Asymmetric hydrogenation with highly active IndolPhos-Rh catalysts: Kinetics and reaction mechanism

Article abstract of DOI:10.1002/chem.200903476

The mechanism of the IndolPhos-Rh-catalyzed asymmetric hydrogenation of prochiral olefins has been investigated by means of X-ray crystal structure determination, kinetic measurements, high-pressure NMR spectroscopy, and DFT calculations. The mechanistic study indicates that the reaction follows an unsaturate/dihydride mechanism according to Michaelis Menten kinetics. A large value of K_M (K_M = 5.01 ± 0.16 M) is obtained, which indicates that the Rh-solvate complex is the catalyst resting state, which has been observed by high-pressure NMR spectroscopy. DFT calculations on the substrate-catalyst complexes, which are undetectable by experimental means, suggest that the major substrate-catalyst complex leads to the product. Such a mechanism is in accordance with previous studies on the mechanism of asymmetric hydrogenation reactions with C₁-symmetric heteroditopic and monodentate ligands.

Full text of DOI:10.1002/chem.200903476

FULL PAPER
DOI: 10.1002/chem.200903476
Asymmetric Hydrogenation with Highly Active IndolPhos–Rh Catalysts:
Kinetics and Reaction ACTHNUGRTENUNGMechanism
Jeroen Wassenaar,[a] Mark Kuil,[b] Martin Lutz,[c] Anthony L. Spek,[c] and
Joost N. H. Reek*[a]
Abstract: The mechanism of the Indol-
Menten kinetics. A large value of KM
(KM =5.01ꢀ0.16m) is obtained, which
indicates that the Rh–solvate complex
is the catalyst resting state, which has
been observed by high-pressure NMR
spectroscopy. DFT calculations on the
Phos–Rh-catalyzed asymmetric hydro-
genation of prochiral olefins has been
investigated by means of X-ray crystal
structure determination, kinetic meas-
urements, high-pressure NMR spec-
troscopy, and DFT calculations. The
mechanistic study indicates that the re-
action follows an unsaturate/dihydride
mechanism according to Michaelis–
substrate–catalyst complexes, which are
undetectable by experimental means,
suggest that the major substrate–cata-
lyst complex leads to the product. Such
a mechanism is in accordance with pre-
vious studies on the mechanism of
asymmetric hydrogenation reactions
with C1-symmetric heteroditopic and
monodentate ligands.
Keywords: asymmetric catalysis
·
hydrogenation · kinetics · reaction
mechanisms · rhodium
Introduction
and is still one of the most important industrial processes in
the production of chiral fine chemical building blocks.[2] The
first catalysts were derived from cationic rhodium and chiral
mono- or bidentate phosphorous ligands, which play a pivo-
tal role in obtaining high selectivities, introduced by
Knowles and Kagan in the early 1970s.[3] Over the subse-
quent decades more than three thousand chiral phosphorous
ligands have been developed guided by an increasing under-
standing of the reaction mechanism.[4]
The amplification of chirality by means of asymmetric catal-
ysis is an elegant and useful concept in the synthesis of com-
plex structures, expressing biological activity relevant to
pharmaceutical use.[1] Homogeneous transition-metal cataly-
sis has emerged in the past four decades as one of the most
efficient and powerful tools to transfer chirality from cata-
lyst to product. Asymmetric hydrogenation (AH) of olefins
was found as one of the first catalytic asymmetric reactions,
In the seminal work by Halpern[5] and Brown[6] it was pro-
posed that the enantioselectivity in the AH of olefins, in
particular, N-acylated dehydroamino esters, is determined
by the oxidative addition of dihydrogen to one of the diaste-
reomeric catalyst–substrate complexes 1B (Scheme 1, top).
The less stable diastereomer is more reactive and leads to
the product. This so-called “unsaturate/dihydride” mecha-
nism is shown to be operative for C2-symmetric diphos-
phines, such as 2,2’-bis(diphenylphosphino)-1,1’-binaphthyl
(BINAP), ethylenebis[(2-methoxyphenyl)phenylphosphine]
(DIPAMP), and 2,3-bis(diphenylphosphino)butane (CHIR-
APHOS). More recently, an alternative mechanism has
been proposed by Gridnev and Imamoto for tBu-BisP*–Rh
catalysts, which involves oxidative addition of H2 to the sol-
vate complex A prior to alkene coordination (“dihydride/
unsaturate” mechanism; Scheme 1, bottom).[7] The enantio-
selectivity is determined in the migratory insertion of one of
the hydrides into the double bond of the substrate. In this
case, two possible diastereomeric substrate–RhH2 complexes
[a] J. Wassenaar, Prof. Dr. J. N. H. Reek
Supramolecular & Homogeneous Catalysis
van ꢀt Hoff Institute for Molecular Sciencesꢁ
University of Amsterdam, Nieuwe Achtergracht 166
1018 WV Amsterdam (The Netherlands)
Fax : (+31) 20 5255604
E-mail: J.N.H.Reek@uva.nl
[b] Dr. M. Kuil
BASF Nederland B.V. Catalysts
Strijkviertel 67, 3454 PK De Meern (The Netherlands)
Current address: Solvay Pharmaceuticals
[c] Dr. M. Lutz, Prof. Dr. A. L. Spek
Crystal and Structural Chemistry
Bijvoet Centre for Biomolecular Research
Faculty of Science, Utrecht University
Padualaan 8, 3584 CH Utrecht (The Netherlands)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200903476.
Chem. Eur. J. 2010, 16, 6509 – 6517
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6509

and one may wonder whether the difference in the trans in-
fluence between phosphine and phosphoramidite allows for
such effects. Our laboratory reported the synthesis of phos-
phine–phosphoramidite IndolPhos (3) and its use as a ligand
in highly selective asymmetric hydrogenations, hydroformy-
lations, and allylic alkylations.[14] In this report, we investi-
gate the mechanism of the IndolPhos–Rh-catalyzed AH to
answer the question of which mechanism is operable, and if
specific substrate binding (lock-and-key mechanism) is
indeed achieved. The latter will be substantiated by X-ray
crystal structure determination, kinetic studies, high-pres-
sure NMR spectroscopy, and DFT calculations.
