Enantioselective hydrogenation with self-assembling rhodium phosphene catalysts: Influence of ligand structure and solvent

Article abstract of DOI:10.1002/chem.200601607

Three sets of new and related chiral phospholane and phosphepine ligands have been prepared for Rh-catalyzed enantioselective hydrogenation. The size and substitution pattern of the cyclic monophosphanes were varied. More importantly, the ligands differ in the nature of the heterocyclic group linked to the trivalent phosphorus atom: 2-pyridone or 2-alkoxypyridine. In the corresponding Rh complexes, the pyridone units of two monodentate P ligands can assemble by hydrogen bonding and form chelates. In contrast, synthetic precursors bearing alkoxypyridine appendages are not able to aggregate via intramolecular hydrogen bonds. The nature of self-assembly is dependent on the nature of the P ligand and the solvent used for the hydrogenation (CH ₂Cl₂ vs. MeOH). These features affect the rate of the reaction as well as the enantioselectivity, which varied in the range of 0-99 % ee Complexation studies and DFT calculations were performed to explain these differences.

Full text of DOI:10.1002/chem.200601607

DOI: 10.1002/chem.200601607
Enantioselective Hydrogenation with Self-Assembling Rhodium Phosphane
Catalysts: Influence of Ligand Structure and Solvent
Mandy-Nicole Birkholz,[a, b] Natalia V. Dubrovina,[b] Haijun Jiao,[b] Dirk Michalik,[b]
Jens Holz,[b] Rocco Paciello,[c] Bernhard Breit,*[d] and Armin Bçrner*[a, b]
Dedicated to Professor Hanswalter Krause
Abstract: Three sets of new and related
chiral phospholane and phosphepine li-
gands have been prepared for Rh-cata-
lyzed enantioselective hydrogenation.
The size and substitution pattern of the
cyclic monophosphanes were varied.
More importantly, the ligands differ in
the nature of the heterocyclic group
linked to the trivalent phosphorus
atom: 2-pyridone or 2-alkoxypyridine.
In the corresponding Rh complexes,
the pyridone units of two monodentate
P ligands can assemble by hydrogen
bonding and form chelates. In contrast,
synthetic precursors bearing alkoxypyr-
idine appendages are not able to aggre-
gate via intramolecular hydrogen
bonds. The nature of self-assembly is
dependent on the nature of the P
ligand and the solvent used for the hy-
drogenation (CH2Cl2 vs. MeOH).
These features affect the rate of the re-
action as well as the enantioselectivity,
which varied in the range of 0–99% ee
Complexation studies and DFT calcu-
lations were performed to explain
these differences.
Keywords: asymmetric catalysis
·
density functional calculations · hy-
drogenation · P ligands · rhodium
Introduction
scale.[1] In particular, chiral ligands bearing a trivalent ligat-
ing phosphorus atom for “soft” metals like rhodium(I), ruth-
enium(II), iridium(I), and palladium(II) play a pivotal role
in this area.[2] Over the last 35 years a tremendous number
of P ligands have been described for this purpose. While sig-
nificant progress has been made in finding new ligand
motifs, a lackof fundamental understanding of how high
enantioselectivity can be achieved still remains. For a long
time, the field was dominated by the use of bidentate li-
gands Z which chelate on the metal center (Scheme 1).
Since 1999, some catalysts based on monodentate P ligands
X have also been found to be highly selective.[3] Evidence
was given that monodentate ligands can sometimes even de-
liver superior performance in comparison to their bidentate
Transition-metal-catalyzed enantioselective hydrogenation is
one of the most versatile reactions for the preparation of
optically active compounds on academic and industrial
[a] Dr. M.-N. Birkholz, Prof. Dr. A. Bçrner
Institut für Chemie der Universität Rostock
A.-Einstein-Strasse 3a, 18059 Rostock(Germany)
E-mail: armin.boerner@catalysis.de
[b] Dr. M.-N. Birkholz, Dr. N. V. Dubrovina, Dr. H. Jiao,
Dr. D. Michalik, Dr. J. Holz, Prof. Dr. A. Bçrner
Leibniz-Institut für Katalyse an der Universität Rostocke.V.
A.-Einstein-Strasse 29a, 18059 Rostock(Germany)
Fax : (+49)381-1281-5202
[c] Dr. R. Paciello
BASF AG
Basic Chemicals Research
GCB/O - M 313, 67056 Ludwigshafen (Germany)
[d] Prof. Dr. B. Breit
Institut für Organische Chemie und Biochemie
Albert-Ludwigs-Universität Freiburg
Albertstrasse 21, 79104 Freiburg (Germany)
Fax : (+49)761-203-8715
E-mail: bernhard.breit@organik.chemie.uni-freiburg.de
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Scheme 1. General types of metal complexes used for enantioselective
hydrogenation.
5896
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 5896 – 5907

FULL PAPER
analogues.[4] However, due to the presence of the essentially
rigid bridge in bidentate ligands, reliable comparison be-
tween the effects of two monodentate ligands and a biden-
tate ligand coordinated to the metal center is difficult. Re-
cently, the gap between these two ligand types was closed
by the concept of self-assembling catalysts, in which two
monodentate ligands are brought together by self-assembly
via hydrogen bonds (Y)[5] or dynamic metal–ligand interac-
tions.[6,7] As a result, chelates are formed.
ready been employed as mono- and bidentate ligands in Rh-
catalyzed enantioselective hydrogenation. For this purpose,
heterocyclic phosphanes of types 1–3 were prepared. Li-
gands of the a series are not able to aggregate, whereas li-
gands of the b series are potential candidates for self-assem-
bly. The phosphanes differ in the steric demand of the phos-
phorus heterocycle, which increases in the order 1<2<3.
In this context, Breit et al.
recently developed the 6-di-
phenylphosphanylpyridone
system IB, which self-assem-
bles through H-bonding with
its tautomer IA in the pres-
ence of a late transition metal
salt (Scheme 2).[8] Rhodium(I)
complexes II derived from this
system turned out to be highly
regioselective and active cata-
lysts for the hydroformylation
of alkenes, capable of operat-
ing even at ambient pressure
and room temperature.[8a–c]
Moreover, P ligands equipped with complementary H-
bonding motifs, on the basis of an A–T base-pair analogy,
have been prepared and advantageously employed in a com-
binatorial approach.[9] The concept was extended to Ru-cat-
alyzed anti-Markovnikov hydration of terminal alkynes.[8d]
Furthermore, a library of chiral P-donor ligands based on
the self-assembly approach through hydrogen-bonding has
been studied in the course of Rh-catalyzed enantioselective
hydrogenations.[10] In a library of 70 chiral self-assembled
catalysts consisting of chiral phosphanes and phosphonites,
phosphonite systems based on the BINOL skeleton turned
out to the best catalysts. High enantioselectivities
(>99% ee) and excellent catalyst activities were observed.
Since it is known that such atropisomeric ligands acting as
ordinary non-assembling monodentate ligands may also
induce high enantioselectivities, we envisaged investigating
the effect of the nature of the ligand on its self-assembly. As
ligands, we chose relatives of phosphanes which have al-
Results and Discussion
Synthesis ofnew ligands : In principle, pyridyl-substituted
phosphanes of the a series can be constructed either by cou-
pling the heterocyclic units (convergent approach) or by
prior synthesis of the primary 2-pyridylphosphane followed
by ring closure to give the cyclic phosphane (linear ap-
proach). Phosphanes of the b series with potential self-as-
sembling properties can be derived from the pyridyl precur-
sors by simple cleavage of the tert-butyl ether.
For the synthesis of 2,5-dimethyl-substituted phospholanes
1a,b, the linear approach, which has seen wide application
in the preparation of other chiral phospholane ligands,[11]
turned out to be the method of choice. Our synthetic route
commenced with 2,6-dibromopyridine (4), which was con-
verted via mono(tert-butyl) ether 5 to diamidophosphonite 6
(Scheme 3). Alcoholysis of the amido group afforded dieth-
ylphosphonite 7, which was reduced with LiAlH4 to give
pyridylphosphane 8. Deproto-
nation of the phosphane with
nBuLi, followed by double nu-
cleophilic substitution with the
cyclic sulfate of (S,S)-hexane-
2,5-diol (9) following the pro-
[12]
cedure of Burket al.,
gave
phospholane 1a. Cleavage of
the tert-butyl ether with formic
acid furnished 5-(2’,5’-dime-
thylphosphanyl)pyridone (1b).
The analogous pyridylphos-
pholanes 2a,b were not acces-
sible by the route detailed in
Scheme 3 due to the high sen-
Scheme 2. Self-assembly of 6-diphenylphosphanylpyridone in the presence of a transition metal salt (1) to give
highly active and regioselective hydroformylation catalysts (2).
Chem. Eur. J. 2007, 13, 5896 – 5907
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5897

B. Breit, A. Bçrner et al.
ence of a Cu catalyst in a one-
pot synthesis failed. Phosphane
oxide 18 was reduced with
LiAlH4/methyl
triflate
or
HSiCl3/NEt3 to afford 2a,
which could be purified via its
BH3 adduct 19. Removal of
the tert-butyl group finally fur-
nished pyridone derivative
(À)-2b. The same route was
used for the preparation of the
opposite enantiomer (+)-2b.
Phosphepines 3a and 3b are
functionalized analogues of the
chiral monodentate phosphe-
pines intensively investigated
by Gladiali et al.[15] and Beller
et al.[16] in several Rh-catalyzed
hydroformylations[15] and enan-
Scheme 3. Synthesis of 2,5-dimethyl-substituted phospholane ligands 1a,b.
tioselective
hydrogena-
sitivity of stereogenic a-carbon atoms in 2,5-diphenylphos-
pholanes with respect to epimerization. Therefore, an alter-
native route based on coupling of the two heterocyclic units
had to be developed. Using a
tions.[16,17] The route depicted in Scheme 5 is based on the
known access to enantiopure chlorophosphepine 24 starting
from (S)-binaphthol (20) according to protocols of Beller
modification of the pioneering
[13]
workof Fiaud et al.,
we pre-
pared racemic phosphinic acid
rac-13 starting from trans,trans-
1,4-diphenylbutadiene (10) via
phospholene (11), followed by
saturation of the heterocycle,
epimerization at one stereo-
genic a-C atom of meso-12 and
final hydrolysis of the amido
group (Scheme 4).
Phosphinic acid rac-13 was
resolved with quinine to afford
stereochemically pure (R,R)-
(+)-13. In analogy to our re-
cently suggested method,[14] the
phosphinic acid was converted
to secondary phosphane oxide
14 by substitution of the OH
group by chlorine followed by
reduction with LiAlH4 and oxi-
dation of the intermediate sec-
ondary phosphane. Compound
14 was coupled in the presence
of a PdII catalyst and Hünig
base with 6-tert-butyloxy-2-io-
dopyridine (17), which is easily
available in two steps starting
from 2-chloropyridine (15), to
give phosphane oxide 18.
All attempts to couple the
chloride derived from 13 with
2-bromopyridine 5 in the pres-
Scheme 4. Synthesis of 2,5-diphenyl-substituted phospholane ligands 2a,b.
5898
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 5896 – 5907

