Asymmetric hydrogenation using rhodium complexes generated from mixtures of monodentate neutral and anionic phosphorus ligands

Article abstract of DOI:10.1002/chem.201202408

A series of monodentate neutral and anionic phosphorus ligands was synthesized and evaluated in the asymmetric rhodium-catalyzed hydrogenation of functionalized olefins by using either catalysts containing identical ligands or catalysts generated from mixtures of two different ligands. We expected that the combination of an anionic ligand with a neutral ligand would favor the formation of hetero over homo bis-ligand complexes due to charge repulsion. NMR spectroscopic studies confirmed that charge effects can indeed shift the equilibrium toward the hetero bis-ligand complexes. In several cases, the combination of a neutral phosphane with an anionic phosphane, one chiral and the other achiral, furnished significantly higher enantioselectivities than analogous mixtures of two neutral ligands. The best results were obtained with a mixture of an anionic phosphoramidite and a neutral phosphoric acid diester. It is supposed that in this case a hydrogen bond between the two ligands additionally stabilizes the hetero ligand combination. Charge effects and hydrogen bonding favor the formation of rhodium hetero bis-ligand complexes from mixtures of neutral and anionic monodentate phosphorus ligands. The combination of a neutral phosphoric acid diester as a hydrogen donor and an anionic phosphoramidite as a hydrogen acceptor gives very high enantioselectivities in Rh-catalyzed hydrogenation reactions, thus exceeding the ee values of the corresponding homo bis-ligand complexes (see scheme). Copyright

Full text of DOI:10.1002/chem.201202408

FULL PAPER
DOI: 10.1002/chem.201202408
Asymmetric Hydrogenation Using Rhodium Complexes Generated from
Mixtures of Monodentate Neutral and Anionic Phosphorus Ligands
Dominik J. Frank,[a, b] Axel Franzke,[a, c] and Andreas Pfaltz*[a]
Abstract:
A
series of monodentate
charge repulsion. NMR spectroscopic
studies confirmed that charge effects
can indeed shift the equilibrium toward
the hetero bis-ligand complexes. In sev-
eral cases, the combination of a neutral
phosphane with an anionic phosphane,
one chiral and the other achiral, fur-
nished significantly higher enantiose-
lectivities than analogous mixtures of
two neutral ligands. The best results
were obtained with a mixture of an
anionic phosphoramidite and a neutral
phosphoric acid diester. It is supposed
that in this case a hydrogen bond be-
tween the two ligands additionally sta-
bilizes the hetero ligand combination.
neutral and anionic phosphorus ligands
was synthesized and evaluated in the
asymmetric rhodium-catalyzed hydro-
genation of functionalized olefins by
using either catalysts containing identi-
cal ligands or catalysts generated from
mixtures of two different ligands. We
expected that the combination of an
anionic ligand with a neutral ligand
would favor the formation of hetero
over homo bis-ligand complexes due to
Keywords: asymmetric catalysis
·
hydrogenation · phosphanes · phos-
phoramidites · phosphoric acid di-
esters
Introduction
phites derived from 1,1’-bi-2-naphthol (BINOL) can induce
high enantioselectivities in Rh-catalyzed asymmetric hydro-
genation reactions. Because ligands of this type are readily
synthesized from inexpensive precursors, they soon found
many applications in asymmetric hydrogenation, including
reactions on an industrial scale.[3b]
Moreover, the work on chiral monodentate phosphorus li-
gands has led to a new approach to combinatorial-catalyst
development. By using mixtures of two different monoden-
tate ligands, Rh catalysts were generated in situ, which in
many reactions gave higher enantioselectivities than the cor-
responding homo-ligand complexes.[3b,6,7] Even mixtures of a
chiral and an achiral ligand proved effective. It may have
come as a surprise that this approach was successful, taking
into account that mixtures of homo- and hetero-ligand com-
plexes are likely to be present in the reaction solution. Al-
though excellent enantioselectivities were achieved in cer-
tain cases, the outcome of combinatorial experiments of this
kind is unpredictable and, in general, a large number of li-
gands has to be tested.
Since Dang and Kagan introduced the chiral diphosphane
1,4-bis(diphenylphosphino)butane (DIOP) in 1971,[1] biden-
tate ligands have been playing a dominant role in asymmet-
ric catalysis.[2] The success of bidentate ligands in a wide
range of enantioselective metal-catalyzed reactions led to
the generally accepted view that chiral chelate complexes
based on bidentate ligands are inherently superior catalysts
relative to complexes based on monodentate ligands. Al-
though many examples of successful applications of mono-
dentate ligands have been reported over the years, in asym-
metric hydrogenation chelate-forming bidentate ligands
indeed seemed to be essential for high enantioselectivity.
This long-standing dogma was refuted in 2000 by the re-
search groups of Feringa and co-workers,[3] Pringle and co-
workers,[4] and Reetz and Mehler,[5] who discovered that
monodentate phosphoramidites, phosphonites, and phos-
[a] Dr. D. J. Frank, Dr. A. Franzke, Prof. Dr. A. Pfaltz
Department of Chemistry
A more selective formation of a specific hetero-ligand
combination can be achieved by means of selective binding
units that promote the self-assembly of two ligands, thus re-
sulting in a supramolecular analogue of a bidentate ligand.
Successful examples of this strategy are the hydrogen-bridg-
ed bis(monophosphane) complexes developed by Breit and
Seiche[8] or the supramolecular metal-bridged systems de-
vised by the groups of Takacs et al.[9] and Reek, van Leeu-
wen, and co-workers.[10] These strategies led to more defined
catalysts than a combinatorial approach based on ligand
mixtures. However, the incorporation of selective recogni-
tion units makes the synthesis of such supramolecular ligand
systems more elaborate and costly.
University of Basel
St. Johanns-Ring 19, 4056 Basel (Switzerland)
Fax : (+41)61-267-1103
E-mail: andreas.pfaltz@unibas.ch
[b] Dr. D. J. Frank
Current Address: School of Chemistry
Joseph Black Building, West Mains Road
Edinburgh, EH93JJ (UK)
[c] Dr. A. Franzke
Current address: Home Care and Formulation Technologies Europe
BASF SE, 67056 Ludwigshafen (Germany)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201202408.
Chem. Eur. J. 2013, 19, 2405 – 2415
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2405

An alternative way to favor the formation of hetero- over
homo-ligand combinations would be to incorporate an
anionic group in one of the ligands. In the case of a cationic
metal ion such as Rh+, charge repulsion should shift the
equilibrium toward the neutral hetero-ligand complex
(Scheme 1). Alternatively, the combination of a phosphor-
Scheme 1. Formation of bis-ligand metal complexes from a mixture of a
neutral (N) and an anionic (A) ligand.
amidite, which possesses a basic nitrogen atom, and a poten-
tial hydrogen-bond donor, such as a phosphoric acid diester,
should also favor the hetero-ligand complex by hydrogen-
bond formation (Scheme 2). This approach seems attractive
Scheme 3. Synthesis of anionic secondary amines 3a and 4a for the prep-
aration of monodentate phosphane ligands. Reagents and conditions:
a) PBr3, NBu4Br, CH2Cl2, 08C, 3 h, then 08C!rt, 17 h (97%);
b) CH3NH2, THF, 08C, then 08C!rt, 3 h (73%); c) piperazine, THF, rt,
16 h (95%).
Scheme 2. Formation of a mixed phosphoramidite/phosphoric acid diester
metal complex.
because complementary ligands of this type are readily
available from simple precursors. Herein, we report the syn-
thesis of a series of anionic phosphite and phosphoramidite
ligands and phosphoric acid diesters, the results of complex-
ation studies, and Rh-catalyzed hydrogenation reactions that
use mixtures of these ligands.
Scheme 4. Synthesis of neutral secondary amines 3b and 4b for the prep-
aration of monodentate phosphane ligands. Reagents and conditions:
a) CH3NH2, THF, 08C, then 08C!rt, 4 h (95%); b) piperazine, toluene,
858C, 2 h (74%).
