New monodentate P,C-stereogenic bicyclic phosphanes: 1-Phenyl-1,3a,4,5,6, 6a-hexahydrocyclopenta[b]phosphole and 1-phenyl-octahydrocyclopenta[b]phosphole

Article abstract of DOI:10.1002/ejoc.200400186

Racemic 1-phenyl-1,3a,4,5,6,6a-hexahydrocyclopenta[b]phosphole (4) was separated into enantiomerically pure 1-phenyl-1,3a,4,5,6,6a- hexahydrocyclopenta[b]phosphole 1-oxides [(R_P)-6 and (S _P)-6] by an oxidative resolution procedure in

Full text of DOI:10.1002/ejoc.200400186

FULL PAPER
New Monodentate P,C-Stereogenic Bicyclic Phosphanes:
1-Phenyl-1,3a,4,5,6,6a-hexahydrocyclopenta[b]phosphole and 1-Phenyl-
octahydrocyclopenta[b]phosphole
Zbigniew Pakulski,[a] Oleg M. Demchuk,[a] Jadwiga Frelek,[a] Roman Luboradzki,[b] and
K. Michał Pietrusiewicz*[a,c]
Keywords: Asymmetric catalysis / Phosphabicyclo[3.3.0]octane / Phosphabicyclo[3.3.0]octene / Phospholene / P-stereogenic
Racemic 1-phenyl-1,3a,4,5,6,6a-hexahydrocyclopenta[b]-
phosphole (4) was separated into enantiomerically pure
1-phenyl-1,3a,4,5,6,6a-hexahydrocyclopenta[b]phosphole 1-
(RP)-6 gave enantiomerically pure (SP)-4. Hydrogenation of
(SP)-6 and subsequent reduction of P=O afforded saturated
1-phenyloctahydrocyclopenta[b]phosphole [(RP)-5]. Mono-
oxides [(RP)-6 and (SP)-6] by an oxidative resolution proced- phosphanes (SP)-4 and (RP)-5 were tested as chiral ligands
ure involving treatment of 4 with menthyl bromoacetate,
crystallization of the resulting diastereoisomeric phosphon-
ium bromides 8, and stereoselective hydrolysis of the diaster-
eomerically pure salts to the corresponding enantiomerically
pure phosphane oxides. Stereoretentive reduction of P=O in
and catalysts in model asymmetric hydrogenation and C−C
bond-forming reactions. Enantioselectivities of up to 95%
were observed.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
Introduction
Phosphanes are well known as most powerful and versa-
tile ligands for use in catalysis involving transition metals.[1]
Recently, monophosphanes with rigid bicyclic structures
containing the phosphorus atom embedded in a five-mem-
bered ring (e.g., 1؊3) have been identified as both highly
efficient chiral ligands and excellent chiral catalysts
(Figure 1).[2Ϫ4]
In this paper we wish to demonstrate that the bicyclic
P,C-stereogenic monophosphanes 1-phenyl-1,3a,4,5,6,6a-
hexahydrocyclopenta[b]phosphole [(SP)-4][5] and 1-phenyl-
octahydrocyclopenta[b]phosphole [(RP)-5] can be readily
obtained in the enantiomerically pure state by an efficient
oxidative resolution procedure utilizing menthyl bromo-
acetates as resolving agents. The use of enantiopure mono-
phosphanes (SP)-4 and (RP)-5 as ligands and catalysts in
model asymmetric CϪC bond-formation and hydrogen-
ation processes is also reported.
Figure 1. Selected chiral phosphanes
Results and Discussion
[a]
Institute of Organic Chemistry, Polish Academy of Sciences,
Kasprzaka 44/52, 01-224 Warszawa, Poland
Fax: (internat.) ϩ 48-22-632-6681
E-mail: kmp@icho.edu.pl
Racemic 1-phenyl-1,3a,4,5,6,6a-hexahydrocyclopenta[-
b]phosphole 1-oxide (6) was synthesized by a single-step
cyclopentannulation route[5,6] involving deprotonation of 1-
phenyl-2,5-dihydro-1H-phosphole 1-oxide (2.1 equiv. of
LDA, Ϫ78 °C) and treatment of the resulting anionic inter-
mediate with 1,3-diiodopropane. The annulation reaction
[b]
Institute of Physical Chemistry, Polish Academy of Sciences,
Kasprzaka 44/52, 01-224 Warszawa, Poland
[c]
Department of Organic Chemistry, Maria Curie-Skłodowska
University,
Gliniana 33, 20-614 Lublin, Poland
Eur. J. Org. Chem. 2004, 3913Ϫ3918
DOI: 10.1002/ejoc.200400186
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3913

Z. Pakulski, O. M. Demchuk, J. Frelek, R. Luboradzki, K. M. Pietrusiewicz
FULL PAPER
was found to be extremely difficult to scale up because of a enantiomerically enriched oxide 6 (40% ee). Reduction of
strong exothermic effect, which caused dramatic decreases this oxide with phenylsilane and repetition of the above res-
in the yields (74% on a 1 mmol scale, 47% on a 5 mmol olution procedure, but now with (ϩ)-menthyl bromoace-
scale, and only 15Ϫ20% on a 10 mmol scale). The anal- tate, gave (SP)-6 (37% based on starting racemic 6, Ͼ98%
ogous annulation with 1-phenyl-2,5-dihydro-1H-phosphole ee[10]). The configuration of the intermediate diastereomer-
1-sulfide proved to be less exothermic and afforded the ex- ically pure phosphonium bromide 8aЈ was established by X-
pected 1-phenyl-1,3a,4,5,6,6a-hexahydrocyclopenta[b]phos- ray crystallographic analysis (Figure 2).