Self-Assembling Rhodium Phosphane Catalysts
FULL PAPER
phosphane groups will be ob-
served if 2-pyridone/pyridin-2-
ol tautomerism is slow on the
NMR timescale.
Structure of ligands and metal
complexes in solution: NMR
studies in CDCl3 or CD2Cl2
were performed to investigate
the self-assembly of phos-
phanes 1b, 2b, and 3b.[20] The
1H, 13C, and 31P NMR spectra
showed that the free ligands
exist in solution at room tem-
perature as a single species.
The single signal was found in
the 31P NMR spectra indicates
equivalence of the two
P
atoms. Furthermore, only one
1
signal for NH in the H NMR
spectra was observed. In ac-
cordance with literature prece-
dence, it can be concluded that
the free ligands aggregate in
nonpolar solvents mainly as
the pyridone/pyridone dimers
A (Scheme 6).[21]
Scheme 5. Synthesis of phosphepine ligands 3a,b.
et al.[16a,c] and Zhang et al.[18] Reduction of the hydroxyl
groups via bis-O-triflate 21 afforded 2,2’-dimethylbinaphtha-
lene (22). Lithiation and subsequent ring closure with di-
chloro(N,N-dimethylamino)phosphane yielded dimethylami-
nophosphepine 23. This was transformed with HCl into the
corresponding chlorophosphepine 24. Reaction with o-lithi-
ated 2-(tert-butyloxy)pyridine (5) gave pyridyl phosphepine
3a, which could be purified via its BH3 adduct 25. Cleavage
of the protecting group gave (S)-(+)-(1H-pyrid-6-on-2-yl)-
phosphepine 3b. The R enantiomer was prepared in the
same straightforward manner.
The Rh complex was formed by reaction of [Rh(cod)2]BF4
A
(cod=1,5-cyclooctadiene) with phospholane 1a and showed
a broad singlet in the 31P NMR spectrum in CDCl3. Hence,
a complex with two equivalent ligands was formed. Con-
versely, when two equivalents of phospholane 1b reacted
with [Rh
spectrum was characterized by two well-separated doublets
of doublets at d=54.5 (J(Rh,P)=146.2 Hz, J(P,P)=36.4 Hz)
and 50.5 ppm (J
(Rh,P)=138.6 Hz).[22,23] This is clear proof
(cod)2]BF4, the main component in the 31P NMR
N
ACHTREUNG
AHCTREUNG
for two different phosphorus nuclei. The P,P coupling con-
stant of about 36 Hz is indicative of a cis arrangement of the
two ligands at the metal center. Moreover, two sets of sig-
nals for the pyridine/pyridone protons were observed in the
1H NMR spectrum, which is in agreement with a ligand ar-
rangement of 1b such as shown for structure C (Scheme 6).
Similar complexation behaviour was observed when phos-
Intermolecular interactions ofphosphanes with pyridone
backbone and formation of metal complexes: Pyridonyl
phosphanes B are intrinsically capable of forming chelating
ligands by metal-induced self-assembly and complementary
H-bonding of the pyridone system with its hydroxypyridine
tautomer B’ (structure C, see
pholane 1b was treated with [PtCl2(cod)]. The typical pat-
U
Scheme 6). Alternatively, the
formation of metal-coordinat-
ed aggregates through intermo-
lecular H-bonding based on
pyridone–pyridone interactions
is conceivable (structure D). A
similar approach was recently
employed as a means to gener-
ate heterogeneous catalysts.[19]
In A, the two phosphorus
atoms are equivalent. In struc-
ture C, however, two different
Scheme 6. Modes of self-assembly of pyridone-based phosphanes and their effect in the reaction with metals.
Chem. Eur. J. 2007, 13, 5896 – 5907
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5899

B. Breit, A. Bçrner et al.
tern for a cis-Pt complex, characterized by Pt,P coupling
constants of 3700 and 3282 Hz, as well as a P,P coupling con-
stant of 17 Hz was found in the 31P NMR spectrum recorded
in CD2Cl2.[24]
garded as Gibbs free energy for discussion and comparison.
We also used this enthalpy difference to calibrate the theo-
retical methods used here.
Geometry optimizations were carried out at the B3LYP/6-
31G* level of density functional theory for all ligands. They
are given as energy-minimum structures at the same level of
theory and the resulting thermal corrections at 298 K have
been used for Gibbs free energy calculations. The obtained
structures were further refined at the B3LYP/6-311+G**
level of theory. Apart from the B3LYP calculated Gibbs
free energies, we also carried out ab initio MP2/6-311+G**
single-point energy calculations using the B3LYP/6-311+
G** optimized geometries. All calculations were done with
the Gaussian03 program suite.[33] The results are summar-
ized in Table 1.
In contrast, the reaction of phospholane 1b with
[Rh(CO)
N
N
terized by a typical P,Rh coupling constant of 125 Hz. Low-
ering the temperature to À608C did not change the spec-
trum. This is additional proof for the trans geometry. More-
over, this spectroscopic feature indicates that both ligands
have the same structure or that tautomerization is fast
under these conditions.
When two equivalents of phospholanes 2a,b reacted with
[Rh(cod)2]BF4, only undefined broad resonances were ob-
U
served in the 31P NMR spectrum.
The complex derived from the reaction of [Rh
C
The relative stability of 2-pyridone in the equilibrium of
2-pyridone and 2-hydroxypyridine is overestimated by
1.65 kcalmolÀ1 at the B3LYP/6-311+G** level compared
with the measured gas-phase stability. This agrees with the
reported literature value.[25] At the MP2/6-311+G**//
B3LYP/6-311+G** level, however, the relative stability of
2-pyridone is underestimated by 2.00 kcalmolÀ1. We also
carried out single-point calculations at the highly correlated
CCSD(T)/6-311+G** level of theory using the B3LYP/6-
311+G** geometry. CCSD(T) underestimates the relative
stability of 2-pyridone by 0.8 kcalmolÀ1. Due to the high
costs, we did not use CCSD(T) for other calculations. Here
we use the calculated energy differences, overestimated at
the B3LYP/6-311+G** level and underestimated at the
MP2/6-311+G** level, to scale the relative energies for
other ligands at both levels of theory for general compari-
son.
with [Rh
terized in CDCl3 by a doublet at d=48.8 ppm and two dou-
blets of doublets at d=60.5 (J(Rh,P)=180 Hz, J(P,P)=
47 Hz) and 68.0 ppm (J(Rh,P)=180 Hz, J(P,P)=46 Hz). The
A
A
A
ACHTREUNG
doublets of doublets are evidence for two different P nuclei
and therefore the formation of a self-assembled complex of
type C (Scheme 6), whereas the doublet indicates the forma-
tion of polynuclear complexes, such as D. A mixture of cis
(d=23.5 ppm, J
49.2 ppm, J(Pt,P)=2280 Hz) was observed in the reaction of
3b with [PtCl
2(cod)] in CDCl3.[25] In CD2Cl2, the solubility
of [PtCl2(3b)2] was too low. BlackPt precipitated after a
A
ACHTREUNG
ACHTREUNG
ACHTREUNG
few days.
As indicated in Table 1, the phospholane-substituted 2-hy-
droxypyridine derivatives are lower in Gibbs free energy
and are therefore the dominant tautomers in the gas phase.
DFT studies: The equilibrium of 2-pyridone and 2-hydroxy-
pyridine has been studied ex-
tensively, both experimental-
ly[26] and theoretically,[27,28] due
to its importance in DNA base
tautomers. In addition, this
equilibrium has been used as a
benchmarkto test the perfor-
mance of theoretical meth-
ods.[29] It is well known that
the equilibrium (constant) de-
pends strongly on the medium
and differs from that in the gas
phase. For example, gas-phase
measurements identified 2-hy-
droxypyridine as the dominant
tautomeric form.[30,31] The en-
thalpy difference in favour of
2-hydroxypyridine is about
0.77 kcalmolÀ1.[32] Due to the
much smaller entropy differ-
ence, this enthalpy difference
can also be approximately re-
Table 1. Computed relative Gibbs free energies DG [kcalmolÀ1].[a]
DG/B3LYP
0.88 (À0.77)
DG/MP2
À2.77 (À0.77)
0.27 (À1.38)
À3.56 (À1.56)
À4.57 (À2.57)
À0.29 (À1.95)
0.95 (À0.70)
À2.25 (À0.25)
[a] Scaled values are given in parentheses.
5900
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 5896 – 5907