Results and Discussion
Synthesis of monodentate phosphite and phosphoramidite li-
gands: In connection with our work on zwitterionic iridium
complexes with anionic N,P ligands,[11] we developed fluori-
nated tetraarylborate 1a,[12] which we thought would be a
suitable building block for the synthesis of phosphites and
phosphoramidites bearing a non-coordinating anionic unit.
Two secondary amines 3a and 4a were prepared from 1a
through bromide 2a (Scheme 3). We also prepared analo-
gous neutral amines 3b and 4b by starting from the fluori-
nated benzyl bromide 2b (Scheme 4).[13]
Following reported procedures[14] a small library of three
pairs of anionic chiral ligands A1–A3 and their neutral ana-
logues N1–N3 was prepared by reaction of (R)-BINOL with
PCl3 and subsequent nucleophilic substitution with the cor-
responding alcohol or secondary amine (Scheme 5). In addi-
tion, achiral phosphites N4, N5, and A6 and the anionic
phosphinite A7 were prepared in a similar way by starting
from biphenol or diphenylphosphanyl chloride (Scheme 6).
Scheme 5. Preparation of BINOL ligands N1–N3 and A1–A3. Reagents
and conditions: a) PCl3, NEt3, THF, 08C, 80 min, then rt, 75 min (79%);
b) amine or alcohol, Et2O, toluene or THF, NEt3, 08C!rt, 19 h (66–
96%).
Hydrogenation studies: These new chiral and achiral ligands
were subsequently evaluated in the Rh-catalyzed hydroge-
nation of selected functionalized olefins. The pairs of struc-
2406
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 2405 – 2415

Binary Mixtures of Monodentate Phosphanes in Rh-Catalyzed Hydrogenations
FULL PAPER
Table 1. Hydrogenation of (Z)-methyl 2-acetamido-3-phenylacrylate (9)
with catalysts derived from the individual ligands.
Entry
L1
Solvent
Conv.
[%][a]
ee [%][b]
1
2
3
4
5
6
N1
N2
N3
A1
A2
A3
toluene
toluene
toluene
toluene
toluene
toluene
80
98
29
>99
61
81
80 (S)
97 (S)
89 (S)
84 (S)
53 (S)
93 (S)
7
8
9
10
11
12
N1
N2
N3
A1
A2
A3
iPrOH
iPrOH
iPrOH
iPrOH
iPrOH
iPrOH
5
17
24
>99
78
81
44 (S)
94 (S)
97 (S)
52 (S)
88 (S)
98 (S)
Scheme 6. Preparation of biphenol ligands N4, N5, and A6 and diphenyl-
phosphinite A7. Reagents and conditions: a) PCl3, NEt3, THF, 08C,
80 min, then rt, 75 min (50%); b) alcohol, THF, NEt3, 08C!rt, 19 h (46–
91%).
13
14
15
16
17
18
N1
N2
N3
A1
A2
A3
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
>99
>99
>99
>99
>99
>99
76 (S)
96 (S)
99 (S)
54 (S)
82 (S)
99 (S)
turally analogous neutral and anionic ligands were included
in this study to evaluate the influence of the anionic group.
First, we studied the performance of the individual neutral
and anionic chiral ligands in the hydrogenation of (Z)-
methyl 2-acetamido-3-phenylacrylate (9) and dimethyl itaco-
nate (10) (Tables 1 and 2). The precatalysts were prepared
19
20
21
22
23
24
N1
N2
N3
A1
A2
A3
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
>99
>99
74
>99
2
84 (S)
97 (S)
91 (S)
52 (S)
90 (S)
95 (S)
in situ from [RhACHTUNGTRENNUNG(cod)2]BF4 and two equivalents of the
ligand. The reactions were carried out under standard condi-
tions at 10 bar hydrogen for 1 hour in iPrOH, EtOAc,
CH2Cl2, or toluene. The highest reactivities and enantiose-
lectivities were observed in CH2Cl2. The neutral and anionic
phosphoramidites N3 and A3 derived from piperazine
proved to be the overall most efficient ligands, thereby
giving up to 99% ee in the hydrogenation of 9 (Table 1, en-
tries 15 and 18) and 98 and 96% ee in the hydrogenation of
dimethyl itaconate (Table 2, entries 15 and 18, respectively).
Clear differences in enantioselectivity and conversion
values were found between anionic ligands and their neutral
counterparts. However, no uniform trends were observed.
Depending on the structure of the fluorinated P-aryl group
and the solvent, either the anionic or neutral ligand per-
formed better.
>99
[a] Conversions determined by GC analysis. [b] Enantioselectivities de-
termined by HPLC on a chiral stationary phase. cod=1,5-cyclooctadiene.
values were measured for all the binary mixtures of two
neutral ligands (Table 3, entries 1, 3, 5, 7, 9, 11, 13, and 15).
In the hydrogenation of 10, the application of binary
ligand mixtures again improved the enantioselectivities in
many cases. Because the phosphoramidites showed lower se-
lectivities as individual ligands relative to the corresponding
phosphites, combinations of N2 and A2 with achiral ligands
were first investigated. Subsequently, the most promising
achiral derivatives were also tested in mixtures with the
more efficient chiral phosphites N1 and A1. Ligand PACHTUNGTRENNUNG(oTol)3
Next, binary mixtures comprising one of the less success-
ful chiral BINOL ligands (N1, N2, A1, or A2) in combina-
tion with the achiral ligands PPh3, PACHTNUGTRENNUG(o-Tol)3, PCAHTUNGTRENN(GUN OPh)3, N4,
in combination with the anionic phosphite A1 in ethyl ace-
tate gave the most spectacular improvement, with a selectiv-
ity of 87% ee instead of only 6% ee induced by A1 alone or
N5, A6, or A7 were evaluated. In the hydrogenation of 9,
the chiral phosphites A1 and N1 were tested that had fur-
nished only moderate enantioselectivities in the screening of
individual ligands. Most combinations of the anionic phos-
phite A1 with neutral achiral phosphites led to improved se-
lectivities. The most significant improvement was obtained
with the mixture of A1 and N4 in EtOAc with 70% ee in-
stead of 52% ee (Table 3, entry 12). On the other hand, the
combination of the chiral phosphoramidite A2 with N4 did
not improve the already quite high enantioselectivities fur-
ther (Table 3, entries 6, 10, and 14). Significantly lower
18% ee for the corresponding mixture of PACTHNGUETRNNU(G oTol)3 with N1
(Table 4, entries 13 and 14). In addition, a large increase in
reactivity was observed as the conversion was raised from 2
to 93%. The mixtures of anionic A2 with PACTHNURTGNENG(U OPh)3, N4, or
N5 showed the best performance in the rather nonpolar sol-
vent toluene with improved enantioselectivities of 82, 81,
and 72% ee, respectively, relative to 56% ee for A2 as an in-
dividual ligand (Table 4, entries 2, 6, and 8).
In general, mixtures of neutral and anionic ligands fur-
nished better results in the hydrogenation of 9 and 10 than
combinations of two neutral or anionic ligands.
Chem. Eur. J. 2013, 19, 2405 – 2415
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2407

A. Pfaltz et al.
Table 2. Hydrogenation of 10 with catalysts derived from individual li-
gands.
We next studied the hydrogenation of simple enam-
ides.[15,16] The reactions were carried out at 20 bar hydrogen
pressure with a reaction time of 4 h in CH2Cl2 or EtOAc,
the conditions under which the best results had been ach-
ieved so far. In the hydrogenation of N-(1-phenylvinyl)ace-
tamide (11), which serves as a benchmark for this class of
olefin, excellent enantioselectivities of up to >99% ee were
obtained in CH2Cl2 by using the anionic phosphoramidite
A3 as a ligand (Table 5, entry 6).
In view of the almost perfect enantioselectivites obtained
for this substrate, the enamide derived from pinacolone (12)
was chosen as a more challenging substrate. The best result
was obtained with the anionic phosphite ligand A1, which
furnished 81% ee in CH2Cl2 (Table 6, entry 4). However, the
enantioselectivities dramatically decreased in EtOAc as the
solvent. Interestingly, phosphite ligands A1 and N1 gave the
(ꢀ)-enantiomer, whereas phosporamidites N2, N3, A2, and
A3 induced the opposite configuration.
Entry
L1
Solvent
Conv.