Results and Discussion
Ligand synthesis: The IndolPhos ligands 3a–f are synthe-
sized according to the previously reported procedures out-
lined in Scheme 2.[14a–b] The 2-lithioindole, generated by in
Scheme 1. Unsaturate/dihydride mechanism for the AH of N-acylated de-
hydroamino esters (P-P=(R,R)-DIPAMP, (R,R)-CHIRAPHOS, or (R)-
BINAP; s=solvent) (top). Dihydride/unsaturated mechanism (P-P=
(R,R)-tBu-BisP*) (bottom).
2C may be formed. Conversely to the unsatuated/dihydride
mechanism, the most stable diastereomer leads to the prod-
uct. In terms of efficiency, the latter mechanism seems at-
tractive as most of the catalyst is involved in the turnover of
substrates to products, whereas in the unsaturated/dihydride
mechanism only a small fraction of the catalyst participates
in turnover, which gives rise to lower reaction rates when
the equilibrium between the diastereomeric substrate–cata-
lyst complexes is slow. In addition to these two general
mechanisms, highly functionalized ligands can give rise to al-
ternative pathways.[8]
Scheme 2. Synthesis of IndolPhos ligands 3a–f. (S)-BINOL-PCI=(S)-
(ꢁ)-1,1-bi(2-naphthol)phosphorochloridite.
situ protection with CO2 of the indole NH and subsequent
lithiation with tBuLi, is reacted with the appropriate phos-
phine chloride. Condensation of the thus obtained indolyl-
phosphines with (S)-bisnaphthol phosphorochloridites gives
IndolPhos ligands in good to excellent yield.
Heteroditopic C1-symmetric ligands enable desymmetriza-
tion of the coordination sphere by differences in the trans
influence of the donor atoms. As a result, specific substrate
coordination is feasible and the exclusion of one diastereo-
mer (1B or 2C) can be achieved. Evans et al. demonstrated
this concept by using a P/S mixed ligand.[9] The mechanism
in that case follows the unsaturate/dihydride catalytic cycle;
however, the stereochemistry of the product suggests that
the major diastereomer leads to the product (lock-and-key
mechanism). Mechanistic studies by Reetz et al. also indi-
cate a similar mechanism for monodentate phosphite li-
gands.[10,11]
Hybrid phosphine–phosphoramidite ligands, which also
feature heteroditopicity and C1 symmetry, have been suc-
cessfully employed to generate highly active and selective
catalysts for the AH of functionalized alkenes.[12,13] It has
been suggested that these ligands too enable specific sub-
strate binding; however, up to now no detailed mechanistic
studies have been reported substantiating this proposition
X-ray crystal structure: Ligand 3e was reacted with di-
chloro-1,5-cyclooctadiene rhodium dimer, and treated subse-
quently with silver tetrafluoroborate to obtain complex 4 of
(cod)]BF4 (cod=1,5-cyclooctadiene).[14a]
the formula [RhACTHNUTRGENN(UG 3e)ACHTUNGTRENNGUN
Crystals suitable for single-crystal X-ray diffraction were ob-
tained as red cubes by slow diffusion of hexane into a di-
chloromethane solution. The absolute structure of 4 in the
noncentrosymmetric space group P212121 was reliably deter-
mined by using the Flack parameter (see the Experimental
Section). Top and side views of displacement ellipsoid plots
of complex 4 are shown in Figures 1 and 2, respectively. Rel-
evant bond lengths and angles are listed in the caption of
Figure 1.
As expected, complex 4 exhibits a square-planar geome-
try, placing the substituents on phosphorous above and
ꢁ
below the coordination plane. The Rh(1) P(1) bond length
ꢁ
is 0.07 ꢃ shorter relative to the Rh(1) P(2) bond, which in-
dicates significant p-backdonation in the case of the phos-
6510
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6509 – 6517

Asymmetric Hydrogenation with Highly Active IndolPhos–Rh Catalysts
FULL PAPER
The view presented in Figure 2 allows for the construction
of a quadrant model. On the right side, intermediate steric
crowding is encountered from the substituents on the phos-
phine. The upper left quadrant exhibits most crowding due
to the chiral bisnaphthol moiety. Consequently, the lower
left quadrant is most accessible to the bulk solution. Fur-
thermore, it is noteworthy that the aromatic six-membered
ring of the indole backbone in part enforces the twisted con-
formation of the bisnaphthol and further rigidifies the struc-
ture. This element of ligand design may be vital to obtain
the high selectivities reported in this paper.
Scope of the IndolPhos–Rh-catalyzed hydrogenation:
Table 1 summarizes the full substrate scope of the Indol-
Phos–Rh catalysts. High efficiencies are obtained for N-
Figure 1. Top view of displacement ellipsoid plot (50% probability level)
of complex 4. Hydrogen atoms, solvent molecules, and the tetrafluorobo-
rate counteranion are omitted for clarity. Selected bond lengths [ꢃ] and
Table 1. Scope of IndolPhos–Rh-catalyzed AH.[a]
ꢁ
ꢁ
angles [8]: Rh(1) P(1): 2.2152(7), Rh(1) P(2): 2.2828(7); P(1)-Rh(1)-
P(2): 84.75(3).
Entry Ligand
Substrate
Conv. [%] ee [%] (configuration)
1
2
3 f
3 f
100
100
97 (R)
97 (R)
3
4
3b
3 f
100
100
94 (S)
99 (R)
5
3a
38
87 (nd)[b]
6
3b
3b
3b
100
100
100
98 (S)
98 (S)
89 (S)
7[c]
8
9
3 f
3 f
45
78 (R)
87 (S)
10
100
11
3a
100
55 (S)
[a] Reactions were performed in CH2Cl2, Rh/L 1:1.1, Rh/substrate 1:100,
10 bar of H2, at 258C for 16 h by using [Rh(nbd)2]BF4 as the metal pre-
AHCTUNGTRENNUNG
cursor. [b] nd=not determined. [c] T=ꢁ408C.