Self-Assembling Rhodium Phosphane Catalysts
FULL PAPER
Compared to the parent 2-pyr-
idone/2-hydroxypyridine
system, the equilibrium for 1b
and 2b is shifted to the 2-hy-
droxypyridine
derivatives,
while that of 3b changes only
slightly. These changes, howev-
er, are too small for making
decisive conclusions without
consideration of the solvent ef-
fects.
Figure 1. B3LYP/LANL2DZ-optimized structures of 1C, 2C, and 3C.
The tautomerism between 2-
pyridone and 2-hydroxypyri-
dine, mediated by dual H-
bonding, was computed by Chou et al.[28] They found that
solvent changes the equilibrium of 2-pyridone and 2-hydrox-
ypyridine, and also that of their dimeric complexes. We also
carried out B3LYP DFT calculations in order to study com-
plex C in Scheme 6. The geometries were optimized at the
B3LYP/LANL2DZ level. They were characterized as energy
minima with only real frequencies. Due to their rather large
size, single-point energy calculations were done at both
B3LYP/LANL2DZ(d) and MP2/LANL2DZ(d) levels of
theory with the B3LYP/LANL2DZ geometries. These re-
sults are given in Table 2. The optimized structures are
shown in Figure 1.
1C-OH with 1b-OH/1b-OH by 10.9 kcalmolÀ1 at the
B3LYP level and by 7.9 kcalmolÀ1 at the MP2 level
(Table 2).
For complex 2C, the Gibbs free energy relative to 2C-OH
becomes smaller compared to 1C: 7.9 kcalmolÀ1 at the
B3LYP level and 3.2 kcalmolÀ1 at the MP2 level. This indi-
cates the effect of changing substituents from methyl to
phenyl. Indeed, this energy difference also reflects the quan-
titative difference of the equilibrium between 2b and 2b-
OH, whereby 2b-OH becomes more dominant upon phenyl
substitution as compared to 1b-OH.
For complex 3C, the Gibbs free energy relative to 3C-OH
is larger (15.3 kcalmolÀ1 at the B3LYP level and 14.3 kcal
molÀ1 at the MP2 level) than those of 1C and 2C. This indi-
cates that formation of 3C is more favoured thermodynami-
cally.
Complexes C-OH with two hydroxypyridine ligands, as
the dominant isomers in the gas phase, were also computed
for comparison in order to understand the strength of the
dual H-bonding in complexes of type C. Complex 1C with
ligand 1b/1b-OH is more stable in Gibbs free energy than
The optimized structures of 1C, 2C, and 3C in Figure 1
show nearly the same dual H-
bonding, as indicated by the
Table 2. Computed Gibbs free energies DG [kcalmolÀ1]; [Rh]=Rh
G
interatomic N N and O O
distances.
À
À
Ligand
Computed complex
X
DG[a]
DG[b]
Enantioselective
hydrogena-
1b
1C-OH
10.9
7.9
tion: The new ligands were
tested in the Rh-catalyzed
enantioselective hydrogenation
of the benchmarksubstrates
methyl N-acetamido acrylate
(aMe), methyl a-(Z)-N-acet-
amido cinnamate (AMe) and
dimethyl itaconate (ItMe2).
Hydrogenations were per-
formed with a molar precata-
lyst:substrate ratio of 1:100 or
1:50 at 1 bar hydrogen pres-
sure and 258C in CH2Cl2 or
MeOH as solvent. Precatalysts
were generated in situ by reac-
1C
0.0
7.9
0.0
0.0
3.2
0.0
2b
2C-OH
2C
3b
3C-OH
3C
15.3
0.0
14.3
0.0
tion of [Rh
ACHTREUNG
equivalents
ligand.
Satisfying enantioselectivi-
ties could only be achieved
[a] B3LYP/LANL2DZ(d)//B3LYP/LANL2DZ level. [b] MP2/LANL2DZ(d)/B3LYP/LANL2DZ level.
Chem. Eur. J. 2007, 13, 5896 – 5907
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5901

B. Breit, A. Bçrner et al.
Table 4. Enantioselective hydrogenation of methyl a-(Z)-N-acetamido
with bulky phosphepines 3a,b as ligands (Table 3) when hy-
drogenating the “small” substrate methyl N-acetamido acry-
late.
cinnamate (AMe).[a]
Run
Ligand
Solvent
t [min][b]
ee [%][c]
1
2
3
4
5
6
7
8
9
1a
1a
1b
1b
2a
2a
2b
2b
3a
3a
3b
3b
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
30
400
1400
1300
60
23 (R)
22 (R)
69 (R)
13 (R)[d]
71 (R)
77 (R)
83 (R)[e,f]
27 (R)[e]
56 (R)
90 (R)
94 (R)
64 (R)
Table 3. Enantioselective hydrogenation of methyl N-acetamido acrylate
(aMe).[a]
Run
Ligand
Solvent
t [min][b]
ee [%][c]
70
1
2
3
4
5
6
7
8
9
1a
1a
1b
1b
2a
2a
2b
2b
3a
3a
3b
3b
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
4
10
80
1400
75
17
960
400
5
2
10
32 (R)
35 (R)
51 (R)
35 (R)[d]
9 (R)
20 (R)
16 (R)[e]
5 (S)[e,f]
38 (R)
74 (R)
85 (R)
91 (R)
3900
2760
7
2
10
10
11
12
50
[a] For reaction conditions, see Table 3. [b] Determined after full conver-
sion. [c] Determined by GC, 25 m g-cyclodextrin, Lipodex E (Machery
and Nagel), fused silica, 1308C. [d] 20% conversion. [e] The substrate/
catalyst ratio is 50:1. [f] 92% conversion.
10
11
12
15
[a] Conditions: Precatalyst generated in situ by reaction of [Rh(cod)2]BF4
A
ing CH2Cl2 with MeOH. The opposite trend was observed
with 3b as ligand.
Similar trends as with AMe as substrate were observed in
the hydrogenation of dimethyl succinate (ItMe2) with self-
assembling ligands 1b, 2b and 3b (Table 5). The catalysts
with 2 equivalents of ligand, substrate:catalyst=100:1, H2 (1 bar), 7.5 mL
solvent. [b] Determined after full conversion. [c] Determined by GC,
25 m Chirasil-Val (Alltech), 1108C. [d] 76% conversion. [e] Substrate:ca-
talyst=50:1. [f] 92% conversion.
In general, the enantioselectivities observed match the
best results obtained with common monodentate phosphe-
pine/Rh catalysts (up to 95% ee).[16] Catalysts with 2,5-dime-
thylphospholanes 1a,b gave low ee values inferior to those
reported with the chelating bis-phospholanes BPE
(91.4% ee) and DuPHOS (99% ee).[12d] Unexpectedly, sub-
stitution of the methyl groups in the 2,5-position of the
phospholane rings (ligands of type 1) by phenyl groups (li-
gands of type 2) did not increase the enantioselectivity, irre-
spective of the solvents used. Interestingly, hydrogenation
with ligand 3b, capable of self-assembly, benefits from being
carried out in MeOH, where the formation of hydrogen
bonds is less likely. The extremely long reaction times when
phospholanes with self-assembling properties were used are
also noteworthy. This finding may be attributed to the for-
mation of catalytically less active polymers of type D
(Scheme 6).[34]
Table 5. Enantioselective hydrogenation of dimethyl succinate (ItMe2).[a]
Run
Ligand
Solvent
t [min][b]
ee [%][c]
1
2
3
4
5
6
7
8
9
1a
1a
1b
1b
2a
2a
2b
2b
3a
3a
3b
3b
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
1400
300
1300
400
40
11 (S)[d]
12 (S)
83 (S)[d,e]
0
49 (R)
76 (R)
51 (S)[d,f]
47 (S)
87 (S)
58 (S)
99 (S)
64 (S)
45
2880
2880
8
2
20
10
11
12
60
[a] For reaction conditions, see Table 3. [b] Determined after full conver-
sion. [c] Determined by GC, 25 m g-cyclodextrin, Lipodex E (Machery
and Nagel), fused silica, 708C. [d] The substrate/catalyst ratio is 50:1.
[e] 88% conversion. [f] 53% conversion.
Self-assembling ligands 1b, 2b, and 3b were advantageous
in terms of enantioselectivity (Table 4) in hydrogenation of
the sterically more demanding substrate methyl a-(Z)-N-
acetamido cinnamate (AMe). Superior ee values were ob-
served in the non-polar solvent CH2Cl2, whereas a dramatic
decrease was noted in MeOH. The bulkier the ligand, the
higher the enantioselectivity. As expected, enantioselectivi-
ties obtained with pyridyl phospholane ligands 1a and 2a
were insensitive to solvent polarity. The enantioselectivities
with 2,5-dimethylphospholanes of type 1a,b are lower than
those obtained with the chelating bis-phospholanes BPE
(85% ee) and DuPHOS (98% ee).[12d] The enantioselectivi-
ties with 2,5-diphenylphospholane ligands 2a,b are close to
those reported with an Rh catalyst based on the related
1,2,5-triphenylphospholane (93% ee).[13] In the case of phos-
phepine 3a, the enantioselectivity was increased by replac-
capable of self-assembly perform better in CH2Cl2. Differen-
ces of up to 83% ee were observed in comparison to the re-
action in MeOH, which is impressive evidence for the bene-
ficial effect of the self-assembling system. For comparison, a
related bidentate bis-phospholane (BPE) gave only 90% ee
in this hydrogenation.[12b] The highest enantioselectivity
(99% ee) was achieved with phosphepine ligand 3b in
CH2Cl2. This result is superior to those achievable with the
usual P-alkyl- or P-aryl-substituted phosphepines (22–
88% ee).[16c] With this substrate, striking differences in the
rate of the hydrogenation on employing self-assembling
phospholanes were also noted.
5902
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 5896 – 5907