[%][a]
ee
[%][b]
1
2
3
4
5
6
N1
N2
N3
A1
A2
A3
toluene
toluene
toluene
toluene
toluene
toluene
33
12
>99
78
37
54
2 (R)
19 (R)
racemic
97 (R)
56 (R)
87 (R)
7
8
9
10
11
12
N1
N2
N3
A1
A2
A3
iPrOH
iPrOH
iPrOH
iPrOH
iPrOH
iPrOH
3
17
20
39
75
49
10 (R)
81 (R)
88 (R)
53 (R)
84 (R)
96 (R)
No heterocombination of chiral and achiral ligands was
found that improved the enantioselectivity of 81% ee ob-
tained with A1 alone in CH2Cl2. However, a beneficial
effect of achiral ligands was observed for the less efficient
chiral derivatives N2 and A3 in CH2Cl2 and for several com-
binations in EtOAc as the solvent (Table 7). In the former
case, enhanced enantioselectivities were found for mixtures
13
14
15
16
17
18
N1
N2
N3
A1
A2
A3
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
>99
>99
>99
>99
>99
>99
93 (R)
86 (R)
98 (R)
94 (R)
91 (R)
96 (R)
19
20
21
22
23
24
N1
N2
N3
A1
A2
A3
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
>99
>99
>99
2
>99
>99
4 (S)
72 (R)
17 (R)
6 (R)
73 (R)
53 (R)
with achiral phosphites PACTHNUTRGNE(UNG OPh)3, N4, or A6, and the stron-
gest increase in selectivity was measured by using a combi-
nation of neutral and anionic phosphites N2 and A6
(Table 7, entry 4). The best result in EtOAc was obtained
for the combination of A1 and PACHTNUGTRENUNG(OPh)3 with an increase in
[a] Conversions determined by GC analysis. [b] Enantioselectivities deter-
mined by GC analysis on a chiral stationary phase.
enantioselectivity from 36 to 65% (Table 7, entry 11). How-
ever, at the same time, the conversion dropped
from 89 to 38%. Significant increases in enantiose-
lectivity were also found for combinations of the
chiral neutral ligands N1 and N2 with achiral anion-
ic phosphites A6 and A7 (Table 7, entries 4, 5, and
12). Interestingly, the achiral ligand A7 inverted the
Table 3. Hydrogenation of substrate 9 by using binary mixtures of ligands.
sense of chiral induction observed with N2 alone
(Table 7, entry 5). Binary mixtures of two anionic li-
gands showed no improvement in enantioselectivity
(see the Supporting Information).
L1 +L2
L1 +L1
Entry
L1
L2
Solvent
Conv.
ee
Conv.
ee
[%][a]
[%][b]
[%][a]
[%][b]
1
2
3
4
5
6
N1
A1
N1
A1
N2
A2
P
P
U
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
>99
>99
>99
>99
>99
>99
49 (S)
58 (S)
35 (S)
61 (S)
72 (S)
62 (S)
>99
>99
>99
>99
>99
>99
76 (S)
54 (S)
76 (S)
54 (S)
96 (S)
82 (S)
The hydrogenation of N-(3,4-dihydronaphthalen-
1-yl)acetamide (13) derived from a-tetralone gener-
ally proceeded with higher enantioselectivities than
the hydrogenation of the sterically hindered enam-
ide 12. The best results were obtained with A1,
with 96 and 97% ee in CH2Cl2 and EtOAc, respec-
tively. However, the conversion reactions were still
incomplete after 4 h (Table 8, entries 4 and 10). In-
terestingly, the + enantiomer was formed when the
anionic phosphoramidite A2 was used (Table 8, en-
tries 5 and 11), whereas all the other ligands gave
the ꢀ enantiomer.
N4
N4
N4
N4
7
8
9
10
N1
A1
N2
A2
N4
N4
N4
N4
iPrOH
iPrOH
iPrOH
iPrOH
>99
>99
38
17 (S)
60 (S)
17 (S)
60 (S)
5
>99
2
44 (S)
52 (S)
94 (S)
88 (S)
1
2
11
12
13
14
15
16
N1
A1
N2
A2
N1
A1
N4
N4
N4
N4
N5
N5
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
>99
>99
>99
93
>99
>99
30 (S)
70 (S)
66 (S)
59 (S)
26 (S)
56 (S)
>99
>99
>99
2
>99
>99
84 (S)
52 (S)
97 (S)
90 (S)
84 (S)
52 (S)
As observed for enamide 12, the application of
ligand mixtures did not furnish higher enantioselec-
tivities for the optimal ligand A1, but again the se-
lectivities obtained for the less efficient phosphor-
[a] Conversions determined by GC. [b] Enantioselectivities determined by HPLC on a
chiral stationary phase.
2408
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 2405 – 2415

Binary Mixtures of Monodentate Phosphanes in Rh-Catalyzed Hydrogenations
FULL PAPER
Table 4. Hydrogenation of alkene 10 using binary mixtures of ligands.
been observed for other achiral ligands. The mix-
ture of the anionic phosphoramidite A3 and
PACTHUNTRGENNG(U OPh)3 in EtOAc resulted in the highest increase
in selectivity from 20 to 52% ee for A3 alone and
the mixture, respectively (Table 9, entry 12).
L1 +L2
L1 +L1
Taken together, in contrast to substrates 9 and
10, no general beneficial effect of the anionic li-
gands in binary mixtures with neutral ligands was
observed for enamides 12 and 13. However, the
anionic phosphite A1 applied as an individual
ligand gave the best result in the hydrogenation of
these substrates.
Finally, we also investigated compounds 14 and
15 as substrates (Figure 1). Only very low enantio-
selectivities of 18% ee at best were obtained when
individual ligands were used. Improved enantiose-
lectivities were observed in several cases for binary
mixtures of chiral BINOL derivatives with achiral
ligands. The neutral phosphoramidite N2 and PPh3
represented the best heterocombination in the
asymmetric hydrogenation of 14, thus yielding
62% ee instead of 10% ee with N2 alone. In the hy-
drogenation of enamide 15, derived from b-tetra-
lone, the anionic phosphoramidite A2 performed
best in combination with PPh3, thus furnishing
36% ee relative to 11% ee with A2 alone (see the
Supporting Information).
Entry
L1
L2
Solvent
Conv.
ee
Conv.
ee
[%][a]
[%][b]
[%][a]
[%][b]
1
2
3
4
5
6
7
8
N2
A2
N1
A1
N2
A2
N2
A2
P
P
U
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
7
70
24
38
18
45
29
37
33 (R)
82 (R)
56 (R)
63 (R)
12 (R)
81 (R)
13 (R)
72 (R)
12
37
33
78
12
37
12
37
19 (R)
56 (R)
2 (R)
97 (R)
19 (R)
56 (R)
19 (R)
56 (R)
N4
N4
N4
N4
N5
N5
9
N2
A2
N1
N2
N4
N4
A6
A6
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
>99
>99
>99
>99
86 (R)
92 (R)
48 (R)
87 (R)
>99
>99
>99
>99
86 (R)
91 (R)
93 (R)
86 (R)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
N1
A1
N2
A2
N1
A1
N2
A2
N2
A2
N1
A1
N2
A2
P
P
P
P
P
P
P
P
U
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
>99
93
18 (R)
87 (R)
80 (R)
64 (R)
10 (R)
39 (R)
65 (R)
78 (R)
76 (R)
48 (R)
47 (R)
26 (R)
87 (R)
79 (R)
>99
2
>99
>99
>99
2
>99
>99
>99
>99
>99
2
4 (R)
6 (R)
72 (R)
73 (R)
4 (R)
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
6 (R)
72 (R)
73 (R)
72 (R)
73 (R)
4 (R)
6 (R)
72 (R)
73 (R)
N4
N4
A7
A7
A7
A7
>99
>99
[a] Conversions determined by GC analysis. [b] Enantioselectivities determined by
GC analysis on a chiral stationary phase.
Figure 1. Tiglic acid (14) and N-(3,4-dihydronaphthalen-2-yl)-
acetamide (15) as substrates for the asymmetric hydrogenation
reaction.
Table 5. Reduction of N-(1-phenylvinyl)acetamide (11).