phole 1-sulfide (7) in very high yield irrespective of the reac-
tion scale used (91% yield for a 10 mmol scale).[5,6] Re-
duction of 6 with phenylsilane[7] and radical-based re-
duction of 7 with tris(trimethylsilyl)silane[8] gave 1-phenyl-
1,3a,4,5,6,6a-hexahydrocyclopenta[b]phosphole (4) quanti-
tatively and with complete retention of configuration at P
(Scheme 1).[5,7,8]
Figure 2. ORTEP drawing of salt 8aЈ
Hydrogenation of (SP)-6 on 10% Pd / C afforded satu-
rated 1-phenyloctahydrocyclopenta[b]phosphole 1-oxide
[(SP)-9, 92%], which, after reduction with phenylsilane, cle-
anly gave (RP)-5 (98%, Scheme 3). As we have found be-
fore,[5] the reduction proceeded with complete retention.
Scheme 1. Synthesis of racemic 1-phenyl-1,3a,4,5,6,6a-hexahydro-
cyclopenta[b]phosphole (4)
Scheme 3. Synthesis of (RP)-5
Separation of 4 into its enantiomers was achieved by an
oxidative resolution route,[9] as shown in Scheme 2. Ad-
dition of an ethyl acetate solution of racemic 4 to a solution
The optical activities of the resolved compounds 6 and 9
of (Ϫ)-menthyl bromoacetate in ethyl acetate afforded crys- were examined by circular dichroism (CD) spectroscopy. As
talline phosphonium bromides in a 1:1 ratio. The crystals can be seen in Figure 3, the CD spectra of compounds
obtained were recrystallized twice from ethyl acetate/ben- (1R,3aS,6aS)-6 and (1S,3aR,6aR)-6 have a perfect mirror-
zene/ethyl alcohol (20:1:1) to give one of the two epimeric image relationship, as expected for enantiomers. In the case
phosphonium salts 8 in the form of a single diastereoisomer of compound (1S,3aR,6aR)-9, however, the expected CD
8a (29%). Hydrolysis of 8a yielded virtually enantiomer- bands could not be detected. This observation seems to
ically pure (RP)-6 (25% based on starting racemic 3, Ͼ98% indicate that the inclusion of the ring double bond in the
ee[10]). The collected filtrates enriched in the second P-epi- chromophoric system in the 2,3-dihydro-1H-phosphole sys-
mer of 8 were evaporated and hydrolyzed to furnish the tems is of key importance. In corroboration of this, a nearly
Scheme 2. Oxidative resolution of rac-4; (a) (Ϫ)-menthyl bromoacetate, (b) double crystallization, (c) 10% aq NaOH, CH2Cl2, (d) PhSiH3,
toluene, 100 °C, (e) (ϩ)-menthyl bromoacetate
3914
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 3913Ϫ3918

New Monodentate P,C-Stereogenic Bicyclic Phosphanes
FULL PAPER
ninefold decrease in the specific rotation value of the satu-
rated 9 relative to the unsaturated 6 was observed. Further
studies to explain this phenomenon are in progress in our
laboratory.
Table 2. Palladium-catalyzed allylic substitution
Entry Starting Solvent
materials
Ligand[a] Product Yield ee
(mol %)
(%) (%)[b]
1
2
3
15 ϩ 16 CH2Cl2/THF 4 (5)
17
17
17
96
83
96
95
69
56
15 ϩ 16 CH2Cl2
4 (5)
15 ϩ 16 CH2Cl2/THF 5 (5)
[a] [Pd(allyl)Cl]2 was used as catalyst precursor (0.5 mol %). [b] % ee
was determined by comparing the optical rotation with literature
value.[15]
Figure 3. CD spectra of (RP)-6 (dashed) and (SP)-6 (solid) in aceto-
nitrile solution
Table 3. CϪC bond formation (24 h, room temp., benzene)
The enantiopure P,C-stereogenic phosphanes (SP)-4 and
(RP)-5 were tested as chiral catalysts and ligands in a series
of model asymmetric reactions known to be efficiently cata-
lyzed by monophosphanes or their organometallic com-
plexes. Firstly, we tested the Pd-catalyzed allylic substi-
tution reaction[11] (Table 1). Treatment of (E)-1,3-di-
phenylallyl acetate (10) with dimethyl malonate (11) in the
presence of (SP)-4 or (RP)-5 and [Pd(allyl)Cl]2 as catalyst
precursor gave the allylated malonate 13[12] (20% ee with 4
and 31% ee with 5). The same procedure with benzylamine
(12) as nucleophile gave the allylamine derivative 14[13] in
86% yield, but with only 15% ee. However, an analogous
reaction with cyclopentenyl carbonate 15[14] as substrate
and phthalimide (16) as nucleophile (Table 2) afforded the
cyclopentenylamine derivative 17 in excellent chemical yield
(96%) and with very high enantioselectivity (95% ee).
