Zwitterionic iridium complexes with P,N-ligands as catalysts for the asymmetric hydrogenation of alkenes

Article abstract of DOI:10.1002/chem.201003314

Several zwitterionic iridium complexes based on chiral P,N-ligands with imidazoline or oxazoline donors and anionic tetraarylborate or aryltrifluoroborate substituents have been synthesized. The corresponding cationic analogues have also been prepared, to evaluate the effect of the covalent linkage between the anion and the cationic metal complex in catalytic reactions. The respective pairs of structurally analogous precatalysts have been compared for their efficacies in the asymmetric hydrogenation of unfunctionalized olefins. In most cases, the anionic derivatization has virtually no influence on the asymmetric induction of the iridium complex. This is in accordance with X-ray structural studies, which have shown that the chiral environment of the cationic metal center is not affected by the anionic substituent. Depending on the nature of the counterion employed, the zwitterionic catalysts proved to be significantly more reactive than their cationic counterparts in nonpolar solvents. Copyright

Full text of DOI:10.1002/chem.201003314

FULL PAPER
DOI: 10.1002/chem.201003314
Zwitterionic Iridium Complexes with P,N-Ligands as Catalysts for the
Asymmetric Hydrogenation of Alkenes
Axel Franzke[a, b] and Andreas Pfaltz*[a]
Abstract: Several zwitterionic iridium
complexes based on chiral P,N-ligands
with imidazoline or oxazoline donors
and anionic tetraarylborate or aryltri-
fluoroborate substituents have been
synthesized. The corresponding cation-
ic analogues have also been prepared,
to evaluate the effect of the covalent
linkage between the anion and the cat-
ionic metal complex in catalytic reac-
tions. The respective pairs of structural-
ly analogous precatalysts have been
compared for their efficacies in the
asymmetric hydrogenation of unfunc-
tionalized olefins. In most cases, the
anionic derivatization has virtually no
influence on the asymmetric induction
of the iridium complex. This is in ac-
cordance with X-ray structural studies,
which have shown that the chiral envi-
ronment of the cationic metal center is
not affected by the anionic substituent.
Depending on the nature of the coun-
terion employed, the zwitterionic cata-
lysts proved to be significantly more re-
active than their cationic counterparts
in nonpolar solvents.
Keywords: asymmetric catalysis
·
borates · hydrogenation · iridium ·
zwitterions
Introduction
complexes of which catalyze several asymmetric transforma-
tions, such as cyclopropanations, allylic oxidations, Henry re-
actions, and enantioselective acylations of secondary alco-
hols and diols.[10]
The use of zwitterionic metal complexes as catalysts for or-
ganic transformations has attracted considerable attention in
the last twenty years. In particular, various well-defined sys-
tems for the polymerization of olefins with late transition
metals[1] or single-site metallocene and half-sandwich cata-
lysts[2] have been described. Furthermore, numerous anionic
ligands for reactions in aqueous media have been devel-
oped,[3] although the corresponding zwitterionic metal com-
plexes have seldom been studied in detail.
Apart from these two applications, there have only been a
few examples of the use of betaines as efficient catalysts.[4]
The zwitterionic rhodium species 1, which was described by
Schrock as early as 1970,[5] was applied by Alper in a multi-
tude of transformations, such as hydroformylations and hy-
drogenations.[6] Complexes such as 2, developed by Marder
and Baker, were used for rhodium-catalyzed hydro- and di-
borations,[7] whereas Petersꢀ borate derivatives
3 were
shown to be good catalysts for several addition reactions to
alkenes.[8] Stradiottoꢀs indenide-based zwitterions 4 effect
the reduction of alkenes and ketones.[9] Finally, our group
recently introduced anionic N,N-ligands 5 and 6, the copper
Compared with their cationic structural analogues, be-
taines are generally characterized by higher electron densi-
ties at their respective metal centers[8] as well as improved
solubilities in apolar media.[9c] However, regarding the use
of zwitterionic complexes as catalysts for hydrogenation re-
actions, the picture is less clear. Although ruthenium deriva-
tive 4 shows much higher activity than the corresponding
cationic complex in the transfer hydrogenation of ketones,[9a]
the related iridium betaine is significantly less efficient in
the reduction of styrene or cyclohexene.[9c]
[a] Dr. A. Franzke, Prof. Dr. A. Pfaltz
Department of Chemistry, University of Basel
St. Johanns-Ring 19, 4056 Basel (Switzerland)
Fax : (+41)61-267-1103
E-mail: andreas.pfaltz@unibas.ch
[b] Dr. A. Franzke
Current address:
Intermediates Research, BASF SE
67056 Ludwigshafen (Germany)
In the last decade, iridium complexes with chiral P,N-
ligands have emerged as efficient catalysts for the asymmet-
ric hydrogenation of unfunctionalized and certain function-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201003314.