Self-Assembling Rhodium Phosphane Catalysts
FULL PAPER
152 (31), 126 (28), 104 (100), 72 (72), 57 (19), 41 (16); HRMS (EI) calcd
for C17H32ON3P: 325.2283; found: 325.2281; elemental analysis calcd (%)
for C17H32ON3P (M=325.43): C 62.74, H 9.91, N 12.91, P 9.52; found: C
61.51, H 11.22, N 12.51, P 10.23.
Conclusion
Three sets of closely related chiral phospholanes 1a,b, 2a,b
and phosphepines 3a,b were prepared by a convergent or
linear synthetic pathway. Owing to the 2-pyridone function-
ality, phosphanes of the b series are capable of forming self-
assembled structures in the coordination sphere of a metal
atom. Phosphanes of the a series lackthis property. Com-
plexation studies with different metals revealed non-uniform
complexation behaviour of ligands of the b series, and this
was dependent on the metal as well as on co-ligands (cod,
CO, Cl). Moreover, in some cases, formation of polymeric
Rh complexes or P,N-coordinated species must be taken
into consideration. Density functional calculations support
the hypothesis that the nature of the heterocyclic phos-
phanes influences the tendency for formation of self-assem-
bled chelating structures. This is most favoured with phos-
phepine 3b as ligand. In the hydrogenation of benchmark
substrates the nature of the phosphane moiety and the sol-
vent used significantly affected the reaction rate and enan-
tioselectivity. In particular, hydrogenations with catalysts
based on the self-assembling phospholanes suffered from
slow reaction rates. Among all catalysts tested, phosphepine
3b gave the highest rates and ee values. Enantioselectivities
with this system exceed those of common monodentate
phosphepines.
6-tert-Butoxypyrid-2-ylphosphonous acid diethyl ester (7): A solution of
6-tert-butoxypyridin-2-ylphosphonous acid bis-diethylamide (6, 20.0 g,
61.5 mmol) in dry ethanol (120 mL) was refluxed for 24 h and cooled to
ambient temperature. The solvent was evaporated and the residue puri-
fied by distillation (b.p.0.4 =798C) to yield the desired product as a color-
less liquid (15.6 g, 95%). 1H NMR (250 MHz, CDCl3): d=7.51 (m, 1H,
CH), 7.21 (m, 1H, CH), 6.58 (m, 1H, CH), 3.88 (m, 4H, CH2), 1.61 (s,
9H, CH3), 1.27 ppm (t, J=7.3 Hz, 6H, CH3); 13C NMR (63 MHz,
CDCl3): d=164.0 (d, J=14 Hz, CP), 161.2 (s, CO), 138.0 (s, CH), 117.6
(d, J=14 Hz, CH), 113.8 (d, J=2 Hz, CH), 80.1 (s, CP), 63.1 (d, J=
11 Hz, CH2), 28.9 (s, CH3), 17.4 ppm (d, J=5 Hz, CH3); 31P NMR
(122 MHz, CDCl3): d=145.3 ppm; MS (EI): m/z (%): 271 (22) [M+], 227
(11), 215 (39), 187 (39), 171 (93), 158 (42), 142 (67), 121 (50), 93 (69), 78
(11), 65 (100), 58 (26), 47 (10), 41 (28); HRMS (EI) calcd for
C13H22O3NP: 271.1337; found: 271.1333; elemental analysis calcd (%) for
C13H22O3NP (M=271.29): C 57.55, H 8.17, N 5.16, P 11.42; found: C
56.78, H 8.29, N 5.07, P 11.48.
6-tert-Butoxypyrid-2-ylphosphane (8): A solution of 6-tert-butoxypyrid-2-
ylphosphonous acid diethyl ester (7, 30.6 g, 112.9 mmol) in dry diethyl
ether (60 mL) was slowly added at 108C to a suspension of LiAlH4
(4.93 g, 129.8 mmol) in dry diethyl ether (300 mL). The suspension was
allowed to warm to ambient temperature and stirred for 16 h. The reac-
tion was quenched by successive addition of water (4.9 mL), aqueous
NaOH (15%, 4.9 mL) and water (14.7 mL). After filtration the solution
was dried over Na2SO4 and evaporated. The crude product was purified
by distillation (b.p.0.7 =388C) to yield 8 as a colorless liquid (10.2 g,
49%). 1H NMR (250 MHz, CDCl3): d=7.35 (m, 1H, CH), 6.98 (m, 1H,
CH), 6.48 (m, 1H, CH), 4.05 (d, J=203.6 Hz, 2H, PH2), 1.59 ppm (s, 9H,
CH3); 13C NMR (63 MHz, CDCl3): d=163.5 (d, J=10 Hz, CO), 153.3 (d,
J=6 Hz, CP), 138.0 (d, J=4 Hz, CH), 121.5 (d, J=17 Hz, CH), 111.8 (s,
CH), 80.1 (s, CO), 28.9 ppm (s, CH3); 31P NMR (122 MHz, CDCl3): d=
À118.0 ppm; MS (CI): m/z (%): 184 (30) [M+], 170 (15), 160 (15), 144
(40), 128 (100); HRMS (EI) calcd for C9H13ONP: 182.0729; found:
182.0736.
Experimental Section
General: Unless otherwise mentioned, all reagents were purchased from
commercial sources and used without additional purification. Solvents
were dried and freshly distilled under argon before use. All reactions in-
volving phosphanes were performed under an argon atmosphere by using
standard Schlenktechniques. Optical rotations were measured on a gyro-
mat-HP instrument (Fa. Dr. Kernchen). NMR spectra were recorded
with Bruker AC 250, ARX 300 and AVANCE 500 spectrometers. Chemi-
cal shifts of 1H, 13C, and 31P NMR spectra are reported in parts per mil-
lion. The solvent peakwas used as internal standard. Chemical shifts of
31P NMR spectra are referred to H3PO4 as external standard. For NMR
measurements on the metal complexes, a 60–100-fold higher concentra-
tion compared to catalytic trials had to be used to accumulate the spectra
in a reasonable time. Elemental analyses were performed with a LECO
CHNS-932. Mass spectra were recorded on a MAT 95XP (Finnigan).
A
(1a):
nBuLi (1.6m in hexane, 14.6 mL, 23.3 mmol) was added dropwise to a so-
lution of 6-tert-butoxypyrid-2-ylphosphane (8, 4.0 g, 22.2 mmol) in dry
THF (120 mL) at À788C. The resulting mixture was stirred for 1 h at this
temperature. Then a solution of (2S,5S)-2,5-hexanediol cyclic sulfate[12d]
(3.94 g, 21.9 mmol) in dry THF (120 mL) was added dropwise. The solu-
tion was allowed to warm to ambient temperature and stirred for 4 h, fol-
lowed by dropwise addition of nBuLi (1.6m in hexane, 14.6 mL,
23.3 mmol) at À788C. The solution was warmed to ambient temperature
and stirred for 36 h, after which the reaction was quenched with metha-
nol (2 mL). The solvent was evaporated. After addition of brine (30 mL)
and diethyl ether (50 mL), standard extraction and purification of the
crude product by distillation (b.p.1.6 =1108C) gave 1a as a colorless liquid
(4.24 g, 72%). [a]=À46.28 (c=0.85, CHCl3); 1H NMR (250 MHz,
CDCl3): d=7.40 (m, 1H, CH), 7.08 (m, 1H, CH), 6.51 (m, 1H, CH), 3.22
(m, 1H), 2.26 (m, 2H), 1.91 (m, 1H), 1.56 (brs, 9H, CH3), 1.55 (m, 1H),
1.31 (dd, J=18.8 Hz, J=7.4 Hz, 3H, CH3), 1.30 (m, 1H), 0.85 ppm (dd,
J=10.8 Hz, J=7.1 Hz, 3H, CH3); 13C NMR (63 MHz, CDCl3): d=163.6
(s, CO), 158.8 (d, J=22 Hz, CP), 137.3 (d, J=11 Hz, CH), 124.6 (d, J=
41 Hz, CH), 112.8 (s, CH), 79.5 (s, CO), 37.5 (d, J=3 Hz, CH2), 36.6 (d,
J=12 Hz, CHP), 33.2 (d, J=7 Hz, CHP), 29.1 (s, CH3), 20.6 (d, J=
33 Hz, CH3), 15.6 ppm (s, CH3); 31P NMR (122 MHz, CDCl3): d=
13.4 ppm; MS (EI): m/z (%): 266 (100) [M+], 210 (15); HRMS (EI)
calcd for C15H25ONP: 266.1668; found: 266.1658; elemental analysis calcd
(%) for C15H24ONP (M=265.33): C 67.90, H 9.12, N 5.28, P 11.67;
found: C 67.46, H 8.84, N 6.00, P 14.00.
6-tert-Butoxypyrid-2-ylphosphonous acid bis-diethylamide (6): nBuLi
(1.6m in hexane, 54.4 mL, 86.9 mmol) was added dropwise to a solution
of 2-bromo-6-tert-butoxypyridine (5, 20.0 g, 86.9 mmol) in dry diethyl
ether (200 mL) at À788C. The reaction mixture was stirred for 15 min at
this temperature and for a further 1.5 h at ambient temperature. Then a
solution of bis(diethylamino)chlorophosphane[35] (18.3 g, 86.9 mmol) in
dry diethyl ether (200 mL) was added dropwise at À788C. The resulting
suspension was warmed to ambient temperature and stirred for 16 h. The
mixture was filtered and the reaction quenched with methanol (3.5 mL).
Evaporation of the solvent followed by purification of the crude product
by distillation (b.p.1 =1208C) gave 6 as a colorless oil (21.6 g, 77%).
1H NMR (250 MHz, CDCl3): d=7.42 (m, 1H, CH), 7.08 (m, 1H, CH),
6.42 (m, 1H, CH), 3.09 (m, 8H, CH2), 1.58 (s, 9H, CH3), 1.11 ppm (t, J=
7.3 Hz, 12H, CH3); 13C NMR (63 MHz, CDCl3): d=164.3 (s, CO), 162.8
(d, J=17 Hz, CP), 137.6 (d, J=3 Hz, CH), 119.6 (d, J=21 Hz, CH), 111.4
(d, J=3 Hz, CH), 79.3 (s, CO), 43.7 (d, J=44 Hz, CH2), 29.0 (s, CH3),
15.0 ppm (d, J=3 Hz, CH3); 31P NMR (122 MHz, CDCl3): d=93.9 ppm;
MS (EI): m/z (%): 325 (8) [M+], 254 (67), 197 (60), 175 (81), 169 (56),
(2R,5R)-2,5-Dimethyl-1-(1H-pyrid-6-on-2-yl)phospholane (1b): A solu-
tion of (2R,5R)-2,5-dimethyl-1-(6-tert-butoxypyrid-2-yl)phospholane 1a
(0.5 g, 2.39 mmol) and concentrated HCOOH (8 mL) was stirred for 3 h.
The solvent was removed under reduced pressure and the remaining
Chem. Eur. J. 2007, 13, 5896 – 5907
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5903