The last result underlines the potential effectiveness of P-
stereogenic bicyclic monophosphane ligands in the asym-
metric allylic substitution of cyclic carbonates and consti-
tutes a good lead for further developments of ligands con-
taining 2-phosphabicyclo[3.3.0]octane skeletons.
Entry
Starting
materials
Product
Cat.
Yield
(%)
ee
(%)
(10 mol %)[a]
1
2
3
18 ϩ 19
18 ϩ 19
18 ϩ 20
21
21
21
4
5
4
63
51[b]
48[b]
29[b]
67
64[c]
[a]
The reaction was carried out under Ar with chiral phosphane
(10 mol %), 19 or 20 (100 mol %), 18 (100 mol %), NaOAc (50
mol %), and AcOH (50 mol %).[3c] [b] % ees were measured by GC
with γ-Dex 225 column.[3c] [c] 24 h, 80 °C.
We also tested phosphane-catalyzed CϪC bond forma-
tion in reactions between 2-butynoates and malonate-type
nucleophiles (Table 3).[3c,16] Activation of ethyl 2,3-butadi-
enoate (19) with chiral phosphane (SP)-4 or (RP)-5, fol-
lowed by addition of the active intermediate[16,17] to 2-
methoxycarbonyl cyclopentanone (18), gave the known op-
tically active adduct 21 in 48Ϫ51% ees depending on the
phosphane used (Table 3, Entries 1,2). Similar transform-
ations utilizing ethyl 2-butynoate (20) as substrate required
higher reaction temperatures and gave significantly lower
enantioselectivities (Table 3, Entry 3).
Table 1. Palladium-catalyzed allylic substitution
The catalytic activity of phosphanes (SP)-4 and (RP)-5
was also screened in [3ϩ2] cycloadditions between ethyl 2,3-
butadienoate (19) and selected olefins to afford func-
tionalized cyclopentenes.[3b,17,18] The tested olefins included
methyl acrylate and isobutyl acrylate as described in the
original paper,[3b] and phenyl vinyl sulfone was also used as
a new substrate. With these three substrates only moderate
enantioselectivities (up to 33% ee) were observed (Table 4).
Apparently, the size of the ester group in the electron-de-
ficient olefins did not affect the enantioselectivity.[3b] In
terms of regioselectivity, phosphane 5 was found to be more
selective than 4 as it favored the formation only of the 1,4-
substituted cyclopentene, whereas in the presence of 4 small
Entry Starting
materials
Solvent Ligand[a] Product Yield ee
(mol %)
(%)
(%)
1
2
3
4
5
10 ϩ 11 CH2Cl2 4 (10)
13
86
52
46
86
86
13[b]
18[b]
20[b]
31[b]
15[d]
10 ϩ 11 toluene 4 (12.5)[c] 13
10 ϩ 11 THF
10 ϩ 11 THF
10 ϩ 12 THF
4 (12.5)
5 (12.5)
5 (10)
13
13
14
[a]
[b]
[Pd(allyl)Cl]2 was used as catalyst precursor (1.25 mol %);
%
ee was measured by 1H NMR spectroscopy in the presence of
[c]
Eu(hfc)2 shift reagent; A solution of KCl (30 mol %) and 18-c-6
(20 mol %) in toluene was added; % ee was measured by HPLC
with Chiralcel OD-R column (methanol, water Ϫ8:2).
[d]
Eur. J. Org. Chem. 2004, 3913Ϫ3918
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3915

Z. Pakulski, O. M. Demchuk, J. Frelek, R. Luboradzki, K. M. Pietrusiewicz
FULL PAPER
The asymmetric induction at the level of 96% ee achieved
in the studied Pd-catalyzed allylic substitution reaction of
a cyclic carbonate points to potential usefulness of the bi-
cyclo[3.3.0]octane structural motif in ligand design.
Table 4. CϪC Bond formation (24 h, room temp., benzene)
Experimental Section
The solvents were purified and dried by literature methods. All
reactions were performed under Ar. TLC was performed on silica
gel (HF-254) and column chromatography on silica gel (230Ϫ400
Entry Starting
Cat.
Product Yield ee (%)
materials (10 mol %)[a]
(%)[b]
mesh, Merck). NMR spectra were recorded with a Varian Mercury
1
400BB instrument at 400 MHz (for H) or 100 MHz (for 13C) and
1
6
5
4
5
6
19 ϩ 22
19 ϩ 22
19 ϩ 23
19 ϩ 23
19 ϩ 24
19 ϩ 24
4
5
4
5
4
5
25
25
26
26
27
27
42[c]
62
21[d]
26[d]
0[f]
1
a Bruker AM 500 spectrometer at 500 MHz (for H) or 125 MHz
(for 13C) in deuteriochloroform (CDCl3) or deuteriobenzene
(C6D6) with Me4Si as internal standard. CD spectra were measured
on a JASCO 715 spectropolarimeter in acetonitrile. High-resolu-
tion mass spectra (HR-MS) were measured with AMD-604 and
MARINER mass spectrometers. Optical rotations were measured
with a JASCO DIP-360 automatic polarimeter.