Chem. Eur. J. 2011, 17, 4131 – 4144
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4131

alized olefins, which give unsatisafactory results with rhodi-
um and ruthenium diphosphine catalysts.[11] These precata-
lysts have been successfully used in the synthesis of various
natural products and useful chiral building blocks.[12] A pro-
nounced counterion effect has been observed in these reac-
tions, with bulky fluorinated aluminates[13] or borates such
tions.[19] Therefore, we decided to first study some cationic
PHIM complexes with different steric properties, in order to
identify the optimum ligand structure for the subsequent
studies. Based on established literature procedures,[20] the
three precatalysts 7a–c were easily synthesized in five steps
starting from ortho-fluorobenzamide (8) and the C2-symmet-
ric diamines 9a–c (Scheme 1). In the course of this work, it
was discovered that the nucleophilic aromatic substitution
to generate phosphanes 12a–c could be efficiently per-
formed even at room temperature, thus avoiding the rather
harsh literature conditions of refluxing THF.[20,21] Although
methods are also available for the synthesis of imidazolines
starting from amino alcohols,[22] the latter were not em-
ployed as chiral starting materials to avoid regioselectivity
problems in the benzylation reaction. The new iridium com-
plexes 7a–c were subsequently evaluated in the asymmetric
hydrogenation of some selected substrates (Table 1).[23]
Both the phenyl-substituted derivative 7b and the cyclo-
hexyl-annelated representative 7a proved to be very reac-
tive, but their enantioselectivities were moderate at best. In
contrast, the enantiomeric excesses induced by complex 7c
with tert-butyl moieties were promising. Especially for the
methylstilbene substrate 14 and the a,b-unsaturated ester
18, this complex clearly outperformed other structurally sim-
ilar catalysts (Table 1, entries 3 and 15).[19b,24] Unfortunately,
precatalyst 7c gave only low reactivity. Even under forcing
conditions (100 bar H2, 20 h), the conversions were not im-
proved significantly.
as tetrakis(pentafluorophenyl)
ACHTUNGTRENNUNG
borate[14] or tetrakis
ACHUTGTNRNEUNG CAHUTNGTRENN[GUN 3,5-bis-
most active catalysts. Although this phenomenon has been
studied in detail[16] and some experimental and theoretical
investigations concerning the mechanism of the hydrogena-
tion have been carried out,[17] the basis for this effect at the
molecular level is not yet fully understood.
In light of these facts, it seemed to be of considerable in-
terest to prepare zwitterionic derivatives of well-established
cationic iridium precatalysts with a covalently attached
anionic group and to study their properties in enantioselec-
tive hydrogenation reactions. If the anionic moiety could be
incorporated into the backbone or at the periphery of the
ligand, the formation of a tight ion pair with the cationic iri-
dium center should be precluded. Therefore, we envisaged
that the absence of a free counterion, which would other-
wise interact with the cationic iridium core, could result in
an increase of catalytic activity.