Table 6. Hydrogenation of N-(3,3-dimethylbut-1-en-2-yl)acetamide (12)
with catalysts derived from individual ligands.
Entry
L1
Solvent
Conv. [%][a]
ee [%][b]
1
2
3
4
5
6
N1
N2
N3
A1
A2
A3
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
>99
>99
>99
>99
>99
>99
95 (ꢀ)
91 (ꢀ)
99 (ꢀ)
Entry
L1
Solvent
Conv. [%][a]
ee [%][b]
89 (ꢀ)
49 (ꢀ)
1
2
3
4
5
6
N1
N2
N3
A1
A2
A3
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
99
80
78
97
93
84
46 (+)
29 (ꢀ)
69 (ꢀ)
81 (+)
71 (ꢀ)
27 (ꢀ)
>99 (ꢀ)
[a] Conversions determined by GC analysis. [b] Enantioselectivities de-
termined by GC analysis on a chiral stationary phase.
ACHTUNGTRENNUNGamidites N2, N3, and A2 were improved by the addition of
7
8
9
10
11
12
N1
N2
N3
A1
A2
A3
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
>99
>99
>99
89
<1
41
4 (+)
8 (ꢀ)
an achiral phosphite ligand. The most significant increases
in enantiomeric excess in CH2Cl2 were measured for the
mixture of neutral ligands N3 and PACTHNUTRGENUG(N OPh)3 (from 47 to
75% ee; Table 9, entry 1) and the combination of the anion-
ic and neutral ligands A2 and N4 (from +24 to ꢀ68% ee;
Table 9, entry 5). In the latter example, the achiral ligand
N4 caused a reversal of enantioselectivity, as had already
racemic
36 (+)
50 (ꢀ)
15 (ꢀ)
[a] Conversions determined by GC analysis. [b] Enantioselectivities de-
termined by GC analysis on a chiral stationary phase.
Chem. Eur. J. 2013, 19, 2405 – 2415
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2409

A. Pfaltz et al.
Table 7. Hydrogenation of 12 by using binary mixtures of ligands.
anionic and a neutral ligand indeed leads to a
higher proportion of the hetero-ligand complex
than the mixture of structurally analogous neutral
ligands.
PACTHNUTRGEN(UGN OPh)3 (N8) was chosen as an exemplary achiral
L1 +L2
L1 +L1
partner ligand L in these investigations. The phos-
phites N1 and A1 and the phosphoramidites N3
and A3 were used as chiral counterparts L*. A
clear trend was observed when the chiral neutral li-
gands on the one hand or the corresponding anionic
derivatives on the other hand were combined with
Entry
L1
L2
(OPh)3
N4
N4
A6
A7
Solvent
Conv.
ee
Conv.
ee
[%][a]
[%][b]
[%][a]
[%][b]
1
2
3
4
5
N2
N2
A3
N2
N2
P
G
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
35
99
>99
98
50 (ꢀ)
48 (ꢀ)
42 (ꢀ)
58 (ꢀ)
46 (+)
80
80
84
80
80
29 (ꢀ)
29 (ꢀ)
27 (ꢀ)
29 (ꢀ)
29 (ꢀ)
>99
PACHTUNGTRENNUG
AHCTUNGTRENNUNG
6
7
8
9
10
11
12
N2
A3
N1
N2
N2
A1
N1
PPh3
PPh3
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
99
87
32
29
5
20 (ꢀ)
19 (ꢀ)
26 (ꢀ)
31 (ꢀ)
45 (ꢀ)
65 (+)
33 (+)
>99
41
>99
>99
>99
89
8 (ꢀ)
15 (ꢀ)
4 (+)
8 (ꢀ)
8 (ꢀ)
36 (+)
4 (+)
AHCTUNGTRENNUNG
P
P
P
P
C
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
38
72
two doublets with large Rh–P couplings of about
J=250 Hz. The hetero complex gave rise to two
doublets of doublets with an additional smaller P–P
coupling close to J=50 Hz. The effect of the nega-
A6
>99
[a] Conversions determined by GC analysis. [b] Enantioselectivities determined by
GC analysis on a chiral stationary phase.
Table 8. Hydrogenation results for 13 with catalysts derived from individ-
Table 9. Hydrogenation of 13 with binary mixtures of ligands.
ual ligands.
L1 +L2
L1 +L1
ee
[%][b]
Entry
L1
Solvent
Conv. [%][a]
ee [%][b]
Entry
L1
L2
Solvent
Conv.
ee
Conv.
[%][a]
[%][b]
[%][a]
1
2
3
4
5
6
N1
N2
N3
A1
A2
A3
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
68
19
80
88
45
68 (ꢀ)
27 (ꢀ)
47 (ꢀ)
96 (ꢀ)
24 (+)
76 (ꢀ)
1
2
3
4
5
6
7
8
N3
A2
N2
N3
A2
N2
N3
A2
P
P
U
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
33
26
83
96
88
95
88
60
75 (ꢀ)
29 (+)
52 (ꢀ)
53 (ꢀ)
68 (ꢀ)
40 (ꢀ)
66 (ꢀ)
42 (ꢀ)
80
45
19
80
45
19
80
45
47 (ꢀ)
24 (+)
27 (ꢀ)
47 (ꢀ)
24 (+)
27 (ꢀ)
47 (ꢀ)
24 (+)
N4
N4
N4
A6
A6
A6
>99
7
8
9
10
11
12
N1
N2
N3
A1
A2
A3
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
19
43
29
63
15
43 (ꢀ)
9 (ꢀ)
34 (ꢀ)
97 (ꢀ)
45 (+)
20 (ꢀ)
9
N2
A3
N2
A3
A6
A6
PPh3
EtOAc
EtOAc
EtOAc
EtOAc
24
24
98
30
41 (ꢀ)
25 (ꢀ)
18 (+)
52 (ꢀ)
43
>99
43
9 (ꢀ)
20 (ꢀ)
9 (ꢀ)
10
11
12
>99
P
N
>99
20 (ꢀ)
[a] Conversions determined by GC analysis. [b] Enantioselectivities de-
termined by HPLC on a chiral stationary phase.
[a] Conversions determined by GC analysis. [b] Enantioselectivities de-
termined by HPLC on a chiral stationary phase.
tive charge on ligands A1 and A3 was reflected by the ratios
31P NMR spectroscopic studies of rhodium(I) complexes
formed from binary ligand mixtures: As a working hypothe-
sis, we assumed that the equilibrium between the hetero-
ligand metal complex and the corresponding homo-ligand
complexes would be shifted toward the hetero-ligand com-
plex by introducing a negative charge on one of the ligands.
In the case of rhodium(I) complexes, the formation of a
neutral zwitterionic hetero-ligand complex should be fa-
vored due to charge repulsion between the anionic ligands
and an attractive interaction between the rhodium cation
and the anionic ligand. This effect should be especially pro-
nounced in apolar media. We carried out 31P NMR spectro-
scopic studies to examine whether the combination of an
of the three species [Rh
ACHTUGTNRNENUG(cod)L2], [RhAHCTUNGTREN(NUNG cod)LL*], and [Rh-
AHCTUNGTRENNUNG
analogous mixtures of the corresponding two neutral li-
gands. The combination of the neutral ligands N1 and N3
with PACTHUNTGRNE(NUG OPh)3 resulted in proportions of the hetero complex
of 60 and 56%, respectively, which is close to the statistical
ratio of 1:2:1. In contrast, the corresponding mixtures of P-
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
close to 75%, thus corresponding to a ratio of 1:6:1. The
31P NMR spectrum obtained for the binary mixture of P-
AHCTUNGTRENNUNG
2410
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 2405 – 2415

Binary Mixtures of Monodentate Phosphanes in Rh-Catalyzed Hydrogenations
FULL PAPER
Figure 2. 31P{1H} NMR spectrum obtained after mixing [Rh
(dd, 1JPRh =259, 2JPP =47 Hz; Rh(PA1 (PN8)), 122.1 (d, 1JPRh =265 Hz; Rh
257 Hz; Rh
(PN8)2).