46[e]
63
14[f]
29[g]
33[g][h]
46
71
[a]
The reaction was carried out under Ar with chiral phosphane
(10 mol %), 19 (100 mol %) and electron-deficient olefins (1000
[b]
mol %).[3b]
1,4 vs. 1,5 substitution was assigned by 1H NMR
6% of the 1,5-isomer was also isolated.
1-Phenyl-1,3a,4,5,6,6a-hexahydrocyclopenta[b]phosphole
1-Oxide
[c]
[d]
spectroscopy.
% ee
(6): A solution of 1-phenyl-2,5-dihydro-1H-phosphole 1-oxide
(178 mg, 1.0 mmol) in THF (2 mL) was added at Ϫ78 °C to a solu-
tion of freshly prepared LDA (2.1 mmol) in THF (6 mL). The re-
sulting deep red solution was stirred at Ϫ78 °C for 10 min, and a
solution of 1,3-diiodopropane (235 mg, 1.1 mmol) in THF (2 mL)
was then added in one portion. Stirring at Ϫ78 °C was continued
for 1 h and the reaction mixture was quenched by addition of a
few drops of water at Ϫ78 °C and then allowed to come to room
temperature. Solvents were evaporated. Column chromatography
of the residue (hexane/ethyl acetate/methanol, 5:3:1 as eluent) gave
pure 6 (161 mg, 74%). HR-MS(EI) calcd. for C13H15PO [M]ϩ:
218.0860; found 218.0857.
values were measured by GC on a γ-Dex 225 column.[3b] [e] 18% of
[f]
the 1,5-isomer was also isolated. % ee values were measured by
HPLC on a Chiralcel OD-H column (hexane, isopropyl alcohol
[g]
99:1). % ee values were measured by HPLC on a Chiralcel OD-
[h]
H column (hexane, isopropyl alcohol 8:2).
0.8, chloroform).
[α]2D0 ϭ Ϫ2.4 (c ϭ
amounts of 1,5-substituted isomers were always detected
(6Ϫ18%). For phenyl vinyl sulfone the formation of only
1,4-substituted product was observed regardless of the
phosphane used.
Recently, monophosphanes have also been found to be
efficient ligands for Rh-, Ru-, and Ir-catalyzed asymmetric
hydrogenation, although high ees of the hydrogenated prod-
ucts are still rarely achieved.[19,20] Rhodium-catalyzed
asymmetric hydrogenation of (Z)-N-acetylaminocinnamic
acid (28) in the presence of (SP)-4 and Rh[COD]2BF4 as
catalyst precursor gave N-acetylphenylalanine (29) with
only 10% ee (Scheme 4). The same reaction in the presence
of (RP)-5, however, afforded the product with 62% ee. In
both cases quantitative conversion of the starting material
took place (Scheme 4).
1-Phenyl-1,3a,4,5,6,6a-hexahydrocyclopenta[b]phosphole 1-Sulfide
(7): A solution of 1-phenyl-2,5-dihydro-1H-phosphole 1-sulfide
(1.94 mg, 10.0 mmol) was added at Ϫ78 °C to a solution of freshly
prepared LDA (30 mmol) in THF (170 mL). The resulting red solu-
tion was stirred at Ϫ78 °C for 10 min, and a solution of 1,3-di-
bromopropane (2.12 g, 10.5 mmol) in THF (5 mL) was then added
in one portion. Stirring at Ϫ78 °C was continued for 1 h, and the
reaction mixture was quenched by addition of a few drops of water
at Ϫ78 °C and then allowed to come to room temperature. Solvents
were evaporated. Column chromatography of the residue (hexane/
ethyl acetate 9:1 as eluent) gave pure 7 (2.13 g, 91%). HR-MS(EI)
calcd. for C13H15PS [M]ϩ: 234.0632; found 234.0631.
1-Phenyl-1,3a,4,5,6,6a-hexahydrocyclopenta[b]phosphole (4). Method A.
From 1-Phenyl-1,3a,4,5,6,6a-hexahydrocyclopenta[b]phosphole 1-
Oxide (6): A solution of racemic phosphane oxide 6 (4.111 g,
18.8 mmol) and phenylsilane (3.36 g, 31.0 mmol) in toluene (8 mL)
was heated under argon at 100 °C (bath temperature) for 5 h. Sol-
vents were evaporated, and the residue was filtered through silica
gel with hexane as eluent and evaporated again to give phosphane
4, which was used for separation of enantiomers without further
purification.
Scheme 4. Hydrogenation of 28 in the presence of Rh[COD]2BF4
and 4 or 5 (MeOH/THF, 24 h, room temp.)
Method B. From 1-Phenyl-1,3a,4,5,6,6a-hexahydrocyclopenta[b]phos-
phole 1-Sulfide (7): Tris(trimethylsilyl)silane (1.58 mL, 5.1 mmol)
and AIBN (40 mg) were added to a solution of racemic phosphane
sulfide 7 (1.171 g, 5.0 mmol) in toluene (25 mL), and the mixture
was heated at 80 °C for 16 h. Volatiles were evaporated to dryness,
and the residue was filtered through silica gel with hexane as eluent
and evaporated again to yield 4, which was used for separation of
Conclusions
In summary, we have demonstrated that the readily avail-
able enantiomerically pure bicyclic phosphanes (SP)-4 and
(RP)-5 are efficient catalysts and ligands for asymmetric
CϪC bond-forming and hydrogenation reactions.