Figure 2. Side view of displacement ellipsoid plot (50% probability level)
of complex 4. Hydrogen atoms, a solvent molecule, the tetrafluoroborate
counteranion, and COD are omitted for clarity (top). Quadrant diagram,
which indicates the spatial distribution of the steric bulk. Darker regions
represent a greater amount of steric crowding (bottom).
acyl-protected enamides (Table 1, entries 1–5), which allow
for the synthesis of protected a- and b-amino acids, and op-
tically active arylamines.[14d] Also double bonds that are not
part of an enamide, are hydrogenated efficiently with high
enantioselectivity (entries 6–9). In addition to dimethylitaco-
nate, 2-hydroxymethylacrylates giving Roche ester deriva-
tives,[15] which serve as valuable building blocks in natural
product synthesis, are hydrogenated in high yield and enan-
tioselectivity (entries 7 and 8).[14b] The functional-group tol-
erance of the catalytic system is illustrated by the hydroge-
nation of a-methylcinnamic acid in good enantiomeric
excess (ee) (entry 9). However, the conversion is decreased
phoramidite. Alternatively, the shorter bond length may be
explained by less steric crowding in the case of the phos-
phoramidite, as the substituents point away from the metal
compared to the phosphine. The bite angle of 858 is small
relative to most chiral bidentate phosphorous ligands. This
may explain the high selectivity for the branched aldehyde
in the rhodium-catalyzed hydroformylation of styrene, re-
ported earlier.[14a]
Chem. Eur. J. 2010, 16, 6509 – 6517
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6511

J. N. H. Reek et al.
and gas-uptake measurements suggest deactivation of the
catalyst over time. Finally, enol- and enamido phosphonate
esters are hydrogenated to the corresponding hydroxy- and
amino phosphonates quantitatively in moderate to good ee
(entries 10 and 11).[14d] For most substrates, ligands contain-
ing an isopropylphosphine give the highest ee values.
Kinetics: To investigate the mechanism of the IndolPhos–
Rh AH, a kinetic study was carried out. These kinetic ex-
periments were carried out in the AMTEC SPR16 consist-
ing of 16 parallel reactors equipped with temperature and
pressure sensors, and a mass flow controller, allowing the re-
action rates to be determined from gas-uptake profiles. We
focused our studies on the AH of methyl 2-acetamidoacry-
late (MAA) by using the catalyst generated in situ from
Figure 4. Ln
(nbd)]BF4 at S/C ratios between 1250–2500 and 10 bar H2 ([olefin] in m,
TOF in molmolꢁ1 minꢁ1).
ACHTUNGTRNE[UNGN olefin]/lnACHTUNTRGEG(NNUN TOF) plot for the AH of MAA by [RhACHTUNGTRENNUNG(3 f)-
AHCTUNGTRENNUNG
ligand 3 f and [RhACHTUNGTRENNUNG(nbd)2]BF4 (nbd=2,5-norbornadiene).
Initial rate-determination experiments revealed the high
activity of the catalytic system. This required us to lower the
catalyst loading to 0.01–0.05 mol%, which resulted in com-
plete conversion after 5 min at room temperature and
10 bar H2 (see the Supporting Information). In addition,
gas-uptake profiles reveal no induction period, which sug-
gests very rapid activation of the precatalyst by hydrogena-
tion of the diolefin. By using Blackmondꢁs kinetic analy-
sis,[16] product inhibition or catalyst decomposition was ob-
served at higher conversions, which required us to deter-
mine the rate at an early stage of the reaction. Rate-deter-
mination from the gas-uptake profiles allowed us to
determine the order in the Rh catalyst from a ln[Rh]/ln-
proposed to be operable, in which the pre-equilibrium form-
ing the catalyst–substrate complex lies on the side of the
educts (vide infra). To arrive at saturation kinetics, we con-
ducted rate measurements at higher S/C ratios of 3400–5600.
Indeed, the order in the olefin concentration is lowered to
approximately 0.5, which indicates that the pre-equilibrium
is shifted (Figure 5). Unfortunately, further lowering of the
amount of catalyst resulted in loss of activity, which is tenta-
tively ascribed to very small impurities in the substrate that
poison the catalyst (note: at S/C ratios of over 10,000, the
purity of the substrate has to exceed 99.9999%).
Due to the limitations of the MAA system, we decided to
perform a full reaction progress kinetic analysis of the AH
of dimethyl itaconate (DMI) by using ligand 3b. Previous
studies by Heller and co-workers have shown that this sub-
strate shows similar behavior in asymmetric hydrogenation
relative to dehydroamino acid esters.[18] Indeed, in this case
we were able to lower the catalyst loading to an S/C ratio of
30,000, presumably because of lower levels of impurities in
this substrate. The corresponding graphical rate equation is
depicted in Figure 6.
ACHTUNGTRENNUNG(TOF) plot (Figure 3). A first order is obtained from the
slope of this plot, which is expected for a mononuclear Rh-
catalyzed AH mechanism. Absolute turnover frequencies of
over 90,000 hꢁ1 are achieved by this catalytic system.
Next, we determined the order in olefin concentration at
S/C ratios of 1250–2500 (Figure 4). The lnACTHNUGTRENNU[G olefin]/lnHCATUNGTRNE(NUGN TOF)
plot yields a straight line with slope=1, indicative of a first-
order reaction. A first-order dependency on the substrate
concentration is rarely observed for AH of olefins,[17] as the
oxidative addition of dihydrogen is proposed to be the rate-
limiting step. In these cases, Michaelis–Menten kinetics are
Figure 3. Ln[Rh]/ln
(TOF) plot for the AH of MAA by [Rh
E
G
Figure 5. LnACHTUNGTREGNU[NN olefin]/lnACHTUNRTEGNN(GUN TOF) plot for the AH of MAA by [RhACHTUNGTRENNUNG(3 f)-
at S/C ratios of 4000–8000 and 10 bar H2 ([Rh] in m, TOF in mol
ACHTUNGTERN(NUNG nbd)]BF4 at S/C ratios between 3400–5600 and 10 bar H2 ([olefin] in m,
molꢁ1 minꢁ1
.
TOF in molmolꢁ1 minꢁ1).
6512
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6509 – 6517

Asymmetric Hydrogenation with Highly Active IndolPhos–Rh Catalysts
FULL PAPER
Figure 6. Rate versus substrate concentration plot for the AH of DMI by
[Rh(3b)
Figure 8. Plot of first-order rate constants versus hydrogen pressure for
the AH of DMI by [RhACTHNUGRTENNUG(3b)ACHTUNGTRENN(UGN nbd)]BF4.
G
E
K
M =5.01ꢀ0.16m.