B. Breit, A. Bçrner et al.
solid was dissolved in dry CH2Cl2 (2 mL). Evaporation of the solvent and
drying of the product in vacuum gave pure 1b as a white solid (0.38 g,
98%), which was used without purification for further transformations.
M.p. 74–828C; [a]=À262.18 (c=1, CHCl3); H NMR (250 MHz, CDCl3):
31P NMR (122 MHz, CDCl3): d=54.3 ppm; MS (EI): m/z (%): 406 [M+]
(35), 350 (100); HRMS (EI) calcd for C25H27O2NP: 404.1756; found:
404.1774; elemental analysis calcd (%) for C25H28O2NP (M=405.47): C
74.05, H 6.96, N 3.45, P 7.64; found: C 73.86, H 6.96, N 3.31, P 7.84.
1
d=11.94 (brs, 1H, NH), 7.43 (m, 1H, CH), 6.53 (d, J=8.4 Hz, 1H, CH),
6.33 (m, 1H, CH), 2.64 (m, 1H), 2.27 (m, 2H), 2.00 (m, 1H), 1.34 (m,
2H), 1.29 (dd, J=19.8 Hz, J=7.4 Hz, 3H, CH3), 0.94 ppm (dd, J=
10.8 Hz, J=6.9 Hz, 3H, CH3); 13C NMR (63 MHz, CDCl3): d=165.4 (s,
CO), 148.4 (d, J=39 Hz, CP), 140.2 (d, J=3 Hz, CH), 120.4 (s, CH),
112.7 (d, J=9 Hz, CH), 37.4 (d, J=4 Hz, CH2), 36.9 (s, CH2), 36.2 (d, J=
12 Hz, CHP), 33.8 (d, J=9 Hz, CHP), 20.8 (d, J=33 Hz, CH3), 15.4 ppm
(s, CH3); 31P NMR (122 MHz, CDCl3): d=10.0 ppm; MS (EI): m/z (%):
210 (100) [M+], 194 (6), 166 (7), 152 (7), 136 (8); HRMS (EI) calcd for
C11H16ONP 209.0964, found 209.0957; elemental analysis calcd (%) for
C11H16ONP (M=209.22): C 63.15, H 7.71, N 6.69, P 14.80; found: C
62.97, H 8.04, N 6.51, P 15.29.
A
((2R,5R)-18b): (2R,5R)-18b was obtained by following the procedure for
the synthesis of the enantiomeric compound (2S,5S)-18b as described
above. M.p. 177–1798C; [a]=+61.68(c=1, CHCl3); MS (EI): m/z (%):
406 (35) [M+], 350 (100); HRMS (EI) calcd for C25H27O2NP 404.1774,
found 404.1756; elemental analysis calcd (%) for C25H28O2NP (M=
405.47): C 74.05, H 6.96, P 7.64; found: C 73.53, H 6.87, P 7.63.
A
2a): Method A: Methyl triflate (0.062 mL, 0.54 mmol) was added to a so-
lution of (2S,5S)-(À)-1-(6-tert-butoxypyrid-2-yl)-1-oxo-2,5-diphenylphos-
pholane ((2S,5S)-18, 0.20 g, 0.49 mmol) in dry DME (4 mL) . After stir-
ring for 2 h the solution was cooled to À208C and LiAlH4 (0.028 g,
0.74 mmol) was added. The solution was warmed to ambient temperature
and stirring was continued for 1.5 h. The solution was quenched with
water (0.028 mL) and dried over Na2SO4 before filtration over Celite
(3 cm) and addition of BH3·SMe2 (2m in toluene, 0.33 mL, 0.66 mmol).
After stirring for 15 h, the solvent was evaporated and the residue was
purified by flash chromatography (heptane/ethyl acetate 20/1) to give
borane adduct 19 as a white solid (0.12 g, 60%). M.p. 118–1218C; [a]=
A
chloride
(5.9 mL, 58.8 mmol) was added dropwise to a suspension of (2S,5S)-(À)-
1-hydroxy-1-oxo-2,5-diphenylphospholane (13,[13b] 4 g, 14.7 mmol) in dry
THF (80 mL) at 08C. After 5 min of stirring at this temperature and a
further 2 h at ambient temperature, the resulting solution was evaporated
and the residue dissolved in dry THF (40 mL). At À208C LiAlH4 (1.12 g,
29.5 mmol) was successively added. The suspension was stirred for 1 h at
this temperature. The reaction was quenched by successive addition of
aqueous NaOH (7%, 12 mL). The suspension was filtered and the filtrate
dried over Na2SO4 and evaporated. The residue was dissolved in iPrOH
(80 mL) and the solution stirred under air for 5 h. After evaporation of
the solvent and column chromatography (CH2Cl2/ethyl acetate 9/1), the
desired product was obtained as a white solid (2.1 g, 57%). Analytical
data were in accordance with those reported in reference [13b].
1
À55.98(c=1, CHCl3); H NMR (250 MHz, CDCl3): d=7.20–6.85 (m, 12H,
CH), 6.46 (m, 1H, CH), 4.80 (m, 1H), 3.63 (m, 1H), 2.60 (m, 1H), 2.32
(m, 3H), 1.60 (s, 9H, CH3), 1.50–0.00 ppm (brs, 3H, BH3); 13C NMR
(63 MHz, CDCl3): d=163.8 (d, J=4 Hz, CO), 140.7 (d, J=9 Hz, CH),
135.5 (d, J=2 Hz, C), 135.0 (d, J=6 Hz, C), 129.3, 129.0, 128.9, 128.8,
128.4, 128.2, 127.9, 127.8, 127.5, 127.3, 123.2, 118.1 (d, J=13 Hz, CH),
71.9 (s, CO), 47.8 (d, J=31 Hz, CHP), 43.9 (d, J=29 Hz, CHP), 32.6 (d,
J=8 Hz, CH2), 30.6 (s, CH2), 29.3 ppm (s, CH3); 31P NMR (122 MHz,
CDCl3): d=45.3 ppm; MS (EI): m/z (%): 403 (7) [M+], 389 (8) [M+
ÀBH3], 346 (21) [M+ÀtBu], 333 (100) [M+ÀBH3ÀtBu], 229 (36), 198
(15), 91 (25); HRMS (EI) calcd for C25H31ONBP: 403.2226, found
403.2231; elemental analysis calcd (%) for C25H31ONBP (M=403.30): C
74.45, H 7.75, N 3.47, P 7.68; found: C 74.10, H 7.74, N 3.33, P 8.02.
Borane adduct 19 was dissolved in THF (3 mL) and after addition of
DABCO (0.040 g, 0.36 mmol) the solution was refluxed for 2 h. After
evaporation of the solvent, the residue was dissolved in hexane (3 mL)
and filtered through Celite (3 cm). Evaporation of the solvent gave pure
2a (0.105 g, 54.7%).
2-(tert-Butoxy)-6-iodopyridine (17): A suspension of 2-chloro-6-iodopyri-
dine (16)[36] (6.77 g, 28.3 mmol) and KOtBu (6.35 g, 56.6 mmol) in dry tol-
uene (150 mL) was stirred at 808C for 24 h. After cooling to ambient
temperature and filtration over a plug of Celite, the filtrate was evaporat-
ed and the residue purified by distillation (b.p.5 =898C) or by flash chro-
matography (heptane/ethyl acetate 14/1) to yield 17 as a colorless liquid
(8.74 g, 56%). 1H NMR (250 MHz, CDCl3): d=7.26 (m, 1H, CH), 7.13
(m, 1H, CH), 6.59 (m, 1H, CH), 1.59 ppm (s, 9H, CH3); 13C NMR
(63 MHz, CDCl3): d=162.8 (s, CO), 139.7 (s, CH), 126.9 (s, CH), 112.9 (s,
CI), 112.2 (s, CH), 81.1 (s, CO), 28.4 ppm (s, CH3); MS (EI): m/z (%):
278 (10) [M+], 222 (100); HRMS (EI) calcd for C9H11ONI 275.9880,
found 275.9866; elemental analysis calcd (%) for C9H12ONI (M=277.10):
C 39.01, H 4.36, N 5.05, I 45.80; found: 38.50, H 4.80, N 4.83, I 45.93.
Method B: HSiCl3 (0.30 mL, 2.96 mmol) was added dropwise to a solu-
tion of (2S,5S)-(À)-1-(6-tert-butoxypyrid-2-yl)-2,5-diphenylphospholane
(18, 0.4 g, 0.98 mmol) and Et3N (0.41 mL, 2.96 mmol) in dry MeCN
(6 mL) at ambient temperature. After stirring for 24 h at reflux, aqueous
KOH (20%, 5.00 g) was added dropwise at 08C. The organic phase was
separated and the aqueous phase extracted with dry diethyl ether (2”
6 mL). The combined organic extracts were dried over Na2SO4 and then
the solvent was evaporated. The crude product was dissolved in diethyl
ether (2 mL) and filtered over Celite (3 cm). Evaporation of the solvent
afforded 2a as a white solid (0.20 g, 52%), which was used without fur-
ther purification. M.p. 111–1158C; [a]=+67.58 (c=1, CHCl3); 1H NMR
(250 MHz, CDCl3): d=7.25 (m, 4H, CH), 7.09 (m, 1H, CH), 6.95 (m,
4H, CH), 6.77 (m, 2H, CH), 6.34 (m, 1H, CH), 6.15 (m, 1H, CH), 4.86
(m, 1H), 3.68 (m, 1H), 2.63 (m, 1H), 2.18 (m, 2H), 1.93 (m, 1H),
1.62 ppm (brs, 9H, CH3); 13C NMR (63 MHz, CDCl3): d=163.3 (d, J=
1 Hz, CO), 156.9 (d, J=26 Hz, CP), 145.3 (d, J=19 Hz, C), 139.1 (d, J=
2 Hz, C), 137.3 (d, J=11 Hz, CH), 128.8, 128.5, 128.1, 127.9, 126.1, 125.8,
125.0, 124.4, 112.8, 79.5 (s, CO), 50.4 (d, J=17 Hz, CHP), 44.2 (d, J=
12 Hz, CHP), 37.4 (s, CH2), 33.0 (s, CH2), 29.3 ppm (s, CH3); 31P NMR
(122 MHz, CDCl3): d=22.6 ppm; MS (EI): m/z (%): 389 (10) [M+], 333
(100), 255 (10), 229 (30), 198 (15), 117 (10), 91 (30), 78 (10); HRMS (EI)
calcd for C25H28ONP: 389.1903; found: 389.1904; elemental analysis calcd
(%) for C25H28ONP (M=389.47): C 77.10, H 7.25, N 3.60, P 7.95; found:
C 76.46, H 6.83, N 3.47, P 7.73.
ACHTREUNG
AHCTREUNG
trans-dibenzylideneacetone) and 1,3-bis(diphenylphosphanyl)propane
(dppp, 0.310 g, 0.75 mmol) in dry DMF (20 mL) was stirred for 15 min.
After addition of 2-tert-butoxy-6-iodopyridine (17, 3.12 g, 11.52 mmol) or
2-bromo-(6-tert-butoxypyridine (2.59 g, 11.52 mmol), stirring was contin-
ued for a further 15 min. A solution of iPr2NEt (4.7 mL, 37.4 mmol) and
(2S,5S)-(À)-1-oxo-2,5-diphenylphospholane (14, 2.40 g, 9.36 mmol) in dry
DMF (20 mL) was added, after which the mixture was stirred at 1058C
for 18 h. After evaporation of the solvent, the residue was dissolved in
CH2Cl2 (100 mL) and the resulting solution was extracted with aqueous
NaHCO3 (5%, 2”20 mL) and brine (20 mL). The organic phase was
dried over Na2SO4 and evaporated. After column chromatographic pu-
rification (CH2Cl2/ethyl acetate 9/1) and crystallisation (CH2Cl2/diethyl
ether 1/10) 18 was obtained as colorless crystals (2.43 g, 64%). M.p. 175–
1808C; [a]=À66.68(c=1, CHCl3); 1H NMR (250 MHz, CDCl3): d=7.24
(m, 6H, CH), 7.16 (m, 1H, CH), 7.02 (m, 4H, CH), 6.97 (m, 1H, CH),
6.49 (m, 1H, CH), 4.18 (m, 1H), 3.72 (m, 1H), 2.65 (m, 1H), 2.35 (m,
3H), 1.57 ppm (brs, 9H, CH3); 13C NMR (63 MHz, CDCl3): d=163.4 (d,
J=21 Hz, CO), 151.0 (d, J=118 Hz, CP), 138.2 (d, J=11 Hz, CH), 136.6
(d, J=3 Hz, C), 135.9 (d, J=4 Hz, C), 129.3 (d, J=6 Hz, CH), 128.9 (d,
J=2 Hz, CH), 128.4 (d, J=2 Hz, CH), 127.4 (d, J=5 Hz, CH), 127.2 (d,
J=2 Hz, CH), 126.7 (d, J=3 Hz, CH), 123.2 (d, J=17 Hz, CH), 116.4 (d,
J=3 Hz, CH), 80.1 (s, CO), 51.6 (d, J=61 Hz, CHP), 43.2 (d, J=63 Hz,
CHP), 32.6 (d, J=7 Hz, CH2), 29.2 (s, CH3), 28.2 ppm (d, J=9 Hz, CH2);
A
((2R,5R)-2a): (2R,5R)-2a was obtained by following the procedure for
the synthesis of the enantiomeric compound as described above. [a]=
5904
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 5896 – 5907