3916
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 3913Ϫ3918

New Monodentate P,C-Stereogenic Bicyclic Phosphanes
FULL PAPER
enantiomers without further purification. 1H NMR (C6D6): δ ϭ
7.47 (m, 2 H, Ar), 7.10 (m, 3 H, Ar), 5.99 (ddd, JH,P ϭ 43.5, J3,4 ϭ
bromoacetate[21] (890 mg, 3.2 mmol) in ethyl acetate (8 mL). The
crystals that precipitated (1.025 g, 71%) were recrystallized twice
7.5, J ϭ 2.4 Hz, 1 H, H-3), 5.93 (ddd, JH,P ϭ 69.0, J ϭ 7.5, J ϭ from ethyl acetate/ethyl alcohol/benzene (20:1:1) as described above
2.6 Hz, 1 H, H-4), 3.28 (m, 1 H), 2.32 (m, 1 H), 1.78 (m, 2 H), 1.43 for salt 8a to afford the diastereomerically pure salt 8aЈ (538 mg,
(m, 2 H), 1.28 (m, 2 H) ppm. 13C NMR (C6D6): δ ϭ 148.1 (d,
JC,P ϭ 1.7 Hz), 132.6 (d, JC,P ϭ 18.9 Hz), 128.5, 128.6, 128.2 (d,
J
37%, Ͼ98% ee). [α]2D0 ϭ Ϫ80.4 (c ϭ 0.34, chloroform). NMR spec-
troscopic data of 8aЈ were identical to those of 8a. HR-MS/LSIMS
C,P ϭ 6.0 Hz), 53.7 (d, JC,P ϭ 4.3 Hz, CH), 45.1 (d, JC,P ϭ 6.8 Hz, calcd. for C25H36O2P [M Ϫ Br]ϩ: 399.2453; found 399.2459.
CH), 33.7 (d, JC,P ϭ 33.5 Hz, CH2), 31.5 (d, JC,P ϭ 3.4 Hz, CH2), The remaining filtrates were combined, evaporated to dryness, and
26.0 (d, JC,P ϭ 5.1 Hz, CH2) ppm. 31P NMR (C6D6): δ ϭ 25.4 ppm.
hydrolyzed as described above, and yielded recovered 6 (355 mg,
54% calculated for starting racemic 6), [α]2D0 ϭ Ϫ2.0 (c ϭ 0.41,
chloroform), which could be recycled in the resolution process.
Hydrolysis of 8a (2.260 g, 4.7 mmol) as described above gave
(1R,3aS,6aS)-(ϩ)-6 (990 mg, 96%). [α]2D0 365.1 (c ϭ 0.55, chloro-
form).
Hydrolysis of 8aЈ (515 mg, 1.07 mmol) as described above gave
(1S,3aR,6aR)-(Ϫ)-6 (222 mg, 95%). [α]2D0 ϭ Ϫ363.6 (c ϭ 0.33,
chloroform).
Resolution of rac-4 into its Enantiomers: Crude phosphane 4
(18.8 mmol) was dissolved in ethyl acetate (50 mL). This solution
was added in one portion to a solution of (Ϫ)-menthyl bromoace-
tate[21] (5.85 g, 21.1 mmol) in ethyl acetate (10 mL) and the re-
sulting mixture was left overnight at room temperature. The pre-
cipitated crystals were collected on sintered glass, washed with ethyl
acetate/hexane (1:1), and dried under vacuum to afford a mixture
of diastereoisomeric phosphonium salts (7.430 g, 82%). Crystalliza-
tion from ethyl acetate/ethyl alcohol/benzene (20:1:1) yielded crys-
tals A (2.635 g, 29%, ee Ͼ 92%) and filtrate A. Crystals were recrys-
tallized from the same solvent mixture to afford diastereomerically
pure salt 8a (2.260 g, 25%, Ͼ98% ee) and filtrate B.
X-Ray Crystallographic Study and Crystal Data for 8aЈ: Diffraction
data were collected on a Kappa CCD diffractometer with graphite
monochromated Mo-Kα radiation. Structure was solved by direct
methods (SHELXS-97) and refined on F2 by the full-matrix, least-
squares method (SHELXL-97).[22] The non-hydrogen atoms were
refined anisotropically; hydrogen atoms were refined isotropically
with a riding model.
Compound 8a: [α]2D0 ϭ 78.1 (c ϭ 0.3, chloroform). 1H NMR
(CDCl3), δ ϭ 8.20 (m, 2 H, Ar), 7.67 (m, 1 H, Ar), 7.60 (m, 2 H,
Ar), 7.22 (ddd, JH,P ϭ 45.9, J ϭ 2.2, J ϭ 7.9 Hz, 1 H, H-3), 7.01
(ddd, JH,P ϭ 26.3, J ϭ 2.4, J ϭ 7.9 Hz, 1 H, H-2), 4.80 (dd, JH,P ϭ
13.5, JH,CH ϭ 17.4 Hz, 1 H, CHP), 4.62 (m, J ϭ 4.4 Hz, J ϭ
Formula: C25H36O2Pϩ BrϪ, M ϭ 479.42, orthorhombic, space
˚
group P212121, a ϭ 7.1060(2), b ϭ 11.3810(4), c ϭ 30.3240(10) A,
3
˚
V ϭ 2452.40(14) A , Z ϭ 4, F(000) ϭ 1008, R1 ϭ 0.0329 [I Ͼ
2σ(I)), wR2 ϭ 0.0844 for all data. The absolute configuration was
known from the reference (1S,2R,5S)-menthyl fragment.