The plot of reaction rate versus substrate concentration
be either oxidative addition of dihydrogen or migratory in-
sertion (see Figure S-1 in the Supporting Information). The
rate equation can, under isobaric conditions (k2 =k2’ [H2]),
be written as depicted in Equation (1), and two limiting
cases can be distinguished. If the olefin concentration is
very high, the pre-equilibrium is completely shifted towards
the catalyst–substrate complex and the rate depends solely
on the rhodium (catalyst) concentration and hydrogen pres-
sure [Eq. (2)]. On the other hand, at lower olefin concentra-
tions the rate also depends linearly on the substrate concen-
tration [Eq. (3)]. The absolute values of these concentra-
tions are determined by the magnitude of KM [Eq. (4)].
shows a strong, almost linear dependency. Therefore, for
substrate concentrations below 3m, a first order approxima-
tion for the rate equation may be used. The influence of cat-
alyst concentration and hydrogen pressure were thus deter-
mined by using first-order rate constants obtained by a fit-
ting of the experimental curves to a first-order function (Fig-
ures 7 and 8). Both plots reveal, as expected, a first-order
d½Pꢂ k2½Rhꢂ ½Sꢂ
KM þ0½Sꢂ
ð1Þ
ð2Þ
ð3Þ
n ¼
¼
dt
d½Pꢂ
dt
¼ k2½Rhꢂ0 for KM ꢃ ½Sꢂ
d½Pꢂ k2½Rhꢂ0
¼
½Sꢂ ¼ kobs½Sꢂ for KM ꢄ ½Sꢂ
dt
KM
½Aꢂ½Sꢂ
½1 Bꢂ
ð4Þ
KM
¼
Figure 7. Plot of first-order rate constants versus catalyst loading for the
AH of DMI by [Rh
(3b)
N
The kinetic profiles for the hydrogenation of MAA and
dimethyl itaconate indicate that under typical conditions,
the reaction exhibits first-order behavior [Eq. (3)]. For di-
methyl itaconate, values for k2 and KM could be obtained
from the graphical rate equation by fitting the data to Equa-
tion (1) (Figure 6). The very high value of KM (KM =5.01ꢀ
0.16m) indicates that most of the catalyst is present as the
solvent complex, that is, the resting state of the catalyst. The
measured value for KM is unusually high for AH reactions,
which generally exhibit zero-order kinetics and the sub-
strate–catalyst complexes 1B are the resting state. Alterna-
tively, the mechanism may follow a dihydride/unsaturated
pathway in which the rate-determining step is the coordina-
tion of the alkene to the solvate–dihydride complex 2B or
subsequent migratory insertion (see Figure S-2 in the Sup-
dependency of the reaction on the amount of catalyst and
hydrogen pressure. For dimethyl itaconate, very high initial
turnover frequencies are obtained (>50000 hꢁ1) and turn-
over numbers up to 30,000.
The kinetics of the AH of MAA and dimethyl itaconate
by IndolPhos–Rh catalysts can best be described by Michae-
lis–Menten kinetics as was shown previously for diphosphine
systems by following the unsaturate/dihydride pathway.[5b,d,19]
In this Michaelis–Menten model, reversible coordination of
the alkene forming the diastereomeric substrate–catalyst
complexes 1B, is followed by an irreversible reaction with
dihydrogen. The irreversible rate-determining step may then
Chem. Eur. J. 2010, 16, 6509 – 6517
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6513

J. N. H. Reek et al.
porting Information). However, the inability to detect sol-
vate–dihydride species by high-pressure NMR spectroscopy
speaks against this (vide infra).
The obtained structures and energies are depicted in
Figure 9.
The calculations indicate that one complex, pro-(S)-trans-
1B, appears to be much more stable than the other three
complexes. The greater stability can be understood in terms
of perturbation from the ideal square-planar geometry pre-
ferred by RhI and the trans effect. In the case of pro-(R)-
trans-1B, the alkene is tilted out of the coordination plane
because of steric repulsion of the a,b-unsaturated ester with
the bisnaphthol moiety, forcing the methyl group of the
ester to be eclipsing with the alkene fragment. Complex
pro-(S)-cis-1B also displays out of plane coordination of the
alkene; however, in this case caused by repulsive interac-
tions between the saturated ester and the bisnaphthol frag-
ment. For pro-(R)-cis-1B, the orientation of the substrate
causes steric crowding between the a,b-unsaturated ester
and the bisnaphthol group and the phosphine–carbonyl
trans disposition is less favorable relative to the phosphora-
midite–carbonyl trans.
High-pressure NMR spectroscopy: The kinetic experiments
described above give valuable information about the rate
equation; however, it does not allow for detection of the in-
termediates in the catalytic cycle. We therefore turned to
high-pressure NMR spectroscopy to obtain structural infor-
mation about these intermediates. Also for these studies we
used the in situ generated complex [RhACHTNUGRTNENUG(3 f)ACHTUGNTREN(NUGN nbd)]BF4 (5),
which exhibits two doublets-of-doublets in the 31P NMR
spectrum at d=141 and 54 ppm with JPP =52.7 Hz. When
this complex is subjected to five bar pressure of molecular
hydrogen in CD2Cl2, very broad signals are obtained. How-
ever, when this species is reacted with alkene substrates, the
hydrogenation products were obtained with the same enan-
tioselectivity. Addition of MeCN to this ill-defined species
yields the MeCN solvate complex, [RhACHTNUGTRENNUG(3 f)AHCUTNGTREN(NUGN MeCN)2]BF4,
which indicates that the species giving broad signals is the
CD2Cl2 solvent complex. Formation of the well-defined sol-
vate complex [RhACHTUNGTRENNUNG(3 f)AHCTUNGTREN(NUNG CD3OD)2]BF4 is also achieved by
We found experimentally that the S enantiomer is ob-
tained in an excellent ee of 98% by using ligand 3b. This
result is in good agreement with the outcome of the DFT
calculations, which predict that the diastereoisomer pro-(S)-
trans-1B, leading to the S-enantiomer is the only populated
one at room temperature according to the Boltzmann distri-
bution. It is possible that alkene decoordination may occur
during the oxidative addition step as proposed by Gridnev
and Imamoto (vide supra). Therefore, a full computational
analysis of the reaction pathway is necessary to substantiate
the true origin of enantioselection. However, theory and ex-
periment are in good agreement and suggest that the major
substrate–catalyst complex leads to the product in the Indol-
Phos-catalyzed AH of dimethyl itaconate and most probably
also for other substrates coordinating in a similar fashion.