Herein, we report the development of iridium betaines
based on anionic phosphinooxazoline (PHOX) or phosphi-
noimidazoline (PHIM or BIPI) ligands and their application
in the asymmetric hydrogenation of unfunctionalized al-
kenes. Furthermore, structurally analogous cationic com-
plexes are also described, and their catalytic properties com-
pared with those of their zwitterionic counterparts.
To shed light on these unexpected reactivity differences,
the respective iridium complexes 7a and 7c were treated
with dihydrogen under atmospheric pressure for 15 min in
deuterodichloromethane and the resulting mixtures were an-
alyzed by NMR spectroscopy. Although the reactive deriva-
tive 7a had been transformed into two hydride-containing
species with 73% conversion,[25] the NMR spectrum of 7c
showed only the unchanged cyclooctadiene complex. This
result indicates that the low activity of 7c might be attribut-
ed to a slow, incomplete conversion to the catalytically
active species.
Results and Discussion
Evaluation of cationic PHIM complexes: In contrast to the
situation for the now well-established iridium PHOX deriva-
tives,[11a,12d,16,17c–e,18] there have only been a few reports con-
cerning the application of the corresponding imidazoline-
based systems in enantioselective hydrogenation reac-
Scheme 1. Preparation of cationic PHIM complexes 7a–c: i) Et3OBF4, CH2Cl2, RT, 19 h, 95%; ii) NEt3, CH2Cl2, RT, 64 h, 75–85% (11a, 11b) or EtOH,
D, 91 h, 64% (11c); iii) KPPh2, THF, D, 2 h, 88% (12b) or 08C!RT, 2.5–4 h, 93–97% (12a, 12c); iv) KH, THF (13a, 13b) or THF/DMF (20:1) (13c),
08C!RT, 2 h then BnBr, NBu4I, RT, 20 h, 73–96%; v) [Ir
ACHTUGNNERT(UNNG COD)Cl]2, CH2Cl2, D, 2 h then NaBArF, H2O, RT, 30 min, 95% to quantitative.
4132
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4131 – 4144

Zwitterionic Iridium Complexes with P,N-Ligands
FULL PAPER
Table 1. Asymmetric hydrogenation reactions with cationic PHIM com-
plexes 7a–c.
Entry Substrate
Precatalyst Conversion [%][a] ee [%][b]
1
2
3
7a
7b
7c
>99
>99
4
64 (S)
41 (S)
90 (S)
4
5
6
7a
7b
7c
>99
>99
54
31 (S)
53 (S)
84 (S)
7
8
9
7a
7b
7c
>99
>99
27
20 (R)
11 (R)
19 (R)
10
11
12
13
14
15
7a
7b
7c
7a
7b
7c
>99
>99
44
>99
>99
51
63 (+)
12 (ꢀ)
85 (+)
50 (S)
15 (S)
86 (S)
[a] Determined by GC. [b] Determined by HPLC on a chiral stationary
phase.
Scheme 2. Different classes of iridium zwitterions featuring P,N-ligands
derived from functionalized borate building blocks 19a and 19b.
In view of these results, further studies were focused on
the highly reactive cyclohexyl-annelated imidazoline com-
plexes and the more selective tert-butyl-substituted deriva-
tives.
Metalation of ligand 22a under the conditions described
in Scheme 1 for the synthesis of cationic iridium complexes
without addition of a base gave, besides the desired betaine
Synthesis of iridium betaines with P,N-ligands: Since cation-
ic complexes with fluorinated tetraarylborates as counter-
ions are among the most active catalysts for asymmetric hy-
drogenation reactions,[16] we decided to first prepare anionic
P,N-ligands with borate moieties at the periphery. In this
context, we have recently developed functionalized anionic
building blocks 19a and 19b (Scheme 2).[26] They resemble
structural analogues of the established tetrakis(pentafluoro)-
phenylborate (BACHTUNGTRENNUNG(C6F5)4) and tetrakisACHUTNGTREN[NNGU 3,5-bis(trifluorome-
thyl)phenyl]borate (BArF), and can easily be introduced in
neutral precursors through nucleophilic substitution reac-
tions.