10.9 Hz, 1 H, OCH-menthyl), 4.51 (dd, JH,P ϭ 13.1, JH,CH
ϭ
17.4 Hz, 1 H, CHP), 3.82 (m, 1 H), 3.54 (m, 1 H), 2.63 (m, 1 H),
2.15 (m, 1 H), 2.06 (m, 1 H), 1.60Ϫ1.88 (m, 7 H), 1.52 (m, 1 H),
1.36 (m, 1 H), 1.29 (m, 1 H), 0.92 (m, 2 H), 0.85 (d, J ϭ 6.5 Hz, 3
H, iPro-CH3), 0.79 (d, J ϭ 7.0 Hz, 3 H, iProϪCH3), 0.58 (d, J ϭ
CCDC-238149 contains the crystallographic data for this paper.
These data can be obtained free of charge at www.ccdc.cam.ac.uk/
conts/retrieving.html [or from the Cambridge Crystallographic
Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; Fax:
ϩ44-1223-336-033; E-mail: deposit@ccdc.cam.ac.uk].
6.9 Hz, 3 H, CH3) ppm. 13C NMR (CDCl3): δ ϭ 165.2 (d, JC,P
ϭ
3.9 Hz), 163.6 (d, JC,P ϭ 20.4 Hz), 134.3 (d, JC,P ϭ 3.1 Hz), 132.6
(d, JC,P ϭ 10.7 Hz), 129.8 (d, JC,P ϭ 13.1 Hz), 120.1 (d, JC,P
84.8 Hz), 113.8 (d, JC,P ϭ 76.2 Hz), 77.7 (s), 53.3 (d, JC,P
ϭ
ϭ
(1S,3aR,6aR)-1-Phenyloctahydrocyclopenta[b]phosphole
1-Oxide
12.7 Hz), 46.4 (s), 40.3 (s, CH2), 37.7 (d, JC,P ϭ 55.0 Hz), 33.8 (s,
CH2), 31.8 (d, JC,P ϭ 54.2 Hz, CH2), 31.4 (s), 31.1 (d, JC,P
(9): Pd/C (10%, 100 mg) was added to a solution of (Ϫ)-1-phenyl-
1,3a,4,5,6,6a-hexahydrocyclopenta[b]phosphole 1-oxide [(Ϫ)-6,
218 mg, 1.0 mmol] in methanol (8 mL), and the reaction mixture
was kept under hydrogen atmosphere (balloon) at room temp. for
24 h. The mixture was then filtered through a short silica column
with methanol as eluent. The filtrate was evaporated to dryness to
yield the title compound (203 mg, 92%) as an oil. [α]2D0 ϭ Ϫ43.2
(c ϭ 0.53, chloroform). M.p. 40Ϫ42 °C. HR-MS(EI) calcd. for
C13H17PO [M]ϩ: 220.1017; found 220.1009.
ϭ
3.8 Hz, CH2), 29.5 (d, JC,P ϭ 1.5 Hz, CH2), 26.7 (d, JC,P ϭ 7.6 Hz,
CH2), 25.8 (s), 22.9 (s, CH2), 21.8 (s), 20.7 (s), 15.8 (s) ppm. 31P
NMR (CDCl3),
δ ϭ 55.6 ppm. HR-MS/LSIMS calcd. for
C25H36O2P [M Ϫ Br]ϩ: 399.2453; found 399.2436.
