohsaka, H. Aoki, H. Imanaka, 1987, , 1249.
, 5031Ϫ5033; T. Kino, H. Ha-
[6b]
Ϫ
: K. Yamada, M. Ojika, T. Ishigaki, Y.
Yoshida, H. Ekimoto, M. Arakawa,
1993,
[6c]
, 11020Ϫ11021. Ϫ
: J. K. Cha, W. J. Christ, J.
M. Finan, H. Fujioka, Y. Kishi, L. L. Klein, S. S. Ko, J. Leder,
W. W. McWhorter, Jr., K.-P. Pfaff, M. Yonaga, D. Uemura, Y.
[6d]
Hirata,
1982,
, 7369Ϫ7371. Ϫ
7.1 Hz, 2 C), 128.68 (2 C) 130.7, 132.02, 133.9 (d,
ϭ 20.8 Hz,
: S. Omura, A. Nakagawa, N. Sadakane,
C,P
[6e]
1979, 4323Ϫ4326. Ϫ
: M. Sasaki,
2 C), 134.27 (d,
ϭ 21.1 Hz, 2 C), 134.36 (signal for C1Ј ex-
C,P
N. Matsumori, T. Maruyama, T. Nonomura, M. Murata, K.
pected as a doublet at ca. 134 is obscured by the signals at 134.27
Tachibana, T. Yasumoto,
1996,
, 1782Ϫ1785;
[6f]
and 134.36), 138.03 (d, C,P ϭ 12.0 Hz), 138.07 (d, C,P ϭ 11.0 Hz),
1996, , 1672. Ϫ
141.12 (d,
ϭ 28.0 Hz), 166.26, 202.84. Ϫ 31P NMR (161.978
: A. T. Bottini, D. G. Gilchrist,
1981,
,
C,P
[6g]
2719Ϫ2722. Ϫ
: S. C. Bezuidenhout, W. C. A.
MHz, CDCl3): δ ϭ Ϫ3.6. Ϫ C29H33O3P (460.6): calcd. C 75.63, H
7.22; found C 75.71, H 7.34.
Gelderblom, C. P. Gorst-Allman, R. M. Horak, W. F. O. Ma-
rasas, G. Spiteller, R. Vleggaar,
1988, 743Ϫ745.
±
[7]
Parts of this work have been published previously as a prelimi-
nary communication: B. Breit,
[(±)-
-22]
±
1997, 591Ϫ592.
[8]
[9]
T. Imai, S. Nishida,
1993, 395Ϫ399.
[(±)- -22]: According to the General Procedure for the
hydroformylation of the homomethallylic -DPPB esters 9, the re-
action of 215 mg (0.5 mmol) of 15 afforded 212 mg (92%) of 21 as
Employing
a catalyst system consisting of 0.35 mol%
[Rh(CO)2acac], 2.8 mol% P(OPh)3 under similar hydrofor-
mylation conditions (conversion of 5a after 24 h at 90°C: quan-
titative) gave a complicated mixture of different products,which
contained only minor amounts of the lactols 6.
B. Bardili, H. Marschall-Weyerstahl, P. Weyerstahl,
1985, 275Ϫ300.
a colorless, viscous oil. Diastereomer ratio 50:50 (
). Spec-
troscopic data of the 1:1 mixture of - and -22 are as follows.
[10]
[11]
1
Ϫ H NMR (300 MHz, CDCl3): δ ϭ 0.8 (m, 24 H, all CH3), 1.65
[m, 2 H, C (CH3)2], 1.8Ϫ2.2 (m, 6 H), 2.4 (m, 2 H, H at C4), 4.9
(m, 2 H, H at C1), 6.88 (m, 2 H, ArH), 7.2Ϫ7.37 (m, 24 H, ArH),
8.1 (m, 2 H, ArH), 9.49 (dd, ϭ 2.6, 1.5 Hz, 1 H, CHO), 9.61 (dd,
ϭ 2.8, 1.1 Hz, 1 H, CHO). Ϫ 13C NMR (75.469 MHz, CDCl3):
δ ϭ 10.39, 11.02, 15.92, 17.49, 17.92, 18.46, 19.33, 19.46, 29.39,
29.73, 30.05, 30.18, 38.15, 39.07, 47.36, 49.0, 79.85, 80.23,
128.1Ϫ141.7 (all aryl-C), 166.15, 166.26, 202.5, 202.56. Ϫ 31P
NMR (161.978 MHz, CDCl3): δ ϭ Ϫ3.8, Ϫ3.9. Ϫ C29H33O3P
(460.6): calcd. C 75.63, H 7.22; found C 75.43, H 7.16.
G. Höfle, W. Steglich, H. Vorbrüggen,
1978,
,
602Ϫ615;
1978, , 569Ϫ583.
[12]
[13]
Compare similar studies in ref.[4]
.