N
ACHTUGNTNREN(UNG OPh)3 (N8; 6 mmol) and A1 (6 mmol) in CD2Cl2 at 295 K: d=122.5
1
U
)
A
E
(PA1 (PN8)), 105.1 ppm (d, JPRh
ACHTUNGTRENNUNG )ACHTUNGERTNNUNG =
ACHTUNGTRENNUNG
These results confirm our expectation that the use of an
anionic ligand in combination with a neutral ligand favors
the formation of the mixed bis-ligand complex over the cor-
responding homo-ligand complexes. However, a word of
caution should be added: Because the catalytic activities of
the homo- and hetero-ligand complexes can differ quite
substantially, a higher proportion of the hetero-ligand com-
plex does not necessarily mean that the reaction is predom-
inantly catalyzed by this species.
Scheme 7. Formation of rhodium complexes from binary mixtures of
phosphoric acid diesters and phosphoramidites.
were readily synthesized as shown in Scheme 8. Protection
of biphenol (7) as the corresponding MOM ether 16, subse-
quent introduction of the methyl substituents, and deprotec-
tion provided the sterically more demanding derivative
17.[18] Finally, the diols were transformed into the corre-
sponding chlorophosphanes and converted into the phos-
phoric acid diesters N9–N11 by the addition of tBuOH.[17]
With these ligands in hand, the rhodium complexes ob-
tained from mixtures of phosphoric acid diesters N9–N11
with BINOL-based ligands N1–N3 and A1–A3 were investi-
gated by 31P NMR spectroscopy to determine the ratios be-
tween the homo- and hetero-ligand complexes. By using
(R)-N9 as a potential hydrogen-donor ligand, only the com-
bination with A3 led to a moderate increase in the propor-
tion of the hetero complex (68%). In contrast, the combina-
tion of the R-configured piperazine-derived phosphorami-
dites N3 and A3 with (S)-N9 resulted in the exclusive for-
mation of the corresponding hetero complexes (Figure 3).
For other combinations of (S)-N9 with phosphites N1 or A1
or phosphoramidites N2 or A2, the preference for the
Binary mixtures of phosphorous acid diesters and phosphor-
amidites: Phosphoric acid diesters exist in two tautomeric
forms in solution, that is, a PV oxoform, which according to
IUPAC recommendations is called a phosphonic acid die-
ster, and a PIII hydroxy species. The trivalent hydroxy tauto-
mer usually binds to transition metals through the phospho-
rus atom, thus forming a complex with the structural subunit
(ArO)2P(OH). The hydroxy function can form a hydrogen
bond within the metal complex that comprises a properly
chosen second ligand with a hydrogen-acceptor group, such
as, for example, a phosphoramidite moiety. Therefore, the
combination of a phosphoric acid diester with a phosphora-
midite species is expected to preferentially form the corre-
sponding mixed-ligand complex (Scheme 7).
To explore binary ligand mixtures of this type, a set of
four phosphonic acid diesters was prepared, namely, both
enantiomers of BINOL-P(O)H (N9)[17] and two achiral bi-
phenol-based ligands N10[14b] and N11. These compounds
Chem. Eur. J. 2013, 19, 2405 – 2415
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2411

A. Pfaltz et al.
The binary mixtures that had led to the quantitative for-
mation of the hetero complexes were then applied to the hy-
drogenation of the demanding enamide substrates 12 and
13. In the hydrogenation of substrate 12, only the combina-
tion of ligand A3 with (S)-N9 led to a slight improvement of
selectivity from 27 to 39% ee, whereas all the other combi-
nations gave lower or similar values relative to N3 or A3
alone (see the Supporting Information).
Much better results were obtained in the hydrogenation
of the dihydronaphthalene derivative 13. A significant in-
crease in enantioselectivity to 91–97% ee was achieved by
using hetero complexes generated from the anionic ligand
A3 in combination with the neutral phosphoric acid diesters
(S)-N9, N10, and N11 relative to 76% ee for the homo-
ligand complex (Table 10, entries 5–7). The corresponding
neutral ligand N3 once again gave inferior results. Only the
Table 10. Application of binary mixtures of phosphoramidites N3 and
A3 and phosphoric acid diesters N9–N11 in the hydrogenation of 13 in
CH2Cl2.
Scheme 8. Preparation of phosphoric acid diesters N9–N11. Reagents and
conditions: a) NaH; THF, 08C!rt, 75 min, then MOMCl, 08C!rt, 16 h
(60%); b) nBuLi, THF, rt, 1 h, then MeI, rt, 1 h, then Amberlyst 15,
THF/MeOH (1:1), D, 17 h (54%); c) PCl3, NEt3, toluene/THF (10:1),
08C!rt, 1 h, then tBuOH, toluene, 08C!rt, 2 h (41–53%). MOMCl=
methoxymethyl chloride.
hetero complex was less pronounced. Apparently, the mix-
ture of (R)-N9 and phosphoramidite A3 seems to represent
the mismatched combination with respect to thermodynamic
stability, whereas the mixture of (S)-N9 and A3 corresponds
to the matched case.
The achiral phosphoric acid diesters N10 and N11 also
formed the corresponding hetero complexes quantitatively
in combination with the neutral and anionic piperazine-de-
rived phosphoramidites N3 and A3. This outcome is illus-
trated by the 31P NMR spectrum of the rhodium complex
obtained from A3 and N11 (Figure 4).
L1 +L2
L1 +L1
Entry
L1
L2
Conv.
ee
Conv.
ee
[%][a]
[%][b]
[%][a]
[%][b]
1
2
3
4
5
6
7
N3
N3
N3
A3
A3
A3
A3
(S)-N9
N10
N11
(R)-N9
(S)-N9
N10
90
74
94
76 (ꢀ)
37 (ꢀ)
50 (ꢀ)
20 (ꢀ)
97 (ꢀ)
91 (ꢀ)
97 (ꢀ)
80
80
80
>99
>99
>99
>99
47 (ꢀ)
47 (ꢀ)
47 (ꢀ)
76 (ꢀ)
76 (ꢀ)
76 (ꢀ)
76 (ꢀ)
89
>99
>99
>99
N11
[a] Conversions determined by GC analysis. [b] Enantioselectivities de-
termined by HPLC on a chiral stationary phase.
Figure 3. 31P{1H} NMR spectrum of the binary mixture of [Rh
ACHTNUGTENRNU(GN cod)2]BF4 (6 mmol), (S)-N9 (6 mmol), and N3 (6 mmol) in CD2Cl2 at 295 K: d=142.8 (dd,
2
2
1
1JPRh =259, JPP =52 Hz; Rh(PN3 (PN9)), 115.7 (dd, JPRh =228, JPP =53 Hz; Rh(PN3 (PN9)), 14.4 ppm (s; N9).
ACHTUNGTRNENUG )ACHUTGTNRENNUG ACHTUNGTNERNUG )CAHTNUGTRENNUGN
2412
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 2405 – 2415

Binary Mixtures of Monodentate Phosphanes in Rh-Catalyzed Hydrogenations
FULL PAPER
Figure 4. 31P{1H} NMR spectrum of the binary mixture of [Rh
ACHTNUGTENRNU(NG cod)2]BF4 (6 mmol), N11 (6 mmol), and A3 (6 mmol) in CD2Cl2 at 295 K: d=143.9 (dd,
2
2
1
1JPRh =262, JPP =53 Hz; Rh(PA3 (PN11)), 111.5 (dd, JPRh =222, JPP =52 Hz; Rh(PA3 (PN11)), 12.1 ppm (s; N11).
ACHTUNGTRNENUG )ACHUTGTNRENNUG ACHTUNGTNERNUG )CAHTNUGTRENNUGN
combination with (S)-N9 led to a notable improvement
from 47 to 76% ee in this case (Table 10, entry 1). Consis-
tent with the results from the NMR spectroscopic studies,
pronounced matched and mismatched effects were observed
with (R)- and (S)-N9. In the case of ligand A3, the selectivi-
ty dropped from 97 to 20% ee, when (R)-N9 was used in-
stead of the S enantiomer (Table 10, entries 4 and 5).