Iridium betaines A, B, and C, containing both imidazoline
and oxazoline subunits, were chosen for these investigations
(Scheme 2). In complexes A and B, the covalently linked
anionic moieties are located at the rearside of the ligands, to
avoid any unfavorable interaction between the opposite
charges. Alternatively, the borate group can be attached to
the side chain of the oxazoline ring as in derivative C, in
which certain steric or electrostatic interactions of the anion
with the cationic metal center are possible.
The preparation of precatalysts 20a–21c as representa-
tives of type A, starting from neutral precursors 12a and
12c, was straightforward (Scheme 3). The intermediate
PHIM ligands were isolated as protonated betaines 22a–23c
after chromatography on silica gel.
Scheme 3. Synthesis of PHIM-based derivatives A: i) KH, THF (22a,
23a) or THF/DMF (20:1) (22c, 23c), 08C!RT, 2 h then 19a or 19b,
08C!RT, 18 h then SiO2, CH2Cl2, 86–97%; ii) [Ir
ACHTUNGTERN(UNGN COD)Cl]2, NEtiPr2,
CH2Cl2, D, 2 h then NaHCO3, H2O, RT, 30 min, 97% to quantitative.
Chem. Eur. J. 2011, 17, 4131 – 4144
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4133

A. Pfaltz and A. Franzke
20a, a second species in a ratio of 2:1. This side product was
shown to be the iridium(III) compound 24 by NMR spec-
Oxazoline-based derivatives 25 and 26, as representatives
of zwitterions B, were easily synthesized in six steps from
commercially available benzylic alcohol 27 (Scheme 4). The
condensation of nitrile 28 yielding oxazoline 29 was not
completed even after prolonged reaction times and 43% of
starting material 28 was recovered. The PHOX ligands were
partially obtained as the respective tetrabutylammonium
salt (33) or as a mixture of the latter and the protonated be-
taine in a ratio of 4:5 (32). This reflects the lower basicity of
oxazolines compared with their structurally related imidazo-
line counterparts.[27]
AHCTUNGTRENNUNG
troscopy, the configuration at the metal center of which
could not be established with certainty. Compound 24 repre-
sents the product of a formal oxidative addition of hydrogen
chloride to complex 20a. Its formation was completely sup-
pressed by the addition of a tertiary amine base.
To determine the influence of the anchoring point of the
anionic borate moiety on the properties of the precatalysts,
iridium complex 34 was prepared as an example of class C
complexes (Scheme 5).
Scheme 4. Preparation of PHOX betaines B: i) TBDMSCl, imidazole, CH2Cl2, 08C!RT, 2 h, 98%; ii) (S)-tert-leucinol, ZnCl2 (cat.), PhCl, D, 71 h, 51%;
iii) sBuLi, TMEDA, pentane, ꢀ788C, 60 min!08C, 15 min then o-Tol2PCl, 08C!RT, 15 h, 82%; iv) TBAF, THF, 08C!RT, 3.5 h, 89%; v) KH, THF,
08C!RT, 2 h then 19a or 19b, 08C!RT, 19 h then SiO2, CH2Cl2; vi) [Ir
ACHTNUGRTNE(NUGN COD)Cl]2, NEtiPr2, CH2Cl2, D, 2 h then NaHCO3, H2O, RT, 30 min, 71–92%
over two steps.
Scheme 5. Variation of the anchoring point: i) 2-FC6H4COCl, NEt3, MeOH, 08C, 2.5 h, 91%; ii) MeMgBr, THF/Et2O, 0 8C, 60 min!RT, 17 h then NH4Cl
(aq.), 08C, 70%; iii) TsCl, NEt3, CH2Cl2, RT, 75 h, 92%; iv) KPPh2, THF, 08C, 2 h, 77%; v) KH, THF, 08C!RT, 2 h then 19a, 08C!RT, 24 h, 83 %;
vi) [IrACHTUNGTRENNUNG(COD)Cl]2, NEtiPr2, CH2Cl2, D, 2 h then NaHCO3, H2O, RT, 30 min, 95%.