For determination of the relative configuration of the aldehydes
14, they were transformed in a similar manner as the -DPPB
derivative 10a via the corresponding dimethyl acetal 11b into
the alcohol 12 (
tal Section.
:
, 71:29). For details, see the Experimen-
[14]
For the combined use of vicinal coupling constants, 2D
NOESY and MM3 force field calculations to determine pre-
ferred conformations of acyclic molecules in solution, see: R.
Göttlich, C. Kahrs, J. Krüger, R. W. Hoffmann,
1997, 247Ϫ251; R. Göttlich, U. Schopfer, M.
Ƞ
Dedicated to Professor
65th birthday.
on the occasion of his
Stahl, R. W. Hoffmann,
1997, 1757Ϫ1764; R. J.
Smith, D. H. Williams, J. C. J. Barna, I. R. McDermott, K. D.
Haegele, F. Piriou, J. Wagner, W. Higgins,
[1]
See for example: [1a] M. Sakai, S. Mano, K. Nozaki, H. Takaya,
[1b]
1985,
, 2849Ϫ2857; D. Neuhaus, M. P. Williamson, “The
1993,
, 7033Ϫ7034. Ϫ
F. Agbossou,
Nuclear Overhauser Effect in Structural and Conformational
J.-F. Carpentier, A. Mortreux,
1995, , 2485Ϫ2506.
1996, , 355Ϫ364. Ϫ
[1c]
[1d]
Analysis”, VCH, New York 1989, pp 412;
, Chapter 11.
Ϫ
B. M. Trost,
[15]
M. Shibasaki, H. Sasai, T. Arai,
1290Ϫ1311;
1997,
,
On going from 30°C to 90°C, the vicinal coupling constants
1997, , 1236Ϫ1256.
for HaHc and HbHc change from 9.0/4.1 Hz to 8.4/4.6 Hz (400
1134
1998, 1123Ϫ1134

Substrate-Directed Diastereoselective Hydroformylations, 2
FULL PAPER
[22]
[23]
MHz, [D8]toluene, digital resolution: 0.125 Hz/pt). These data
indicate a change in conformer populations for 10a on going
to higher temperatures, which is reflected in the temperature
dependence of the diastereoselectivities of the hydrofor-
mylation reaction.
S. Berger, S. Braun, H.-O. Kalinowski,
, Bd. 3, Thieme, Stuttgart, 1993, Chapter 3.3.11.1,
p. 164Ϫ166.
Compare [Rh(acac)(PPh3)(CO)]: 13C NMR (CH2Cl2): δ ϭ
189.5 (1
ϭ 74.8 Hz, 1 C,P ϭ 24.8 Hz) ; see L. J. Todd, J. R.
1974, , 1-25; and L. S.
[16]
MACROMODEL: F. Mohadi, N. G. J. Richardson, W. C. Gu-
WilkinsoCn,,Rh
ida, R. Liskamp, M. Lipto, C. Caufield, G. Chang, T. Hen-
Bresler, N. A. Buzina, Y. S. Varshavsky, N. V. Kiseleva, T. G.
drickson, W. C. Still,
1990, , 440Ϫ467. A
Cherkasova,
1979,
, 229Ϫ235.
[24]
[25]
[26]
[27]
[28]
[29]
[30]
Boltzmann distribution over all low energy conformers (
<
B. Ragonnet et, M. Santelli, M. Bertrand,
rel
25 kJ·molϪ1 above the most stable conformer) resulted in the
conformer population reported, whereas different conformers
of the PPh2 unit and the isopropyl substituent with identical
conformation of the main chain have been combined additively.
The intramolecular -DPPB-directed hydroformylation has
been shown to be at least 15 times faster than a corresponding
intermolecular hydroformylation process (see ref.[4]).
1973, 3119Ϫ3121.
G. P. Boldrini, L. Lodi, E. Tagliavini, C. Tarasco, C. Trombini,
A. Umani-Ronchi
1987, , 5447Ϫ5452.
L. S. Hegedus, S. D. Wagner, E. L. Waterman, K. Siirala-
[17]
[18]
Hansen,
R. Dran, T. Prange
, 492Ϫ494.
1975, , 593Ϫ598.
1966,
R. W. Hoffmann,
1992,
1992, , 1124.
, 1147Ϫ1157;
J. E. Hoots, T. B. Rauchfuss, D. A. Wrobleski, H. C. Knachel,
1982, , 175Ϫ179.
[19]
[20]
R. W. Hoffmann,
1989, , 1841Ϫ1860.
W. Wykypiel, J.-J. Lohmann, D. Seebach,
1981, , 1337Ϫ1346.
J. Szymoniak, S. Pagneux, D. Felix, C. Moise,
1996,
46Ϫ48.
Y. Gao, H. Urabe, F. Sato,
1994, 59, 5521Ϫ5523.
[21]
S. Weigand, R. Brückner,
1996, 475Ϫ482.
[97394]
1998, 1123Ϫ1134
1135