The proposed mechanism differs significantly to the Hal-
pern and Brown mechanism for C2-symmetric diphosphines,
in which the least-stable diastereomeric substrate–catalyst
complex leads to the product (vide supra). Landis and co-
workers calculated the whole reaction pathway of the rhodi-
um-catalyzed AH, by using DuPHOS as the chiral ligand.[20]
They found energy differences between the major and
minor catalyst substrate complexes of 3.7 kcalmolꢁ1 and a
total energy difference between the two pathways of
4.4 kcalmolꢁ1. In our case, the energy differences in the sub-
strate–catalyst complexes are much larger (9.4 kcalmolꢁ1) at
the same level of theory. In addition, the overall reaction ac-
tivation energy (Ea) is approximately 15.3 kcalmolꢁ1 for the
hydrogenation of DMI, which is derived from the value of
k2 by using the Arrhenius equation. Consequently, a mecha-
nism similar to the one proposed by Evans and Reetz for
C1-symmetric P/S and monodentate ligands is much more
likely to be operable for IndolPhos–Rh catalysts.[9,10]
changing the solvent to deuteromethanol, characterized by a
new set of doublets-of-doublets at d=146.4 and 78.4 ppm
with JPP =72.4 Hz. Attempts to detect the solvate–dihydride
complex 2B at low temperatures by following the protocol
by Gridnev and Imamoto were unsuccessful.[7a] Further-
more, efforts to replace solvent molecules with substrates
did not result in the formation of diastereomeric catalyst–
substrate complexes 1B, which is in line with the kinetic
profile (KM is very large).
The high-pressure NMR spectroscopic experiments are in
agreement with the kinetic data as no substrate–catalyst
complexes could be obtained. Instead, stable solvate com-
plexes could be detected, representing the major resting
state. The inability of the solvate complexes to form dihy-
drides at low temperatures indicates that an unsaturate/dihy-
dride pathway is followed. Recent contributions from Grid-
nev and Imamoto suggest that dihydride formation may be
accelerated by partial substrate coordination through the
carbonyl function.[7d] This may indeed also occur in our
system under catalytic conditions, but we are unable to
study this as no substrate–catalyst complexes can be pre-
pared.
DFT calculations: The kinetic studies in conjunction with
the high-pressure NMR spectroscopic studies discussed
above point towards an unsaturate/dihydride pathway in
which the absolute configuration of the product is deter-
mined in the oxidative addition of dihydrogen to one of the
catalyst–substrate complexes 1B. As these species cannot be
detected experimentally because of the large value of KM,
we decided to calculate the four possible diastereomeric cat-
alyst–substrate complexes 1B at a high level of theory
(B3LYP, 6-31G*/LANL2DZ) without any constraints, for
the hydrogenation of dimethyl itaconate, by using ligand 3b.
6514
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6509 – 6517

Asymmetric Hydrogenation with Highly Active IndolPhos–Rh Catalysts
FULL PAPER
Figure 9. Calculated (B3LYP, 6-31G*/LANL2DZ) structures and energies of diastereomeric catalyst–substrate complexes 1B.
Conclusion
In conclusion, the mechanistic studies on IndolPhos–Rh
catalysts reported in this paper are in line with a lock-and-
key type mechanism of enantiodiscrimination as has been
previously proposed for C1-symmetric P/S and monodentate
ligands.[9,10] However, on the basis of the current data we
cannot exclude other mechanistic pathways. Relative to the
anti-lock-and-key mechanism observed for C2-symmetric li-
gands, this mechanism is more elegant and efficient as the
major flux of the reaction proceeds through the major
isomer, that is, a higher effective catalyst concentration. For
IndolPhos–Rh catalysts, however, a large amount of the cat-
alysts exists as the solvate complex at typical conditions.
IndolPhos–Rh complexes are shown to be highly active
(TOF>90,000 hꢁ1) and enantioselective (up to 99% ee) hy-
drogenation catalysts for a broad range of prochiral alkenes,
making them suitable candidates for industrial large scale
AH processes.[21] Their facile preparation allows for easy de-
rivatization leading to a rapid construction of a small ligand
library. Kinetic studies indicate that the AH of MAA and
dimethyl itaconate follows Michaelis–Menten kinetics, ex-
hibiting a very large value for KM, which indicates that the
Rh–solvate complex is the major resting state of the cata-
lyst. Several solvate complexes have been observed by high-
pressure NMR spectroscopy and form neither substrate–cat-
alyst complexes nor solvate–dihydride species. This suggests
that an unsaturate/dihydride mechanism is followed, but we
cannot exclude a dihydride/unsaturate mechanism complete-
ly. High-level DFT calculations indicate that only one of
four diastereomeric substrate–catalyst complexes is energeti-
cally accessible, which leads to the experimentally found ab-
solute configuration of the product. We are currently inves-
tigating the full reaction pathway by DFT calculations to
substantiate these findings.
Experimental Section
Kinetic gas-uptake measurements: The experiments were carried out in
an AMTEC SPR16 consisting of 16 parallel reactors equipped with tem-
perature and pressure sensors, and a mass flow controller. The apparatus
is suited for monitoring gas-uptake profiles during the catalytic reactions.
The autoclaves were heated to 1108C and flushed with argon (22 bar)
five times. Next, the reactors were cooled to 258C and flushed again with
argon (22 bar) five times. The autoclaves were charged with the appropri-
ate amount of [RhACHTUNRGTENNG(U nbd)2HCATUNGTNERN(UGN BF4)], ligand 3, and substrate in CH2Cl2
Chem. Eur. J. 2010, 16, 6509 – 6517
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6515

J. N. H. Reek et al.
(8.00 mL) under argon. The reactors were pressurized to the desired
pressure with H2 and the pressure was kept constant during the whole re-
action. The reaction mixtures were stirred at 258C and the hydrogen
uptake was monitored and recorded for every reactor. After catalysis, the
pressure was reduced to 2.0 bar and samples (0.2 mL) were taken for
chiral GC analysis.
cules. Their contribution to the structure factors was taken into account
by using a back-Fourier transformation with the SQUEEZE routine of
the program PLATON[25], resulting in 151 electrons/unit cell. 647 param-
eters were refined with 11 restraints. R1/wR2 [I>2s(I)]: 0.0351/0.0840.