Filtrates A and B were combined, and the solvents were evaporated
to dryness to give a mixture of salts (4.892 g, 54%) enriched in the
1
P-epimer of 8a (ca. 40% ee). H NMR (CDCl3): δ selected signals
of the predominant epimer: 7.23 (ddd, JH,P ϭ 46.0, J ϭ 2.2 Hz,
J ϭ 8.0 Hz, 1 H, H-3), 6.90 (ddd, JH,P ϭ 29.0, J ϭ 2.3, J ϭ 7.9 Hz,
1 H, H-2), 4.60 (d, 2 H, PCH2), 4.56 (m, J ϭ 4.5 Hz, J ϭ 10.9 Hz,
1 H, OCHϪmenthyl), 0.49 (d, J ϭ 7.0 Hz, 3 H, CH3) ppm. 31P
NMR (CDCl3): δ ϭ 55.5 ppm. The mixture was dissolved in di-
chloromethane (60 mL), aqueous sodium hydroxide (10%, 50 mL)
was added, and the resulting two-phase reaction mixture was
stirred at room temp. for 3 h. The organic layer was separated, and
the aqueous layer was extracted with dichloromethane (2 ϫ
20 mL). The organic layers were combined, washed with water (2
ϫ 10 mL), dried over Na2SO4, and filtered, and the solvents were
evaporated to dryness. Column chromatography of the residue
(hexane/ethyl acetate/methanol, 5:3:1) gave enantioenriched phos-
phane oxide 6 (2.027 g, 91%, 40% ee). [α]2D5 ϭ Ϫ152.5 (c ϭ 0.42,
(1R,3aR,6aR)-1-Phenyloctahydrocyclopenta[b]phosphole (5): Re-
duction of (Ϫ)-1-phenyloctahydrocyclopenta[b]phosphole 1-oxide
[(Ϫ)-9, 191 mg, 0.87 mmol] with phenylsilane as described for 4
gave the title compound (175 mg, 98%) as an oil. 1H NMR (C6D6):
δ ϭ 7.26 (m, 2 H, Ar), 7.08 (m, 2 H, Ar), 7.01 (m, 1 H, Ar), 2.54
(m, 1 H), 2.46 (m, 1 H), 2.00 (m, 1 H), 1.46Ϫ1.80 (m, 7 H), 1.22
(m, 1 H), 0.90 (m, 1 H) ppm. 13C NMR (C6D6): δ ϭ 131.0 (d,
JC,P ϭ 14.6 Hz), 128.2, 128.0, 127.8, 46.2 (JC,P ϭ 3.4 Hz), 45.3 (d,
JC,P ϭ 12.1 Hz), 34.1 (d, JC,P ϭ 30.2 Hz, CH2), 33.9 (s, CH2), 33.8
(d, JC,P ϭ 3.4 Hz, CH2), 27.9 (d, JC,P ϭ 8.6 Hz, CH2), 26.0 (d,
JC,P ϭ 14.6 Hz, CH2) ppm. 31P NMR (C6D6): δ ϭ 7.1 ppm.
General Procedure for Palladium-Catalyzed Allylic Substitution: Di-
chloroform). The enantioenriched phosphane oxide 6 (655 mg, methyl malonate (0.76 mmol) and BSA (1 mmol) were dissolved in
3.0 mmol) was treated with phenylsilane (815 mg, 7.5 mmol) as de-
scribed above, and the resulting crude phosphane was dissolved
in ethyl acetate (2 mL) and added to a solution of (ϩ)-menthyl
THF (2 mL), and the solution was stirred at room temp. for 30 min.
It was then cooled to Ϫ20 °C, and a solution of [Pd(allyl)Cl]2 (1.25
mol %) and ligand (4 or 5, 12.5 mol %) in benzene (0.2 mL) was
Eur. J. Org. Chem. 2004, 3913Ϫ3918
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3917

Z. Pakulski, O. M. Demchuk, J. Frelek, R. Luboradzki, K. M. Pietrusiewicz
FULL PAPER
added. The resulting mixture was stirred at Ϫ20 °C for 30 min and
at 0 °C for 20 min and was cooled again to Ϫ20 °C, a solution of
(E)-1,3-diphenylallyl acetate (0.4 mmol) in THF (1 mL) was added,
and the mixture was stirred at room temp. overnight. Solvents were
evaporated to dryness, and column chromatography of the residue
gave 13 (hexane/acetone, 15:1, as eluent), 14 (hexane/ethyl acetate,
9:1), or 17 (hexane/acetone, 10:1 as eluent).
[1]
[2]
Comprehensive Asymmetric Catalysis (Eds.: E. N. Jacobsen, A.
Pfaltz, H. Yamamoto) Springer, Berlin, 2000.
[2a]
For reviews see:
M. Ohff, J. Holz, M. Quirmback, A.
[2b]
Börner, Synthesis 1998, 1391Ϫ1415.
G. Buono, O. Chiodi,
F. Lagasse, H. B. Kagan,
[2c]
M. Wills, Synlett 1999, 377Ϫ388.
Chem. Pharm. Bull. 2000, 48, 315Ϫ324.
[3]
[3a]
For selected recent results see:
Y. Hamada, N. Seto, H.
Ohmori, K. Hatano, Tetrahedron Lett. 1996, 37, 7565Ϫ7568.
Ethyl 4-(Phenylsulfonyl)cyclopent-1-ene-1-carboxylate (27): The chi-
ral phosphane 5 (120 µL of a 0.42  solution in benzene,
0.05 mmol) was added to a solution of ethyl 2,3-butadienoate (19,
56 mg, 0.5 mmol) and methyl phenyl sulfone (24, 101 mg,
0.6 mmol) in benzene (2 mL), and the mixture was stirred at room
temp. for 24 h. Solvents were evaporated, and the residue was puri-
fied by column chromatography (hexane/ethyl acetate, 7:3), which
[3b]
G. Zhu, Z. Chen, Q. Jiang, D. Xiao, P. Cao, X. Zhang, J.
[3c]
Am. Chem. Soc. 1997, 119, 3836Ϫ3837.
Z. Chen, G. Zhu,
Q. Jiang, D. Xiao, P. Cao, X. Zhang, J. Org. Chem. 1998, 63,
[3d]
5631Ϫ5635.
E. Vedejs, O. Daugulis, J. Am. Chem. Soc.
1999, 121, 5813Ϫ5814.
I. V. Komarov, A. Börner, Angew. Chem. Int. Ed. 2001, 40,
1197Ϫ1200.
Z. Pakulski, R. Kwiatosz, K. M. Pietrusiewicz, Tetrahedron
Lett. 2003, 44, 5469Ϫ5472.