The results obtained for combinations of phosphoric acid
diesters with phosphoramidites are very promising. The
strong preference for the formation of hetero- rather than
homo-ligand complexes and the high enantioselectivities in
the hydrogenation of enamide 13 indicate that mixtures of
potential hydrogen-donor and -acceptor ligands of this type
offer new possibilities for combinatorial-catalyst develop-
ment.
in some cases, indicate that the introduction of a negative
charge may result in a useful diversification of monophos-
phane libraries.
Moreover, anionic ligands such as A1–A3 offer new possi-
bilities for catalyst preparation from binary mixtures of
monodentate phosphanes. As we have shown in NMR stud-
ies, the combination of an anionic ligand with a neutral
ligand can shift the equilibrium from the statistical distribu-
tion of complexes toward the hetero bis-ligand complex.
This behavior may have a beneficial effect on the enantiose-
lectivity, especially when a mixture of an achiral and a chiral
ligand is used. Indeed, in the hydrogenation of substrates 9
and 10, the combination of a neutral with an anionic phos-
phorus donor, one chiral and the other achiral, often fur-
nished significantly better enantioselectivities than mixtures
of two structurally analogous neutral ligands.
The formation of hetero bis-ligand complexes can be fur-
ther promoted by a hydrogen bond between the two ligands.
On the basis of this concept, we explored combinations of
phosphoric acid diesters as hydrogen donors with phosphor-
amidites as hydrogen acceptors. Some of these ligand mix-
tures formed the hetero bis-ligand complex exclusively, as
evidenced by NMR analysis, and also induced very high
enantioselectivities in rhodium-catalyzed hydrogenation re-
actions that clearly exceeded the enantioselectivities of the
corresponding homo bis-ligand complexes. By far the best
results were obtained with the anionic phosphoramidite A3
in combination with neutral phosphoric acid diesters (S)-N9
or N11. Evidently, the combination of negatively charged
and neutral monophosphanes with complimentary hydro-
gen-donor and -acceptor properties seems to be a strategy
Conclusion
With the incorporation of an anionic group in chiral mono-
dentate phosphite and phosphoramidite ligands, we have
added an additional structural feature that can have a dis-
tinct effect on the catalytic activity and enantioselectivity of
a metal complex. When these ligands were tested in rhodi-
um-catalyzed hydrogenation reactions, pronounced differen-
ces in enantiomeric excess and conversion were observed
between anionic ligands and their neutral counterparts. De-
pending on the substrate, the ligand structure, and the sol-
vent, the charge effects were either positive or negative, so
no uniform trends could be discerned from the results. How-
ever, the high enantioselectivities induced by anionic ligands
Chem. Eur. J. 2013, 19, 2405 – 2415
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2413

A. Pfaltz et al.
and washing of the remaining solid with THF, toluene, or Et2O (10 mL),
the filtrate was concentrated under reduced pressure and the product
was purified by column chromatography.
with considerable potential for combinatorial-catalyst devel-
opment.
The preparation of phosphoramidite A3: Compound A3 was prepared
according to the general procedure from phosphorochloridite 7 (328 mg,
925 mmol, 1.05 equiv), amine 4a (954 mg, 836 mmol, 1.00 equiv), and NEt3
(369 mL, 268 mg, 2.65 mmol, 3.00 equiv) in THF (15 mL). After purifica-
tion by column chromatography (SiO2, 3ꢂ11 cm, CH2Cl2/MeOH 25:1)
A3 was obtained as a colorless solid (1.08 g, 89%). Rf =0.83–0.44 (SiO2,
CH2Cl2/MeOH=25:1); [a]D20 =ꢀ128.5 (c=0.930, CHCl3); 1H NMR
(400.1 MHz, CD2Cl2, 300 K): d=8.00 (d, J=8.8 Hz, 1H; Naph-H), 7.96–
7.91 (m, 3H; Naph-H), 7.88 (s, 6H; ArF-o-H), 7.57 (s, 3H; ArF-p-H), 7.54
(d, J=8.8 Hz, 1H; Naph-H), 7.45–7.41 (m, 2H; Naph-H), 7.37 (d, J=
8.9 Hz, 2H; Naph-H), 7.33 (d, J=8.3 Hz, 1H; Naph-H), 7.30–7.24 (m,
Experimental Section
All the reactions were performed in flame-dried glassware under argon
by using Schlenk techniques. The solvents, PCl3, and NEt3 were dried by
using standard procedures in a nitrogen or argon atmosphere or by pas-
sage over a column of activated alumina under nitrogen (PureSolv, Inno-
vative Technology Inc).[19,20] All the other commercial reagents were used
as received. Deuterated solvents for NMR analysis were degassed by
mean of three freeze/pump/thaw cycles, dried over 4 ꢁ molecular sieves,
and stored in argon. The solvents for the workup and the chromatograph-
ic purification of air-sensitive compounds were purged with a stream of
argon for at least 15 min prior to use. Chromatographic separations were
performed on silica gel 60 (Merck, Darmstadt; 40–63 nm). Precoated Ma-
cherey–Nagel Polygram SIL G/UV254 plates were used for TLC analysis,
and the compounds were visualized with UV light.
2H; Naph-H), 3.62 (s, 2H; ArCH2N), 3.12–3.06 (m, 2H; PN
3.01 (br s, 2H; PN(CHH)2), 2.90 (mc, 8H; NCH2CH2CH2CH3), 2.38 (br s,
4H; CH2N(CH2)2), 1.47 (mc, 8H; NCH2CH2CH2CH3), 1.27 (sext, J=
ACHTUNGTRENNUNG(CHH)2),
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
7.4 Hz, 8H; NCH2CH2CH2CH3), 0.91 ppm (t, J=7.3 Hz, 12H;
NCH2CH2CH2CH3); 11B{1H} NMR (128.4 MHz, CD2Cl2, 295 K): d=
ꢀ8.1 ppm; 13C{1H} NMR (100.6 MHz, CD2Cl2, 300 K): d=161.1 (dmc, J=
60 Hz; ArF-i-C), 149.7 (d, J=4 Hz; Naph-C), 149.6 (Naph-C), 148.2 (dmc,
J=241 Hz; Ar-m-C), 145.5 (dmc, J=244 Hz; Ar-o-C), 134.1 (ArF-o-CH),
132.8 (Naph-C), 132.6 (Naph-C), 131.5 (Naph-C), 131.1 (Naph-C), 130.4
(Naph-CH), 130.3 (Naph-CH), 128.8 (q, J=31 Hz; ArF-m-C), 128.6
(Naph-CH), 128.6 (Naph-CH), 126.8 (Naph-CH), 126.7 (Naph-CH),
126.3 (Naph-CH), 126.3 (Naph-CH), 125.0 (Naph-CH), 124.8 (q, J=
272 Hz; CF3), 124.8 (Naph-CH), 124.1 (d, J=5 Hz; Naph-C), 122.7 (d,
J=2 Hz; Naph-C), 122.2 (Naph-CH), 122.1 (Naph-CH), 117.4 (mc; ArF-
p-CH), 111.6 (t, J=18 Hz; Ar-p-C), 58.9 (t, J=3 Hz; NCH2CH2CH2CH3),
NMR spectroscopic experiments were performed on Bruker Avance 400
1
or 500 spectrometers. The H and 13C spectra were referenced relative to
SiMe4 by using the solvent signals as internal standards.[21, 22] The 31P, 19F,
and 11B spectra were calibrated with H3PO4 (85 wt%), CFCl3, and
BF3·OEt2 as external standards. The term dmc refers to a doublet of cen-
tered multiplets. Mass spectra were measured on VG70–250, Finnigan
MAT 95Q (EI), Finnigan MAT 312, Finnigan MAR 8400 (FAB), or Fin-
nigan MAT LCQ (ESI) spectrometers. Elemental analyses were per-
formed by the Micro-Analysis Laboratory at the University of Basel. IR
spectra were measured on a Perkin–Elmer 1600 FTIR spectrometer. Spe-
cific rotations were measured on a Perkin–Elmer 314 polarimeter. HPLC
analyses were performed on a Shimadzu system and GC measurements
on Carlo Erba Instruments. The abbreviation ArF refers to the 3,5-bis(tri-
fluoromethyl)phenyl substituent.