Self-Assembling Rhodium Phosphane Catalysts
FULL PAPER
À768 (c=1, CHCl3); MS (EI): m/z (%): 390 (100) [M+], 376 (32), 334
(81), 81 (25); HRMS (EI) calcd for C25H27ONP: 388.1825, found
388.1818; elemental analysis calcd (%) for C25H28ONP (M=389.47): C
77.10, H 7.25, N 3.60, P 7.95; found: C 72.58, H 7.03, N 3.33, P 7.63.
(100); HRMS (EI) calcd for C31H28NOP (M=461.53): 461.1903; found:
461.1895.
(R)-(+)-4-(6-tert-Butoxypyrid-2-yl)-4,5-dihydro-3H-dinaphtho[2,1-a;1’,2’-
e]phosphepine ((R)-(+)-3a): (R)-(+)-3a was prepared in the same
A
manner from
(R)-(+)-4-chloro-4,5-dihydro-3H-dinaphtho[2,1-a;1’,2’-
A solution of (2S,5S)-(À)-(6-tert-butoxypyrid-2-yl)-2,5-diphenylphospho-
lane ((2S,5S)-2a, 0.05 g, 0.043 mmol) and concentrated HCOOH (1 mL)
was stirred for 3 h. The solvent was evaporated and the residue dried in
vacuum at 408C to give the product as a white solid in quantitative yield.
e]phosphepine (24). [a]=+1338 (c=0.5, CHCl3); m.p. 91.0–91.58C.
(S)-(À)-4-(1H-pyridin-2-on-6-yl)-4,5-dihydro-3H-dinaphtho[2,1-a;1’,2’-e]-
A
and concentrated formic acid (8 mL) was stirred at ambient temperature
for 10 min and then the solvent was evaporated. The residue was dried in
vacuum at 408C for 5 h to give the pure product as colorless crystals
(0.203 g, 100%). M.p. 158.5–159.58C; [a]=À2578 (c=0.5, CHCl3);
1H NMR (CDCl3) d=11.01 (brs, 1H, NH), 7.85–8.01 (m, CH), 7.76 (d,
1H, J=9 Hz, CH), 7.58 (d, 1H, J=9 Hz, CH), 7.14–7.49 (m, CH), 7.06
(d, J=8 Hz, 1H, CH), 6.71 (d, J=9 Hz, 1H, CH), 6.14 (m, 1H, CH), 3.10
(m, 2H, CH2), 2.85 ppm (m, 2H, CH2); 13C NMR (CDCl3) d=165.1 (C=
O), 147.0, 141.7, 124.7–129.7, 118.5, 112.7, 29.3 (d, J=23 Hz, CH2),
25.7 ppm (d, J=15 Hz, CH2); 31P NMR (CDCl3): d=4.3 ppm; MS (EI):
m/z (%): 406 [M++1] (60), 329 (60), 283 (20); HRMS (EI) calcd for
C27H20NOP (M=405.43): 405.1277; found: 405.1272.
M.p. 69–738C; [a]=À123.68 (c=1, CHCl3); H NMR (250 MHz, CDCl3):
d=9.60 (brs, 1H, NH), 7.40–6.90 (m, 11H, CH), 6.28 (m, 1H, CH), 6.15
(m, 1H, CH), 3.96 (m, 1H), 3.80 (m, 1H), 2.65 (m, 1H), 2.38 (m, 2H),
1.97 ppm (m, 1H); 13C NMR (63 MHz, CDCl3): d=164.2 (s, CO), 146.7
(d, J=46 Hz, CP), 143.0 (d, J=19 Hz, C), 139.9 (d, J=5 Hz, CH) 137.4
(d, J=3 Hz, C), 129.0, 128.3, 127.5, 127.0, 126.9, 120.9, 113.0 (d, J=
15 Hz, CH), 48.9 (d, J=16 Hz, CHP), 46.3 (d, J=15 Hz, CHP), 37.3 (d,
J=2 Hz, CH2), 33.3 ppm (d, J=4 Hz, CH2); 31P NMR (122 MHz, CDCl3):
d=18.7 ppm; MS (EI): m/z (%): 333 (100) [M+], 259 (38), 229 (33), 198
(16), 152 (7), 127 (19), 91 (38), 77 (7); HRMS (EI) calcd for C21H20ONP
333.1277, found 333.1278; elemental analysis calcd (%) for C21H20ONP
(M=333.36): C 75.66, H 6.05, N 4.20, P 9.29; found: C 74.86, H 6.16, N
4.48, P 9.67.
(R)-(+)-4-(1H-pyrid-2-on-6-yl)-4,5-dihydro-3H-dinaphtho[2,1-a;1’,2’-e]-
AHCTREUNG
A
((2R,5R)-
2b): (2R,5R)-2b was obtained by following the procedure for the synthe-
sis of the enantiomeric compound described above. M.p. 66–718C; [a]=
+127.68 (c=1, CHCl3); MS (EI): m/z (%): 333 (100) [M+], 259 (13), 229
(37), 198 (17), 185.0 (5), 152 (7), 127 (20), 104 (10), 91 (42), 78 (6);
HRMS (EI) calcd for C21H20ONP: 333.1278; found: 333.1277; elemental
analysis calcd (%) for C21H20ONP (M=333.36): C 75.66, H 6.05, P 9.29;
found: C 74.22, H 6.25, P 9.79.
A
ACHTREUNG
AHCTREUNG
0.44 mmol) in dry THF (3 mL) at À508C. The resulting solution was al-
lowed to warm and stirred for 16 h at ambient temperature. The solution
was concentrated to half-volume and hexane (10 mL) added. After filtra-
tion, the residue was washed with hexane (2”3 mL) and dried at 408C
overnight in vacuum to give the Rh complex as a yellow powder in
nearly quantitative yield. 1H NMR (250 MHz, CDCl3): d=7.77 (m, 2H,
CH), 7.39 (m, 2H, CH), 6.62 (m, 2H, CH), 4.98 (brs, 2H, CH), 4.69 (brs,
2H, CH), 3.62 (m, 2H), 2.58 (m, 2H), 2.32 (m, 8H, CH2), 2.13 (m, 4H),
1.88 (m, 2H), 1.63 (dd, J=18.4 Hz, J=7.7 Hz, 6H, CH3), 1.51 (brs, 18H,
CH3), 1.21 (m, 2H), 0.38 ppm (brs, 6H, CH3); 13C NMR (63 MHz,
CDCl3): d=163.2 (d, J=12 Hz, CO), 150.5 (d, J=47 Hz, CP), 139.2 (d,
J=12 Hz, CH), 125.6 (d, J=29 Hz, CH), 115.7 (s, CH), 92.9 (brs, CH),
80.2 (s, CO), 35.6 (d, J=42 Hz, CH2), 33.2 (s, CH2), 32.5 (d, J=33 Hz,
CHP), 28.9 (s, CH3), 28.8 (s, CH2), 20.5 (brd, J=12 Hz, CH3), 16.5 ppm
(s, CH3); 31P NMR (122 MHz, CDCl3): d=45.6 ppm (brs); MS (FAB+):
m/z (%): 783 (15), 742 (30) [M+ÀBF4], 593 (65), 521 (40), 476 (100); ele-
mental analysis calcd (%) for C38H60BF4O2N2P2Rh (M=828.55): C 55.08,
H 7.30, N 3.38, P 7.48, Rh 12.42; found: C 54.36, H 7.36, N 3.91, P 7.43,
Rh 12.48.
(S)-(À)-
and
(R)-(+)-4-chloro-4,5-dihydro-3H-dinaphtho[2,1-a;1’,2’-
e]phosphepine (24): (S)-(À)- and (R)-(+)-24 were obtained from com-
mercially available enantiopure 2,2’-binaphthol (20) according to the pro-
tocols of Beller et al.[16] and Zhang et al.[18]
N,N-Dimethylaminodichlorophosphane was prepared according to Nçth
and Vetter.[37]
(S)-(À)-4-(6-tert-Butoxypyrid-2-yl)-4,5-dihydro-3H-dinaphtho[2,1-a;1’,2’-
e]phosphepine ((S)-(À)-3a):
A solution of nBuLi (1.6m in hexane,
2.06 mL, 3.30 mmol) was added dropwise to a solution of 2-bromo-6-tert-
butoxypyridine (0.50 g, 2.2 mmol) in THF (10 mL) with stirring at
À788C. The reaction mixture was stirred for 15 min at À788C, allowed to
warm to ambient temperature and stirred for 1.5 h. Then the solution
was cooled to À1008C and a solution of chlorophosphepine 24[16a,c]
(1.14 g, 3.30 mmol) in THF (10 mL) was slowly added. The solution was
allowed to warm to room temperature and stirred overnight. Then meth-
anol (5 mL) was added. The mixture was concentrated in vacuum to give
the crude product (S)-(À)-3a. It could be converted into its BH3 adduct
25 and purified by column chromatography (Rf =0.55, hexane/ethyl ace-
tate 5/1). 1H NMR (CDCl3) d=1.49 (s, 9H, CH3), 0.81–1.93 (brm, 3H,
BH3), 3.38–3.42 (m, 2H, CH2), 3.44 (m, 1H, CH2), 4.22 (dd, 1H, J=
13 Hz, J=18 Hz, CH2), 7.05 (m, 1H, CH), 7.23 (dd, J=8 Hz, J=1 Hz,
1H, CH), 7.47–8.39 ppm (m, 10H, CH); 13C NMR (CDCl3) d=27.3 (d,
J=35 Hz, CH2), 28.5 (CH3), 32.8 (d, J=29 Hz, CH2) 80.6 (CO), 115.9,
121.3, 125.9–129.3, 132.6, 133.5–134.1, 139.0, 149.2, 150.0 (s), 163.5 ppm
(d, J=13 Hz); 31P NMR (CDCl3) d=48.0 ppm (s); MS (EI): m/z (%):
475 [M+] (7), 461 [M+ÀBH3] (12), 405 [M+ÀBH3ÀtBu] (100), 265 (30).
[Rh
A
A
G
and ligand 1b (0.011 g, 0.536 mmol) in dry CDCl3 was mixed and stored
in an NMR tube for 1 h at ambient temperature. The resulting brownish
solution was used for NMR experiments. 1H NMR (500 MHz, CDCl3):
d=11.75 (br, 2H, NH, OH), 7.75 (brm, 1H, CH), 7.59 (brm, 1H, CH),
7.19 (br, 1H, CH), 6.77 (brm, 2H, CH), 6.68 (m, 1H, CH), 5.30–4.50 (br,
4H, CH), 3.23 (brm, 1H, CH), 3.08 (brm, 1H, CH), 2.60–1.55 (several
m, 18H, CH2, CH), 1.51 (dd, J=19.5 Hz, J=7.2 Hz, 6H, CH3), 0.74 (dd,
J=15.0 Hz, J=7.0 Hz, 3H, CH3), 0.11 ppm (dd, J=14.0 Hz, J=7.0 Hz,
3H, CH3); 31P NMR (122 MHz, CDCl3): d=54.5 (dd, J=146.2 Hz, J=
36.4 Hz), 50.7 ppm (dd, J=138.6 Hz, J=36.4 Hz).
[PtCl
ACHTREUNG
The phosphane was liberated from BH3 adduct 25 by treatment with
DABCO in THF under reflux and subsequent chromatography on a
short column of Celite (0.52 g, 51%). M.p. 89.5–90.58C; [a]=À1298 (c=
0.5, CHCl3); 1H NMR (CDCl3) d=7.93 (d, J=8 Hz, 2H, CH), 7.67 (d,
J=8 Hz, 1H, CH), 7.57 (d, J=8 Hz, 1H, CH), 7.32–7.45 (m, 3H, CH),
7.15–7.26 (m, 4H, CH), 6.85 (m, 1H, CH), 6.78 (d, 1H, J=8 Hz, CH),
6.45 (d, J=8 Hz, 1H, CH), 3.11 (dd, 1H, J=12 Hz, J=15 Hz, CH2),
2.93–3.06 (m, 2H, CH2), 2.78 (dd, J=12 Hz, J=13 Hz, 1H, CH2),
1.38 ppm (s, 9H, CH3), 13C NMR (CDCl3) d=163.5 (d, J=7 Hz), 160.2
(s), 148.2, 137.9, 125.0–128.7, 119.0, 111.7, 79.7 (CO), 31.4 (d, J=23 Hz,
CH2), 29.0 (d, J=14 Hz, CH2), 28.5 ppm (CH3); 31P NMR (CDCl3): d=
13.4 ppm (s); MS (EI): m/z (%): 406 [M+] (1), 405 [M+ÀtBu] (3), 282
(3 mL) was added dropwise to
a
ACHTREUNG
0.282 mmol) in dry CH2Cl2 (3 mL) at ambient temperature. After 20 min
of stirring, the resulting suspension was evaporated, and the residue
washed with hexane (2”3 mL) and dried in vacuum to give the Pt com-
plex in nearly quantitative yield. Due to the poor solubility of the result-
ing white powder, only
a
31P NMR spectrum could be registered.
31P NMR (122 MHZ, CD2Cl2): d=39.5 (dd, J=3700.0, 17 Hz), 39.4 ppm
(dd, J=3282.0, 17.0 Hz). MS (EI): m/z (%): 611 (61) [M+ÀCl2], 596 (15),
569 (14), 556 (37), 529 (32), 515 (30), 497 (12), 474 (14), 461 (11), 433
(15), 402 (14), 306 (26), 209 (25), 136 (12), 96 (27); elemental analysis
calcd (%) for C22H32O2N2Cl2P2Pt (M=684.43): C 38.61, H 4.71, N 4.09,
Cl 10.36; found: C 37.86, H 4.33, N 3.77, Cl 11.15.
Chem. Eur. J. 2007, 13, 5896 – 5907
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5905