D. Magiera, S. Moeller, Z. Drzazga, Z. Pakulski, K. M. Pietru-
siewicz, H. Duddeck, Chirality 2003, 15, 391Ϫ399.
K. L. Marsi, J. Org. Chem. 1974, 39, 265Ϫ267.
R. Romeo, L. A. Wozniak, C. Chatgilialoglu, Tetrahedron Lett.
2000, 41, 9899Ϫ9902.
[4]
[5]
[6]
1
gave 27 (100 mg, 71%) as a thick oil. H NMR (CDCl3): δ ϭ 7.90
(m, 2 H, Ar), 7.65 (m, 1 H, Ar), 7.56 (m, 2 H, Ar), 6.58 (m, 1 H,
CHϭ), 4.15 (q, J ϭ 7.1 Hz, 2 H, OCH2), 3.92 (m, 1 H, CHS), 3.08
(m, 2 H, CH2), 2.80 (m, 2 H, CH2), 1.24 (t, 3 H, CH3) ppm. 13C
NMR (CDCl3): δ ϭ 163.6 (CϭO), 139.4 (CHϭ), 137.9 (Cϭ),
134.4, 133.8, 129.3, 128.5, 61.6 (CHS), 60.5 (OCH2), 34.1 (CH2),
32.9 (CH2), 14.1 (CH3) ppm. HR-MS (ESI) calcd. for
C14H16NaO4S [M ϩ Na]ϩ: 303.0662; found 303.0677.
[7]
[8]
[9]
K. M. Pietrusiewicz, M. Zabłocka, Chem. Rev. 1994, 94,
1375Ϫ1411.
ees were determined by 1H NMR spectroscopy in the presence
of Kagan’s amide: Z. Pakulski, O. M. Demchuk, R. Kwiatosz,
[10]
N-Acetylphenylalanine (29): A ligand (4 or 5, 5 mol %) was added
to a cooled (Ϫ70 °C) solution of Rh[COD]2BF4 (1 mol %) in THF
(2 mL) and the resulting solution was stirred at Ϫ70 °C for 30 min
and additionally at 0 °C for 20 min. The solution of rhodium cata-
lyst was transferred to an autoclave, (Z)-N-acetylaminocinnamic
acid (28, 1.5 mmol) and Et3N (0.02 mL) in methanol (7 mL) was
added, and hydrogen pressure was applied (5 atm, 24 h). Solvents
were evaporated, and the residue was dissolved in aqueous NaOH
(1 ) and washed with dichloromethane (2 ϫ 15 mL). The aqueous
phase was acidified with H2SO4 (3 ), saturated with sodium sul-
fate, and extracted with chloroform. The organic extracts were
dried (Na2SO4), and the solvents were evaporated to yield N-acetyl-
phenylalanine (29) of [α]2D0 ϭ Ϫ4.1 (c ϭ 1.0, methanol) in the case
of 4, and [α]2D0 ϭ Ϫ24.9 (c ϭ 1.2, methanol) in case of 5. ref.[23]
[α]2D0 ϭ ϩ40.1 (c ϭ 1.0, methanol) for (S) enantiomer.
´
´
´
P. W. Osinski, W. Swierczynska, K. M. Pietrusiewicz, Tetra-
hedron: Asymmetry 2003, 14, 1459Ϫ1462.
B. M. Trost, D. L. Van Vranken, Chem. Rev. 1996, 96,
395Ϫ422.
[11]
[12]
[13]
[14]
[15]
J. M. Brown, D. I. Humes, P. J. Guiry, Tetrahedron 1994, 50,
4493Ϫ4506.
Y. M. Malone, P. J. Guiry, J. Organomet. Chem. 2000, 603,
110Ϫ115.
ˇ
ˇ
ˇ
I. J. S. Fairlamb, G. C. Lloyd-Jones, S. Vyskocil, P. Kocovsky,
Chem. Eur. J. 2002, 8, 4443Ϫ4453.
H. Braun, H. Felber, G. Kresse, A. Ritter, F. P. Schmidtchen,
F. P. Schneider, Tetrahedron 1991, 47, 3313Ϫ3328.
B. M. Trost, C. J. Li, J. Am. Chem. Soc. 1994, 116, 3167Ϫ3168.
C. Zhang, X. Lu, J. Org. Chem. 1995, 60, 2906Ϫ2908.
T. Hudlicky, J. D. Price, Chem. Rev. 1989, 89, 1467Ϫ1486.
W. Tang, X. Zhang, Chem. Rev. 2003, 103, 3029Ϫ3069.
W. Tang, W. Wang, X. Zhang, Angew. Chem. Int. Ed. 2003,
42, 943Ϫ946.
K. Burgess, I. Henderson, Tetrahedron 1991, 47, 6601Ϫ6616.
G. M. Sheldrick, SHELXS-97 and SHELXL-97, University of
Göttingen, Germany, 1997.
M. Murata, N. Yamamoto, K. Yoshikawa, T. Morimoto, K.
Achiwa, Chem. Pharm. Bull. 1991, 39, 2767Ϫ2769.
Received March 22, 2004
[16]
[17]
[18]
[19]
[20]
[21]
[22]
Acknowledgments
This work was supported by the State Committee for Scientific
Research, Poland, grant no. 7 TO9A 068 20.
[23]
3918
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 3913Ϫ3918