49.5
(CH2N
G
44.4
(ArCH2),
44.3
(PN
G
23.7
(NCH2CH2CH2CH3), 19.6 (t, J=1 Hz; NCH2CH2CH2CH3), 13.2 ppm
(NCH2CH2CH2CH3); despite prolonged data acquisition time, the signal
for Ar-i-C was not detected; 19F{1H} NMR (376.5 MHz, CD2Cl2, 300 K):
d=ꢀ63.7 (s, 18F; CF3), ꢀ128.6 (dd, J=25, 14 Hz, 2F; Ar-o-F),
ꢀ146.8 ppm (dd, J=24, 14 Hz, 2F; Ar-m-F); 31P{1H} NMR (202.5 MHz,
CD2Cl2, 295 K): d=142.2 ppm; IR (KBr): n˜ =3063, 2971, 2882, 1615,
1441, 1359, 1279, 1126, 943, 680 cmꢀ1; MS (ESI, CH2Cl2, 323 K): m/z (%):
1211 (100) [MꢀNBu4]ꢀ; elemental analysis calcd (%) for
C71H67BF22N3O2P (1454.06): C 58.65, H 4.64, N 2.89; found: C 58.77, H
4.59, N 2.92.
Preparation of anionic secondary amine 4a: Piperazine (304 mg,
3.53 mmol, 4.00 equiv) was added to a solution of anionic benzyl bromide
2a (1.00 g, 880 mmol, 1.00 equiv) in THF (5 mL). After stirring the mix-
ture overnight (19 h) at room temperature, a solution of 1m NaOH
(20 mL) was added to the resulting colorless suspension, and the reaction
mixture was extracted with CH2Cl2 (3ꢂ30 mL). The combined organic
layers were washed with water (2ꢂ20 mL) and brine (20 mL), dried over
Na2SO4, filtered, and evaporated. The product was obtained as a color-
less solid, which was used without further purification (954 mg, 95%).
1H NMR (400.1 MHz, CDCl3, 300 K): d=7.78 (s, 6H; ArF-o-H), 7.48 (s,
3H; ArF-p-H), 3.65 (s, 2H; ArCH2), 2.91 (mc, 8H; NCH2CH2CH2CH3),
General procedure for the synthesis of phosphoric acid diesters:[17] A sol-
ution of the diol (1.00 equiv) in absolute toluene (3 mLmmolꢀ1) and a
small amount of THF (<0.1 mLmmolꢀ1) to ensure solubilization was
added dropwise to an ice-cooled solution of PCl3 (2.00 equiv) and NEt3
(3.00 equiv) in toluene (1.5 mLmmolꢀ1
) over 30 min. The resulting
yellow mixture was stirred for 1 h at room temperature and filtered
through MgSO4, and the filtrate was concentrated in high vacuum. The
remaining colorless foam was redissolved in CH2Cl2 (2ꢂ5 mL), and the
solvent was removed by evaporation twice. tBuOH (1.00 equiv) was
added to the resulting phosphorochloridite dissolved in toluene
(1 mLmmolꢀ1) at 08C, and the reaction mixture stirred for 2 h at room
temperature. The precipitated product was collected by filtration and
washed with pentane. The filtrate was concentrated, the crude product
redissolved in toluene, and the ligand precipitated by addition of a triple
volume of pentane. The product was obtained as a colorless solid after
filtration.
2.87 (d, J=4.8 Hz, 2H; HN
ACHTUTGNRENNGU(CHH)2), 2.49 (s, 4H; CH2NACHTUNGTRNE(NUGN CH2)2), 1.54 (s,
2H; HN(CHH)2), 1.47 (mc, 8H; NCH2CH2CH2CH3), 1.25 (sext, J=
ACHTUNGTRENNUNG
7.4 Hz, 8H; NCH2CH2CH2CH3), 0.89 ppm (t, J=7.2 Hz, 12H;
NCH2CH2CH2CH3); 11B{1H} NMR (160.2 MHz, CDCl3, 295 K): d=
ꢀ7.5 ppm; 13C{1H} NMR (100.6 MHz, CDCl3, 300 K): d=134.1 (ArF-o-
CH), 128.8 (q, J=31 Hz; ArF-m-C), 124.7 (q, J=273 Hz; CF3), 117.3 (mc;
ArF-p-CH), 111.2 (Ar-p-C), 58.9 (NCH2CH2CH2CH3), 53.7 (CH2N-
ACHTUNGTRENNUNG(CH2)2), 49.9 (CH2Ar), 46.2 (HNAHCUTGNTREN(NUGN CH2)2), 23.7 (NCH2CH2CH2CH3), 19.6
(NCH2CH2CH2CH3), 13.3 ppm (NCH2CH2CH2CH3); despite prolonged
data acquisition, the signals for Ar-i-C, Ar-o-C, Ar-m-C, and ArF-i-C
were not detected; 19F{1H} NMR (376.5 MHz, CDCl3, 300 K): d=ꢀ63.3
(s, 18F; CF3), ꢀ128.0 (dd, J=25, 14 Hz, 2F; Ar-o-F), ꢀ146.1 ppm (dd,
J=25, 14 Hz, 2F; Ar-m-F); IR (KBr): n˜ =2973, 2885, 1616, 1443, 1360,
1280, 1125, 1012, 887, 837, 713, 680 cmꢀ1; MS (ESI, CH2Cl2, 323 K): m/z
(%): 897 (100) [MꢀNBu4]ꢀ; elemental analysis calcd (%) for
C51H56BF22N3 (1139.79): C 53.74, H 4.95, N 3.69; found: C 53.98, H 4.90,
N 3.63.
Preparation of phosphoric acid diester (S)-N9: According to the general
procedure, the PCl species was formed from PCl3 (506 mL, 796 mg,
5.80 mmol, 2.00 equiv), NEt3 (1.21 mL, 880 mg, 8.70 mmol, 3.00 equiv),
and (S)-BINOL (830 mg, 2.90 mmol, 1.00 equiv) in toluene (5 mL) and
THF (1.5 mL), filtered over MgSO4, and transformed into the phospho-
nate by reaction with tBuOH (272 mL, 2.90 mmol, 1.00 equiv) in toluene
(5 mL). Compound (S)-N9 precipitated as a colorless solid from the reac-
tion mixture and was washed with pentane after filtration (10 mL) to
yield the product (510 mg, 53%). [a]D20 = +673 (c=0.740, CHCl3);
1H NMR (400.1 MHz, CDCl3, 300 K): d=8.07 (d, J=8.8 Hz, 1H; Naph-
H), 8.06 (d, J=8.8 Hz, 1H; Naph-H), 7.98 (d, J=8.3 Hz, 2H; Naph-H),
7.58 (d, J=8.8 Hz, 1H; Naph-H), 7.55–7.48 (m, 3H; Naph-H), 7.37–7.28
General procedure for the synthesis of monodentate phosphite and phos-
phoramidite ligands:[14] Triethylamine (2.20 equiv) and the amine or alco-
hol (1.10 equiv) were subsequently added to a solution of the respective
phosphorochloridite (1.00 equiv) in THF (10 mLmmolꢀ1), toluene
(40 mLmmolꢀ1), or Et2O (10 mLmmolꢀ1) at 08C. The resulting colorless
suspension was stirred at room temperature overnight. After filtration
2414
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 2405 – 2415

Binary Mixtures of Monodentate Phosphanes in Rh-Catalyzed Hydrogenations
FULL PAPER
(m, 4H; Naph-H), 7.30 ppm (d, J=734 Hz, 1H; P(O)H); 13C{1H} NMR
(100.6 MHz, CDCl3, 300 K): d=145.8 (d, J=10 Hz; Naph-C), 147.0 (d,
J=11 Hz; Naph-C), 132.5 (Naph-C), 132.3 (Naph-C), 132.0 (d, J=2 Hz;
Naph-C), 132.0 (d, J=1 Hz; Naph-C), 131.4 (Naph-CH), 131.4 (Naph-
CH), 128.6 (Naph-CH), 128.6 (Naph-CH), 127.1 (Naph-CH), 127.1
(Naph-CH), 127.0 (Naph-CH), 126.9 (Naph-CH), 126.1 (Naph-CH),
126.0 (Naph-CH), 122.1 (d, J=2 Hz; Naph-C), 121.7 (d, J=3 Hz; Naph-
C), 120.8 (d, J=2 Hz; Naph-CH), 120.2 ppm (d, J=3 Hz; Naph-CH);
31P NMR (162.0 MHz, CDCl3, 300 K): d=14.4 ppm (d, J=733 Hz); IR
(KBr): n˜ =3509, 3433, 3055, 1671, 1595, 1507, 1469, 1383, 1221, 1183,
1148, 1021, 815, 749 cmꢀ1; MS (EI, 70 eV): m/z (%): 332 (100) [M+], 286
(10), 268 (78), 239 (53), 119 (12); elemental analysis calcd (%) for
C20H13O3P (332.29): C 72.29, H 3.94; found: C 72.22, H 4.17.