4134
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4131 – 4144

Zwitterionic Iridium Complexes with P,N-Ligands
FULL PAPER
The synthesis of phosphane 39 from (S)-serine methyl
ester hydrochloride (35) has been previously described by
Helmchen and co-workers.[28] As a major improvement of
their procedures, we found that the nucleophilic aromatic
substitution to give 39 can be accomplished in the presence
of the free hydroxyl group with high efficiency, avoiding
protection/deprotection steps. The intermediate PHOX
ligand was isolated as tetrabutylammonium salt 40 after
chromatography on silica gel.
Preparation of cationic analogues: The aim of these investi-
gations was to study the influence of the covalent linkage
between the cationic iridium complexes and the respective
anions on the reactivities and enantioselectivities in the hy-
drogenation of different olefins. Therefore, it was mandatory
to also synthesize the cationic structural analogues of all of
the zwitterions described above as reference systems
(Scheme 6).
Scheme 8. Synthesis of cationic PHIM reference systems 43a–44c: i) KH,
THF (45a, 45b) or THF/DMF (20:1) (45c), 08C!RT, 2 h then 4-
XC6F4CH2Br, 08C!RT, 17–24 h, 33–58%; ii) [Ir
ACHTUNGRTENN(NUG COD)Cl]2, CH2Cl2, D,
2 h then Li[BACHTNUGTRNEUNG(C6F5)4] or 42, H2O, RT, 30 min, 75–97%.
45a–c were only moderate (Scheme 8). Complexes 43a and
43b with X=F or H were prepared to exclude an unlikely
but possible influence of this substituent on the properties
of the catalysts.
Scheme 6. The betaines and their cationic structural analogues.
Initial attempts to prepare the cationic analogues 46 and
47 of PHOX-derived zwitterions 25 and 26 were unsuccess-
ful. When pentafluorobenzyl bromide was used in the ether-
ification, exclusively the dimeric P,N-ligand 48 was isolated
in good yield (Scheme 9). The formation of 48 is a result of
selective nucleophilic aromatic substitution of the para-
fluoro atom by the potassium alkoxide.
However, the corresponding reaction with 2,3,5,6-tetra-
fluorobenzyl bromide as the electrophile gave the desired
derivative 49, which could then be easily transformed into
complexes 46 and 47. Reference compound 51 was obtained
in the same manner.
Although the free anion for the representatives with Ar=
C6F5
corresponds
to
the
well-established
tetrakis(pentafluoro
N
(C6F5)4), the respective
counterion (C6F5)BAr3 for the derivatives with Ar=3,5-
(CF3)2C6H3 is not known in the literature. However, it was
easily synthesized via the reported aryltrifluoroborate 41
(Scheme 7).[29]
Enantioselective hydrogenation reactions in dichloro-
AHCTUNGTREGmNNNU ethane: To investigate the properties of the new iridium
betaines 20a–21c, 25, 26, and 34 and their cationic counter-
parts 43a–44c, 46, 47, and 51, the complexes were evaluated
for their efficacies in the asymmetric hydrogenation of a
series of representative alkenes. As a starting point, we ap-
plied the conditions that had proven to be optimal for cat-
ionic iridium catalysts.[11,16] In particular, dichloromethane
was used as solvent. To ensure maximum comparability of
the results, the respective reactions with the zwitterionic
complex and its cationic analogue were performed on the
same day, with the same batches of substrate and solvent,
Scheme 7. Preparation of borate 42: i) Mg, Et2O, D, 2 h then BACHTNUGTRENUNG(OiPr)3,
08C!RT, 20 h then KHF2, H2O, RT, 65 min, 55%; ii) ArMgBr, Et2O,
08C!RT, 6 h then Na2CO3, H2O, RT, 60 min, 67% (Ar=3,5-
(CF3)2C6H3).