R1/wR2 [all refl.]: 0.0418/0.0871. S=1.041. Flack x-parameter:[26]
ꢁ0.024(16). Residual electron density between ꢁ0.81 and 0.93 eꢃ3. Ge-
ometry calculations and checking for higher symmetry were performed
with the PLATON program.[25]
High-pressure NMR spectroscopic experiments: In a typical experiment,
a 5 mm sapphire high-pressure NMR spectroscopic tube was filled with a
solution of [Rh
(nbd)2]BF4 (5.6 mg, 0.015 mmol), ligand 3 f (8.8 mg,
CCDC-756580 contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
0.015 mmol), and CD2Cl2 or CD3OD (0.5 mL). The tube was purged
three times with 5 bar of H2, and then pressurized with 5 bar of H2. After
vigorous manual shaking for approximately 2 min, the tube was inserted
in the NMR spectrometer.
Computational details: DFT calculations were performed by using the
Spartan ꢀ04 for windows program package,[27] employing the B3LYP func-
tional.[28] The basis set was 6–31G* for all atoms,[29] except for Rh, which
was described by an effective core potential and the associated basis set
LANL2DZ.[30] Graphics were generated by using MacPyMOL.[31]
[RhACHTUNGTRENNUNG(3 f)ACHTUNGTRENNUNG
(CD3OD)2]BF4: 1H NMR (500 MHz, CD3OD, 253 K): d=8.15 (s,
1H), 8.06 (d, J=8.3 Hz, 1H), 7.92 (d, J=8.3 Hz, 1H), 7.82–7.79 (m, 2H),
7.63 (d, J=8.0 Hz, 1H), 7.51 (d, J=7.8 Hz, 1H), 7.13–7.09 (m, 1H), 6.98
(t, J=7.6 Hz, 1H), 6.89 (d, J=8.3 Hz, 1H), 6.17 (t, J=7.5 Hz, 1H), 6.01
(t, J=1.8 Hz, 2H), 5.88 (d, J=8.4 Hz, 1H), 3.09 (s, 3H), 2.85 (m, 3H),
1.66–1.64 (m, 2H), 1.62 (s, 3H), 1.52–0.93 ppm (m, 12H); 31P NMR
(202 MHz, CD3OD): d=146.43 (dd, J=331.8, 72.6 Hz, 1P), 78.42 ppm
(dd, J=180.3, 72.3 Hz, 1P).
Acknowledgements
[Rh
lution of [RhACHTUNGTRENNUNG
ACHTUNGTRENNUNG(3b)ACHTUNGTRENNUNG
This work was supported by the NRSC-C and the European Union
(RTN Revcat MRTN-CT-2006-035866). We thank R.J. Detz for assis-
tance with the AMTEC experiments. This work was also supported in
part (M.L, A.L.S.) by the Council for the Chemical Sciences of The Neth-
erlands Organization for Scientific Research (CW-NWO).
for 30 min at RT. Dihydrogen was bubbled through the solution for
30 min. Filtration through Celite followed by evaporation of the solvent
in vacuo. Washing of the resulting orange solid with hexane (3ꢄ5 mL)
and drying in vacuo gave the characteristic broad NMR spectra for the
CH2Cl2–solvate complex. The compound was redissolved in MeCN
(5 mL), stirred for 20 min at RT, and the solvent was removed in vacuo
to give a yellow solid. The solid can be handled in air but decomposes
overnight in CDCl3 and upon storage in air for several days. Yield:
122 mg (82%); 1H NMR (500 MHz, CDCl3, 298 K): d=8.16 (d, J=
8.8 Hz, 1H), 8.02 (d, J=8.2 Hz, 1H), 7.92 (d, J=8.1 Hz, 1H), 7.76 (d, J=
8.8 Hz, 1H), 7.71 (d, J=8.8 Hz, 1H), 7.54–7.50 (m, 2H), 7.43–7.32 (m,
5H), 6.99 (t, J=7.5 Hz, 1H), 6.92 (d, J=8.8 Hz, 1H), 6.37 (t, J=7.8 Hz,
1H), 6.00 (d, J=8.5 Hz, 1H), 2.68 (2ꢄtd, J=15.3, 7.8 Hz, 2H), 2.43 (s,
3H), 2.14 (brs, 6H), 1.49–1.33 (m, 9H), 1.17 ppm (dd, J=16.9, 6.9 Hz,
3H); 13C NMR (126 MHz, CDCl3, 298 K): d=149.63 (d, J=16.0 Hz),
147.70 (d, J=5.2 Hz), 136.86 (d, J=6.9 Hz), 132.68 (s), 132.33 (s), 131.86
(s), 131.57 (s), 131.22 (s), 130.71 (s), 128.55 (d, J=3.1 Hz), 127.22 (s),
127.05 (s), 127.02 (s), 126.95 (s), 126.33 (s), 125.81 (s), 124.39 (s), 123.27
(d, J=1.9 Hz), 122.99 (s), 122.34 (s), 121.46 (s), 121.10 (s), 119.55 (s),
115.28 (s), 70.68 (s), 27.52 (d, J=8.7 Hz), 27.29 (d, J=10.7 Hz), 26.61 (s),
20.23 (s), 20.08 (d, J=4.5 Hz), 19.99 (s), 19.82 (d, J=4.5 Hz), 11.13 (s),
2.26 ppm (s); 31P NMR (202 MHz, CDCl3, 298 K): d=148.45 (dd, J=
294.5, 63.1 Hz; 1P), 71.60 ppm (dd, J=159.2, 63.0 Hz; 1P); MS (ESI)
m/z: calcd for C39H39N3O2P2Rh: 746.16 [MꢁBF4]+; found: 746.20.
[1] J. S. Carey, D. Laffan, C. Thomson, M. T. Williams, Org. Biomol.
Chem. 2006, 4, 2337–2347.
[2] a) J. G. de Vries, C. J. Elsevier, The Handbook of Homogeneous
Hydrogenation, Wiley-VCH, Weinheim, 2007; b) H. U. Blaser, B.
Pugin, F. Spindler, M. Thommen, Acc. Chem. Res. 2007, 40, 1240–
1250.
[3] T. Ohkuma, M. Kitamura, R. Noyori in New Frontiers in Asymmet-
ric Catalysis (Eds.: K. Mikami, M. Lautens), Wiley, New York, 2007,
pp. 1–32.
[4] W. Tang, X. Zhang, Chem. Rev. 2003, 103, 3029–3069.
[5] a) A. S. C. Chan, J. Halpern, J. Am. Chem. Soc. 1980, 102, 838–840;
b) A. S. C. Chan, J. J. Pluth, J. Halpern, J. Am. Chem. Soc. 1980, 102,
5952–5954; c) J. Halpern, Science 1982, 217, 401–407; d) C. R.