[4] C. Claver, E. Fernandez, A. Gillon, K. Heslop, D. J. Hyett, A. Mar-
torell, A. G. Orpen, P. G. Pringle, Chem. Commun. 2000, 961–962.
[5] M. T. Reetz, G. Mehler, Angew. Chem. 2000, 112, 4047–4049;
Angew. Chem. Int. Ed. 2000, 39, 3889–3890.
[6] a) M. T. Reetz, T. Sell, A. Meiswinkel, G. Mehler, Angew. Chem.
2003, 115, 814–817; Angew. Chem. Int. Ed. 2003, 42, 790–793;
b) M. T. Reetz, X. Li, Angew. Chem. 2005, 117, 3019–3021; Angew.
Chem. Int. Ed. 2005, 44, 2959–2962; c) M. T. Reetz, Angew. Chem.
2008, 120, 2592–2626; Angew. Chem. Int. Ed. 2008, 47, 2556–2588.
[7] D. PeÇa, A. J. Minnaard, J. A. F. Boogers, A. H. M. de Vries, J. G.
de Vries, B. L. Feringa, Org. Biomol. Chem. 2003, 1, 1087–1089.
[8] a) B. Breit, W. Seiche, J. Am. Chem. Soc. 2003, 125, 6608–6609;
b) B. Breit, W. Seiche, Angew. Chem. 2005, 117, 1666–1669; Angew.
Chem. Int. Ed. 2005, 44, 1640–1643; c) B. Breit, Angew. Chem.
2005, 117, 6976–6986; Angew. Chem. Int. Ed. 2005, 44, 6816–6825.
[9] a) J. M. Takacs, D. S. Reddy, S. A. Moteki, D. Wu, H. Palencia, J.
Am. Chem. Soc. 2004, 126, 4494–4495; b) J. M. Takacs, K. Chaisee-
da, S. A. Moteki, D. S. Reddy, D. Wu, K. Chandra, Pure Appl.
Chem. 2006, 78, 501–509; S. A. Moteki, K. Toyama, Z. Liu, J. Ma,
A. E. Holmes, J. M. Takacs, Chem. Commun. 2012, 48, 263–265.
[10] a) V. F. Slagt, P. W. N. M. van Leeuwen, J. N. H. Reek, Chem.
Commun. 2003, 2474–2475; b) M. Kuil, P. E. Goudriaan, P. W. N. M.
van Leeuwen, J. N. H. Reek, Chem. Commun. 2006, 4679–4681;
c) X.-B. Jiang, P. W. N. M. van Leeuwen, J. N. H. Reek, Chem.
Commun. 2007, 2287–2289.
[11] A. Franzke, A. Pfaltz, Chem. Eur. J. 2011, 17, 4131–4144.
[12] A. Franzke, A. Pfaltz, Synthesis 2008, 2, 245–252.
[13] B. Capuano, I. T. Crosby, E. J. Lloyd, D. A. Taylor, Aust. J. Chem.
2002, 55, 565–576.
[14] a) J. Scherer, G. Huttner, M. Bꢃchner, J. Bakos, J. Organomet.
Chem. 1996, 520, 45–58; b) V. V. Ovchinnikov, O. A. Cherkasova,
L. V. Verizhnikov, Zh. Obshch. Khim. 1982, 52, 707–708.
[15] M. J. Burk, G. Casy, N. B. Johnson, J. Org. Chem. 1998, 63, 6084–
6085.
General procedure for Rh-catalyzed hydrogenation reactions: The sub-
strate (0.33m, usually 200 mmol, 1.00 equiv), [RhACHTUNTGRNEUNG(cod)2]BF4 (0.010 equiv),
and the monodentate ligand(s) (0.020 equiv in total) were dissolved in
the desired solvent (usually 600 mL) in a 2 mL screw-cap vial equipped
with
a magnetic stirrer bar. Alternatively, stock solutions of [Rh-
ACHTUNGTRENNUNG(cod)2]BF4 and the olefin were sometimes used. Reactions in CH2Cl2,
EtOAc, and toluene were set up in a glove box in a nitrogen atmosphere.
Hydrogenation reactions in iPrOH were quickly set up in air. Four vials
were transferred into a 50 mL autoclave, which was pressurized with hy-
drogen (usually 10–20 bar). The reaction was started by switching on the
stirrer (700 minꢀ1). After the desired time, hydrogen was released and
hexanes or pentanes (1 mL) added. The resulting mixture was filtered
through a pad of silica gel, which was washed with hexanes/EtOAc or
pentanes/EtOAc (1:1). The filtrate was concentrated under reduced pres-
sure and the residue redissolved in Et2O, pentane, or EtOH (3 mL). The
conversion and enantioselectivity were directly determined by GC or
HPLC analysis (see the Supporting Information).
Acknowledgements
[16] J. L. Renaud, P. Dupau, A.-E. Hay, M. Guingouain, P. H. Dixneuf,
C. Bruneau, Adv. Synth. Catal. 2003, 345, 230–238.
[17] a) M. T. Reetz, T. Sell, R. Goddard, Chimia 2003, 57, 290–292;
b) M. T. Reetz, O. Bondarev, Angew. Chem. 2007, 119, 4607–4610;
Angew. Chem. Int. Ed. 2007, 46, 4523–4526.
Financial support by the Swiss National Science Foundation and the Uni-
versity of Basel is gratefully acknowledged. We thank Dr. Ivana Fleischer
for measuring the 31P NMR spectra of the ligand mixtures and Beat
Amrein for preparing batches of substrate 12.
[18] T. R. Wu, L. Shen, J. M. Chong, Org. Lett. 2004, 6, 2701–2704.
[19] Purification of Laboratory Chemicals (Eds.: D. D. Perrin, W. L. F.
Armarego), Pergamon Press, Oxford, 1988.
[20] A. B. Pangborn, M. A. Giardello, R. H. Grubbs, R. K. Rosen, F. J.
Timmers, Organometallics 1996, 15, 1518–1520.
[1] T. P. Dang, H. B. Kagan, J. Chem. Soc. Chem. Commun. 1971, 481.
[2] a) Comprehensive Asymmetric Catalysis, Vol. I–III (Eds.: E. N. Ja-
cobsen, A. Pfaltz, H. Yamamoto), Springer, Berlin, 1999; b) Asym-
metric Catalysis on Industrial Scales, (Eds.: H.-U. Blaser, E.
Schmidt), Wiley-VCH; Weinheim, 2004; c) W. S. Knowles, R.
Noyori, Acc. Chem. Res. 2007, 40, 1238–1239.
[3] a) M. van den Berg, A. J. Minnaard, E. P. Schudde, J. van Esch,
A. H. M. de Vries, J. G. de Vries, B. L. Feringa, J. Am. Chem. Soc.
2000, 122, 11539–11540; b) A. J. Minnaard, B. L. Feringa, L. Lefort,
J. G. de Vries, Acc. Chem. Res. 2007, 40, 1267–1277.
[21] H. E. Gottlieb, V. Kotylar, A. Nudelman, J. Org. Chem. 1997, 62,
7512–7515.
[22] The solvent shifts of d=5.32 and 54.0 ppm (for the 1H and 13C
nuclei, respectively) in CD2Cl2 were used to calibrate the spectra.
Received: July 6, 2012
Published online: December 27, 2012
Chem. Eur. J. 2013, 19, 2405 – 2415
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2415