With the required borates in hand, the cationic structural
analogues of PHIM-based betaines 43a–44c were easily ac-
cessible, although the yields of the intermediate P,N-ligands
Chem. Eur. J. 2011, 17, 4131 – 4144
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4135

A. Pfaltz and A. Franzke
Hydrogenations of other un-
functionalized acyclic or endo-
cyclic alkenes, such as 15 and
16 (Tables 3 and 4), allylic alco-
hol 17 (Table 5), and a,b-unsa-
turated ester 18 (Table 6), gave
similar results. PHIM com-
plexes 20c, 21c, 43c, and 44c
and PHOX derivatives 34 and
51 gave considerably higher
conversions with these sub-
strates, with the cationic sys-
tems being slightly more reac-
tive than their zwitterionic ana-
logues. Compared with their cy-
clohexyl-annelated counterparts
20a, 21a, 43a, 43b, and 44a,
PHIM precatalysts 20c, 21c,
43c, and 44c with tert-butyl
subunits were clearly more se-
lective. In general, PHOX-
based representatives 25, 26, 34,
46, 47, and 51 are considerably
more efficient in terms of both
reactivity and enantioselectivity,
Scheme 9. Preparation of cationic PHOX precatalysts: i) KH, THF, 08C!RT, 2 h then C6F5CH2Br, 08C!RT,
15 h, 72%; ii) KH, THF, 08C!RT, 2 h then 4-HC6F4CH2Br, 08C!RT, 19–23 h, 53–86%; iii) [Ir
CH2Cl2, D, 2 h then Li[B(C6F5)4] or 42, H2O, RT, 30 min, 97% to quantitative.
(COD)Cl]2, especially the latter.
ACHTUNGTRENNUNG
Table 2. Hydrogenation of (E)-1,2-diphenyl-1-propene (14).
and in the same autoclave.
Tables 2–6 show the results of
hydrogenations of olefins 14–
18, respectively.[23] For compari-
son, the corresponding data for
the standard PHOX catalyst 52
are also included.[11a,d]
Entry
Precatalyst
Conversion [%][a]
ee [%][b]
1
2
3
4
5
6
7
8
20a
43a
43b
21a
44a
20c
43c
21c
44c
25
46
26
47
34
51
52
>99
>99
>99
>99
>99
1
65 (S)
73 (S)
72 (S)
68 (S)
73 (S)
n.d.
In the asymmetric hydro-
AHCTUNGTRENGNgUN enation of (E)-1,2-diphenyl-1-
propene (14), a test reaction that has been extensively stud-
ied in the past,[11,16a,c] both cyclohexyl-annelated PHIM com-
plexes 20a, 21a, 43a, 43b, and 44a and PHOX precatalysts
25, 26, 46, and 47 based on tert-leucinol exhibited high reac-
tivities, yielding full conversions after 2 h at 50 bar hydrogen
pressure (Table 2, entries 1–5 and 10–13). Although the
PHOX complexes gave highly enantioenriched products
(Table 2, entries 10–13), the selectivities of the PHIM-based
representatives were moderate (Table 2, entries 1–5). Most
importantly, the data for the betaines and their cationic ref-
erence systems were virtually identical (Table 2, entries 10/
11 and 12/13) or at least very similar (Table 2, entries 1–3
and 4/5).
2
2
7
n.d.
n.d.
9
88 (S)
96 (R)
98 (R)
97 (R)
97 (R)
82 (S)
93 (S)
97 (R)
10
11
12
13
14
15
16[11a]
>99
>99
99
>99
34
99
>99
[a] Determined by GC. [b] Determined by HPLC on a chiral stationary
phase. n.d. denotes “not determined”.
In contrast, the pair of PHOX complexes 34 and 51 yield-
ed rather different results, with cationic derivative 51 being
more efficient (Table 2, entries 14 and 15). The tert-butyl-
substituted PHIM complexes 20c, 21c, 43c, and 44c gave
only trace amounts of product under the chosen reaction
conditions (Table 2, entries 6–9).