B. Breit, A. Bçrner et al.
[PdCl2
A
Z. Freixa, P. W. N. M. van Leeuwen, J. N. H. Reek, Organometallics
2006, 25, 954–960.
(10 mL) was added dropwise to
a
ACHTREUNG
0.14 mmol) in dry CH2Cl2 (10 mL) at ambient temperature. After 20 min
of stirring, the resulting suspension was evaporated, and the residue
washed with hexane (2”3 mL) and dried in vacuum to give the Pd com-
plex in nearly quantitative yield. Due to the poor solubility of the result-
[6] X.-B. Jiang, L. Lefort, P. E. Goudriaan, A. H. M. de Vries,
P. W. N. M. van Leeuwen, J. G. de Vries, J. N. H. Reek, Angew.
Chem. 2006, 118, 1245–1249; Angew. Chem. Int. Ed. 2006, 45, 1223–
1227; J. M. Takacs, D. S. Reddy, S. A. Moteki, D. Wu, H. Palencia, J.
Am. Chem. Soc. 2004, 126, 4494–4495; J. M. Atkins, S. A. Moteki,
S. G. DiMagno, J. M. Takacs, Org. Lett. 2006, 8, 2759–2762.
[7] For reviews, see: M. J. Wilkinson, P. W. N. M. van Leeuwen, J. N. H.
Reek, Org. Biomol. Chem. 2005, 3, 2371–2383; S. J. Reyes, K. Bur-
gess, Chem. Soc. Rev. 2006, 35, 416–423; A. W. Kleij, J. N. H. Reek,
Chem. Eur. J. 2006, 12, 4218–4272.
[8] a) B. Breit, W. Seiche, J. Am. Chem. Soc. 2003, 125, 6608–6609;
b) B. Breit, W. Seiche, Pure Appl. Chem. 2006, 78, 249–256; c) W.
Seiche, A. Schuschkowski, B. Breit, Adv. Synth. Catal. 2005, 347,
1488–1494; d) F. Chevallier, B. Breit, Angew. Chem. 2006, 118,
1629–1632; Angew. Chem. Int. Ed. 2006, 45, 1599–1602.
[9] B. Breit, W. Seiche, Angew. Chem. 2005, 117, 1666–1669; Angew.
Chem. Int. Ed. 2005, 44, 1640–1643; B. Breit, Angew. Chem. 2005,
117, 6976–6978; Angew. Chem. Int. Ed. 2005, 44, 6816–6825.
[10] M. Weis, C. Walloch, W. Seiche, B. Breit, J. Am. Chem. Soc. 2006,
128, 4188–4189.
ing white powder, only
a
31P NMR spectrum could be registered.
31P NMR (122 MHz, CD2Cl2): d=59.5, 59.3; elemental analysis calcd (%)
for C22H32O2N2Cl2P2Pd (M=595.78): C 44.35, H 5.41, N 4.70, Cl 11.90;
found: C 43.11, H 5.21, N 4.39, Cl 14.01.
trans-[Rh(CO)Cl
A
A
solution of [{Rh(CO)Cl2}2] (0.025 g,
0.07 mmol) in dry CH2Cl2 (2 mL) was added to a solution of ligand 1b
(0.053 g, 0.26 mmol) in dry CH2Cl2 (5 mL) at ambient temperature. After
2 h of stirring, the resulting red solution was evaporated and the residue
washed with hexane (2”3 mL) and dried in vacuum to give the Rh com-
plex in nearly quantitative yield. 1H NMR (250 MHz, CDCl3): d=10.49
(brs, 2H, NH), 7.37 (m, 2H, CH), 6.55 (m, 4H, CH), 2.87 (m, 4H), 2.30
(m, 2H), 2.09 (m, 2H), 1.50 (dd, J=17.1 Hz, J=9.7 Hz, 6H, CH3), 1.48
(m, 2H), 1.18 (m, 2H), 1.00 ppm (dd, J=14.6 Hz, J=7.3 Hz, 6H, CH3);
31P NMR (122 MHz, CDCl3): d=51.4 ppm (brd, J=124.6 Hz); elemental
analysis calcd (%) for C23H32O3N2ClP2Rh (M=584.82): C 47.24, H 5.52,
N 4.79, Cl 6.06 P 10.59; found: C 46.98, H 5.46, N 4.51, Cl 6.06 P 10.62.
General procedure for enantioselective hydrogenation: All hydrogena-
tion reactions were carried out under isothermal (258C) and isobaric
(1 bar) conditions in an automatically registrating hydrogenation device.
[11] For reviews, see: T. P. Clark, C. R. Landis, Tetrahedron: Asymmetry
2004, 15, 2123–2213; J. Holz, M.-N. Gensow, O. Zayas, A. Bçrner,
Current Org. Chem. 2007, 11, 61–106.
[12] a) M. J. Burk, J. E. Feaster, R. L. Harlow, Organometallics 1990, 9,
2653–2655; b) M. J. Burk, J. E. Feaster, R. L. Harlow, Tetrahedron:
Asymmetry 1991, 2, 569–592; c) M. J. Burk, J. Am. Chem. Soc. 1991,
113, 8518–8519; d) M. J. Burk, J. E. Feaster, W. A. Nugent, R. L.
Harlow, J. Am. Chem. Soc. 1993, 115, 10125–10138.
Precatalysts were prepared by mixing [Rh(cod)2]BF4 (1–2 mol%) and li-
A
gands (2–4 mol%) under an argon atmosphere prior to hydrogenation.
Anhydrous CH2Cl2 or MeOH (7.5 mL) was added and the solution
stirred for 20 min at ambient temperature. Then the prochiral substrate
(0.5 mmol) was added and, after replacement of argon by hydrogen, hy-
drogenation was started.
[13] a) F. Guillen, J.-C. Fiaud, Tetrahedron Lett. 1999, 40, 2939–2942;
b) F. Guillen, M. Rivard, M. Toffano, J.-C. Legros, J.-C. Fiaud, Tetra-
hedron 2002, 58, 5895–5904.
[14] N. V. Dubrovina, H. Jiao, V. I. Tararov, A. Spannenberg, R. Kadyr-
ov, A. Monsees, A. Christansen, A. Bçrner, Eur. J. Org. Chem. 2006,
3412–3420.
Acknowledgement
[15] S. Gladiali, A. Dore, D. Fabbri, O. De Lucci, M. Manassero, Tetrahe-
dron: Asymmetry 1994, 5, 511–514.
The authors are grateful to BASF AG (Ludwigshafen) for financial sup-
port of this work. We thank Dr. C. Fischer and Mrs. S. Buchholz for the
analysis of the chiral hydrogenation products.
[16] a) K. Junge, G. Oehme, A. Monsees, T. Riermeier, U. Dingerdissen,
M Beller, Tetrahedron Lett. 2002, 43, 4977–4980; b) K. Junge, B.
Hagemann, S. Enthaler, G. Oehme, M. Michalik, A. Monsees, T.
Riermeier, U. Dingerdissen, M. Beller, Angew. Chem. 2004, 116,
5176–5179; Angew. Chem. Int. Ed. 2004, 43, 5066–5069; c) K.
Junge, B. Hagemann, S. Enthaler, A. Spannenberg, M. Michalik, G.
Oehme, A. Monsees, T. Riermeier, M. Beller, M. Tetrahedron:
Asymmetry 2004, 15, 2621–2631; d) B. Hagemann, K. Junge, S. En-
thaler, M. Michalik, T. Riermeier, A. Monsees, B. Beller, Adv.
Synth. Catal. 2005, 347, 1978–1986; e) S. Enthaler, B. Hagemann, K.
Junge, G. Erre, M. Beller, Eur. J. Org. Chem. 2006, 2912–2917.
[17] For other applications of phosphepines, see: J. E. Wilson, G. C. Fu,
Angew. Chem. 2006, 118, 1454–1457; Angew. Chem. Int. Ed. 2006,
45, 1426–1429.
[1] J. M. Brown in Comprehensive Asymmetric Catalysis, Vol. I (Eds.:
E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer, Berlin, 1999,
pp. 121–182; T. Ohkuma, M. Kitamura, R. Noyori in Catalytic
Asymmetric Synthesis (Ed.: I. Ojima), Wiley, New York, 2000,
pp. 1–110; H.-U. Blaser, B. Pugin, F. Spindler, J. Mol. Catal. A 2005,
231, 1–20.
[2] For recent reviews, see: M. DiØguez, O. Pàmies, A. Ruiz, Y. Diaz, S.
Castillón, C. Claver, Coord. Chem. Rev. 2004, 248, 2165–2192; T. T.-
L. Au-Yueng, A. S. C. Chan, Coord. Chem. Rev. 2004, 248, 2151–
2164; P. Barbara, C. Bianchini, G. Giambastiani, S. L. Parisel,
Coord. Chem. Rev. 2004, 248, 2131–2150; W. Tang, X. Zhang,
Chem. Rev. 2003, 103, 3029–3069.
[3] For reviews, see: F. Lagasse, H. B. Kagan, Chem. Pharm. Bull. 2000,
48, 315–324; I. V. Komarov, A. Bçrner, Angew. Chem. 2001, 113,
1237–1239; Angew. Chem. Int. Ed. 2001, 40, 1197–1200; T. Jerphag-
non, J.-L. Renaud, C. Bruneau, Tetrahedron: Asymmetry 2004, 15,
2101–2111.
[4] For some recent results, see: B. Zhao, Z. Wang, K. Ding, Adv.
Synth. Catal. 2006, 348, 1049–1057; Q.-H. Zeng, X.-P. Hu, Z.-C.
Duan, X.-M. Liang, Z. Zheng, J. Org. Chem. 2006, 71, 393–396; H.
Bernsmann, M. van den Berg, R. Hoen, A. J. Minnaard, G. Mehler,
M. T. Reetz, J. G. de Vries, B. L. Feringa, J. Org. Chem. 2005, 70,
943–951; Y. Liu, K. Ding, J. Am. Chem. Soc. 2005, 127, 10488–
10489; S. Wu, W. Zhang, Z. Zhang, X. Zhang, Org. Lett. 2004, 6,
3565–3567.
[18] D. Xiao, Z. Zhang, X. Zhang, Org. Lett. 1999, 1, 1679–1681.
[19] L. Shi, X. Wang, C. A. Sandoval, M. Li, Q. Qi, Z. Li, K. Ding,
Angew. Chem. 2006, 118, 4214–4218; Angew. Chem. Int. Ed. 2006,
45, 4108–4112; P. A. Duckmanton, A. J. Blake, J. B. Love, Inorg.
Chem. 2005, 44, 7708–7710.
[20] IR spectroscopic investigations in solution were not successful due
to fast decomposition of the free ligands in aerobic solvents.
[21] a) C. B. Fischer, H. Steininger, D. S. Stephenson, H. Zipse, J. Phys.
Org. Chem. 2005, 18, 901–907; b) P. L. Wash, E. Maverick, J. Chief-
ari, D. A. Lightner, J. Am. Chem. Soc. 1997, 119, 3802–3806; c) M.
Gallant, M. T. Phan Viet, J. D. Wuest, J. Am. Chem. Soc. 1991, 113,
721–723.
[22] The 31P NMR spectrum of [Rh
N
N
broad singlet at d=49.7 ppm indicating equivalence of the two phos-
phorus atoms due to cleavage of the hydrogen bonds.
[5] Y. Liu, C. A. Sandoval, Y. Yamaguchi, X. Zhang, Z. Wang, K. Kato,
K. Ding, J. Am. Chem. Soc. 2006, 128, 14212–14213; L. K. Knight,
5906
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 5896 – 5907