Landis, J. Halpern, J. Am. Chem. Soc. 1987, 109, 1746–1754.
[6] a) J. M. Brown, P. A. Chaloner, Tetrahedron Lett. 1978, 19, 1877–
1880; b) J. M. Brown, P. A. Chaloner, J. Am. Chem. Soc. 1980, 102,
3040–3048.
[7] a) I. D. Gridnev, N. Higashi, K. Asakura, T. Imamoto, J. Am. Chem.
Soc. 2000, 122, 7183–7194; b) I. D. Gridnev, T. Imamoto, Acc.
Chem. Res. 2004, 37, 633–644; c) I. D. Gridnev, T. Imamoto, Chem.
Commun. 2009, 7447–7464; d) I. D. Gridnev, T. Imamoto, G. Hoge,
M. Kouchi, H. Takahashi, J. Am. Chem. Soc. 2008, 130, 2560–2572.
[8] a) F. W. Patureau, M. Kuil, A. J. Sandee, J. N. H. Reek, Angew.
Chem. 2008, 120, 3224–3227; Angew. Chem. Int. Ed. 2008, 47, 3180–
3183; b) F. W. Patureau, S. de Boer, M. Kuil, J. Meeuwissen, P. A. R.
Breuil, M. A. Siegler, A. L. Spek, A. J. Sandee, B. de Bruin, J. N. H.
Reek, J. Am. Chem. Soc. 2009, 131, 6683–6685.
[9] D. A. Evans, F. E. Michael, J. S. Tedrow, K. R. Campos, J. Am.
Chem. Soc. 2003, 125, 3534–3543.
[10] M. T. Reetz, A. Meiswinkel, G. Mehler, K. Angermund, M. Graf, W.
Thiel, R. Mynott, D. G. Blackmond, J. Am. Chem. Soc. 2005, 127,
10305–10313.
[11] I. D. Gridnev, C. Fan, P. G. Pringle, Chem. Commun. 2007, 1319–
1321.
[12] For reviews, see: a) D. Amoroso, T. W. Graham, R. W. Guo, C. W.
Tsang, K. A. Rashid, Aldrichimica Acta 2008, 41, 15–26; b) F.
Boeda, T. Beneyton, C. Crꢅvisy, Mini-Rev. Org. Chem. 2008, 5, 96–
127.
X-ray crystal-structure determination of 4: [C55H57NO2P2RhSi2]-
ACHTUNGTRENNUNG(BF4)·CH2Cl2 +disordered solvent; Fw =1156.78; orange block; 0.25ꢄ
0.12ꢄ0.09 mm3; orthorhombic; P212121 (no. 19); a=14.1237(1), b=
15.3310(1), c=26.7687(2) ꢃ; V=5796.24(7) ꢃ3; Z=4; Dx =1.33 gcmꢁ3
;
m=0.54 mmꢁ1
.
74726 reflections were measured on a Nonius Kappa
1
CCD diffractometer with rotating anode (graphite monochromator, l=
0.71073 ꢃ) up to a resolution of (sin q/l)max =0.65 ꢃꢁ1 at a temperature
of 150(2) K. Intensities were integrated with HKL2000.[22]. An absorption
correction based on multiple measured reflections was performed by
using the program SADABS[23] (correction range 0.74–0.95). 13196 re-
flections were unique (Rint =0.041), of which 11865 were observed [I>
2s(I)]. The structure was solved with direct methods by using the pro-
gram SHELXS-97.[24] The structure was refined with SHELXL-97[24]
against F2 of all reflections. Non-hydrogen atoms were refined with ani-
sotropic displacement parameters. All hydrogen atoms were introduced
in calculated positions and refined with a riding model. The crystal struc-
ture contains ordered dichloromethane solvent molecules, which were re-
fined with full occupancies. The crystal structure also contains voids
(542 ꢃ3/unit cell) filled with severely disordered dichloromethane mole-
1
Derived parameters do not contain the contribution of the disordered
[13] a) G. Franciꢆ, F. Faraone, W. Leitner, Angew. Chem. 2000, 112,
1486–1488; Angew. Chem. Int. Ed. 2000, 39, 1428–1430; b) X. P.
solvent.
6516
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6509 – 6517

Asymmetric Hydrogenation with Highly Active IndolPhos–Rh Catalysts
FULL PAPER
Hu, Z. Zheng, Org. Lett. 2004, 6, 3585–3588; c) X. P. Hu, Z. Zheng,
Org. Lett. 2005, 7, 419–422; d) Q. H. Zeng, X. P. Hu, Z. C. Duan,
X. M. Liang, Z. Zheng, Tetrahedron: Asymmetry 2005, 16, 1233–
1238; e) J. D. Huang, X. P. Hu, Z. C. Duan, Q. H. Zeng, S. B. Yu, J.
Deng, D. Y. Wang, Z. Zheng, Org. Lett. 2006, 8, 4367–4370; f) D. Y.
Wang, X. P. Hu, J. D. Huang, J. Deng, S. B. Yu, Z. C. Duan, X. F.
Xu, Z. Zheng, Angew. Chem. 2007, 119, 7956–7959; Angew. Chem.
Int. Ed. 2007, 46, 7810–7813; g) M. Qiu, X. P. Hu, D. Y. Wang, J.
Deng, J. D. Huang, S. B. Yu, Z. C. Duan, Z. Zheng, Adv. Synth.
Catal. 2008, 350, 1413–1418; h) S. B. Yu, J.-D. Huang, D. Y. Wang,
X. P. Hu, J. Deng, Z. C. Duan, Z. Zheng, Tetrahedron: Asymmetry
2008, 19, 1862–1866; i) K. A. Vallianatou, I. D. Kostas, J. Holz, A.
Bçrner, Tetrahedron Lett. 2006, 47, 7947–7950; j) Y. Yan, X. Zhang,
J. Am. Chem. Soc. 2006, 128, 7198–7202; k) X. Zhang, B. Cao, Y.
Yan, S. Yu, B. Ji, X. Zhang, Chem. Eur. J. 2010, 16, 871–877; l) Z. C.
Duan, X. P. Hu, J. Deng, S. B. Yu, D. Y. Wang, Z. Zheng, Tetrahe-
dron: Asymmetry 2009, 20, 588–592; m) M. Qiu, D. Y. Wang, X. P.