Self-Assembling Rhodium Phosphane Catalysts
FULL PAPER
[23] In CD3OD or CDCl3 some minor species were found with resonan-
ces between d=À9 and À15 ppm. This may indicate the formation
of P,N-coordinated mononuclear or binuclear species.
y, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyen-
gar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.
Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O.
Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Fores-
man, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Na-
nayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wallingford,
CT, 2004.
[24] In CDCl3 no P,P coupling was observed.
[25] These data correlate well with those of related cis and trans Pt com-
plexes: D. W. W. Anderson, E. A. V. Ebsworth, D. W. H. Rankin, J.
Chem. Soc. Dalton 1973, 2370–2373.
[26] P. Beak, Acc. Chem. Res. 1977, 10, 186–192, and references therein;
A. Dkhissi, R. Ramaekers, L. Houben, L. Adamowicz, G. Maes,
Chem. Phys. Lett. 2000, 331, 553–560; L. Forlani, G. Cristoni, C.
Boga, P. E. Todesco, E. Del Vecchio, S. Selva, M. Monari, ARKI-
VOC 2002, 11, 198–215; J. A. Frey, S. Leutwyler, Chimia 2005, 59,
511–516.
[27] M. Piacenza, S. Grimme, J. Comput. Chem. 2004, 25, 83–98.
[28] P.-T. Chou, C-Y. Wie, F-T. Hung, J. Phys. Chem. B 1997, 101, 9119–
9126.
[29] M. W. Wong, K. B. Wiberg, M. J. Frisch, J. Am. Chem. Soc. 1992,
114, 1645–1652, and references therein.
[34] In general such polymeric catalysts display diminished reactivity,
long reaction times and sometimes elevated hydrogen pressure are
required.[19]
[35] J. Sakaki, W. B. Schweizer, D. Seebach, Helv. Chim. Acta 1993, 76,
2654–2665.
[30] a) T. Gronneberg, K. Undheim, Tetrahedron Lett. 1972, 13, 3193–
3196; b) A. Maquestian, Y. van Haverbeke, C. de Meyer, A. R. Ka-
tritzky, M. J. Cook, A. D. Page, Can. J. Chem. 1975, 53, 490–496.
[31] P. Beak, F. S. Fry, J. Am. Chem. Soc. 1973, 95, 1700–1702.
[32] a) M. J. Nowak, L. Lapinski, J. Fulara, A. Les, L. Adamowicz, J.
Phys. Chem. 1992, 96, 1562–1569; b) L. D. Hatherley, R. D. Brown,
P. D. Godfrey, A. P. Pierlot, W. Caminati, D. Damiani, S. Melandri,
L. B. Favero, J. Phys. Chem. 1993, 97, 46–51.
[36] S. Choppin, F. Gros, Y. Fort, Org. Lett. 2000, 2, 803–805.
[37] H. Nçth, H.-J. Vetter, Chem. Ber. 1963, 96, 1109–1118.
Received: November 10, 2006
Revised: February 6, 2007
Published online: April 11, 2007
[33] Gaussian03, RevisionC.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomer-
Chem. Eur. J. 2007, 13, 5896 – 5907
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5907