Hu, J. D. Huang, S. B. Yu, J. Deng, Z. C. Duan, Z. Zheng, Tetrahe-
dron: Asymmetry 2009, 20, 210–213; n) S. B. Yu, X. P. Hu, J. Deng,
D. Y. Wang, Z. C. Duan, Z. Zheng, Tetrahedron: Asymmetry 2009,
20, 621–625.
1188; c) A. Preetz, H. J. Drexler, C. Fischer, Z. Dai, A. Bçrner, W.
Baumann, A. Spannenberg, R. Thede, D. Heller, Chem. Eur. J. 2008,
14, 1445–1451; d) T. Schmidt, Z. Dai, H. J. Drexler, W. Baumann, C.
Jager, D. Pfeifer, D. Heller, Chem. Eur. J. 2008, 14, 4469–4471; e) T.
Schmidt, Z. Dai, H. J. Drexler, M. Hapke, A. Preetz, D. Heller,
Chem. Asian J. 2008, 3, 1170–1180; f) T. Schmidt, H. J. Drexler, J.
Sun, Z. Dai, W. Baumann, A. Preetz, D. Heller, Adv. Synth. Catal.
2009, 351, 750–754.
[19] M. Kitamura, M. Tsukamoto, Y. Bessho, M. Yoshimura, U. Kobs, M.
Widhalm, R. Noyori, J. Am. Chem. Soc. 2002, 124, 6649–6667.
[20] a) C. R. Landis, P. Hilfenhaus, S. Feldgus, J. Am. Chem. Soc. 1999,
121, 8741–8754; b) S. Feldgus, C. R. Landis, J. Am. Chem. Soc. 2000,
122, 12714–12727; c) C. R. Landis, S. Feldgus, Angew. Chem. 2000,
112, 2985–2988; Angew. Chem. Int. Ed. 2000, 39, 2863–2866; d) S.
Feldgus, C. R. Landis, Organometallics 2001, 20, 2374–2386.
[21] H. U. Blaser, F. Spindler, M. Thommen in The Handbook of Homo-
geneous Catalysis, Vol. 3 (Eds.: J. G. de Vries, C. J. Elsevier), Wiley-
VCH, Weinheim, 2007, pp. 1279–1324.
[22] Z. Otwinowski, W. Minor in Methods in Enzymology (Eds.: C. W.
Carter Jr., R. M. Sweet), Academic Press, New York, 1997, pp. 307–
326.
[14] a) J. Wassenaar, J. N. H. Reek, Dalton Trans. 2007, 3750–3753; b) J.
Wassenaar, M. Kuil, J. N. H. Reek, Adv. Synth. Catal. 2008, 350,
1610–1614; c) J. Wassenaar, S. van Zutphen, G. Mora, P. Le Floch,
M. A. Siegler, A. L. Spek, J. N. H. Reek, Organometallics 2009, 28,
2724–2734; d) J. Wassenaar, J. N. H. Reek, J. Org. Chem. 2009, 74,
8403–8406.
[15] The name Roche ester, which is nowadays frequently used for (S)-3-
hydroxy-2-methylpropionate, should not be confused with the
Roche company. The name is derived from the university of Roches-
ter (New York, USA) where its synthesis has been described for the
first time by Herrmann and Schlessinger, see: J. L. Herrmann, R. H.
Schlessinger, Tetrahedron Lett. 1973, 14, 2429–2432.
[23] SADABS 2006/1, G. M. Sheldrick, University of Gçttingen, Gçttin-
gen, 2006.
[24] G. M. Sheldrick, Acta Crystallogr. Sect. A 2008, 64, 112–122.
[25] A. L. Spek, Acta Crystallogr. Sect. D 2009, 65, 148–155.
[26] H. D. Flack, Acta Crystallogr. Sect. A 1983, 39, 876–881.
[27] Spartan ꢀ04 for windows, Wavefunction, Inc., Irvine, California,
2004.
[28] a) S. H. Vosko, L. Wilk, M. Nusair, Can. J. Phys. 1980, 58, 1200–
1211; b) C. T. Lee, W. T. Yang, R. G. Parr, Phys. Rev. B 1988, 37,
785–789; c) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652;
d) P. J. Stephens, F. J. Devlin, C. F. Chabalowski, M. J. Frisch, J.
Phys. Chem. 1994, 98, 11623–11627.
[16] D. G. Blackmond, Angew. Chem. 2005, 117, 4374–4393; Angew.
Chem. Int. Ed. 2005, 44, 4302–4320.
[29] a) P. C. Hariharan, J. A. Pople, Theor. Chim. Acta 1973, 28, 213–
222; b) M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, M. S.
Gordon, D. J. Defrees, J. A. Pople, J. Chem. Phys. 1982, 77, 3654–
3665.
[30] P. J. Hay, W. R. J. Wadt, J. Chem. Phys. 1985, 82, 270; P. J. Hay,
W. R. J. Wadt, J. Chem. Phys. 1985, 82, 284; P. J. Hay, W. R. J. Wadt,
J. Chem. Phys. 1985, 82, 299.
[17] a) D. Heller, R. Kadyrov, M. Michalik, T. Freier, U. Schmidt, H. W.
Krause, Tetrahedron: Asymmetry 1996, 7, 3025–3035; b) H. J. Drex-
ler, A. Preetz, T. Schmidt, D. Heller in The Handbook of Homoge-
neous Hydrogenation, Vol. 1 (Eds.: J. G. de Vries, C. J. Elsevier),
Wiley-VCH, Weinheim, 2007, pp. 257–293.
[18] a) H. J. Drexler, S. L. Zhang, A. L. Sun, A. Spannenberg, A. Arrieta,
A. Preetz, D. Heller, Tetrahedron: Asymmetry 2004, 15, 2139–2150;
b) H. J. Drexler, W. Baumann, T. Schmidt, S. L. Zhang, A. L. Sun,
A. Spannenberg, C. Fischer, H. Buschmann, D. Heller, Angew.
Chem. 2005, 117, 1208–1212; Angew. Chem. Int. Ed. 2005, 44, 1184–
[31] The PyMOL Molecular Graphics System, W. L. DeLano, DeLano
Scientific LLC, Palo Alto, California, http://www.pymol.org, 2009.
Received: December 18, 2009
Published online: April 22, 2010
Chem. Eur. J. 2010, 16, 6509 – 6517
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6517