It should be mentioned that the lower conversions of sub-
strates 14, 17, and 18 with iridium betaine 34 were easily en-
hanced by increasing the reaction time without affecting the
enantioselectivities.[23] This indicates that the diminished re-
4136
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4131 – 4144

Zwitterionic Iridium Complexes with P,N-Ligands
FULL PAPER
Table 3. Hydrogenation of (E)-2-(4’-methoxyphenyl)-2-butene (15).
Table 5. Hydrogenation of (E)-2-methyl-3-phenyl-2-propenol (17).
Entry
Precatalyst
Conversion [%][a]
ee [%][b]
Entry
Precatalyst
Conversion [%][a]
ee [%][b]
1
2
3
4
5
6
7
8
20a
43a
43b
21a
44a
20c
43c
21c
44c
25
46
26
47
34
51
52
>99
>99
>99
>99
>99
8
28
15
44
>99
>99
>99
>99
73
54 (+)
58 (+)
57 (+)
58 (+)
57 (+)
76 (+)
69 (+)
80 (+)
71 (+)
93 (ꢀ)
94 (ꢀ)
94 (ꢀ)
94 (ꢀ)
69 (+)
82 (+)
96 (ꢀ)
1
2
3
4
5
6
7
8
20a
43a
43b
21a
44a
20c
43c
21c
44c
25
46
26
47
34
51
52
>99
>99
>99
>99
>99
16
60
9
49
>99
>99
>99
>99
>99
>99
>99
22 (S)
25 (S)
22 (S)
21 (S)
24 (S)
73 (S)
71 (S)
66 (S)
65 (S)
70 (R)
74 (R)
68 (R)
71 (R)
93 (S)
93 (S)
61–81 (R)
9
10
11
12
13
14
15
16[11a]
9
10
11
12
13
14
15
16[11a,d]
>99
95
[a] Determined by GC. [b] Determined by HPLC on a chiral stationary
phase.
[a] Determined by GC. [b] Determined by HPLC on a chiral stationary
phase.
Table 6. Hydrogenation of (E)-ethyl 3-phenyl-2-butenoate (18).
Table 4. Hydrogenation of 6-methoxy-1-methyl-3,4-dihydronaphthalene
(16).
Entry
Precatalyst
Conversion [%][a]
ee [%][b]
1
2
3
4
5
6
7
8
20a
43a
43b
21a
44a
20c
43c
21c
44c
25
46
26
47
34
51
52
>99
>99
>99
>99
>99
7
23
12
45
>99
>99
>99
>99
74
46 (S)
46 (S)
46 (S)
50 (S)
47 (S)
70 (S)
84 (S)
76 (S)
86 (S)
81 (R)
80 (R)
83 (R)
81 (R)
26 (S)
40 (S)
84 (R)
Entry
Precatalyst
Conversion [%][a]
ee [%][b]
1
2
3
4
5
6
7
8
20a
43a
43b
21a
44a
20c
43c
21c
44c
25
46
26
47
34
51
52
>99
>99
>99
>99
>99
8
26
13
49
>99
>99
>99
>99
97
17 (R)
14 (R)
15 (R)
19 (R)
15 (R)
10 (S)
15 (R)
rac.
21 (R)
67 (S)
73 (S)
69 (S)
72 (S)
75 (R)
71 (R)
72 (S)
9
10
11
12
13
14
15
16[11a]
9
10
11
12
13
14
15
16[11d]
99
96
[a] Determined by GC. [b] Determined by HPLC on a chiral stationary
phase.
98
>99
[a] Determined by GC. [b] Determined by HPLC on a chiral stationary
phase.
enantioselectivities afforded by PHOX-derived betaines 25
and 26 or PHIM zwitterions 20a and 21a and their cationic
reference systems 43a, 43b, 44a, 46, and 47. In addition, all
of these derivatives show approximately the same efficien-
cies, yielding full conversions with 1 mol% catalyst loading.
Finally, it is evident from the consistency of results with de-
rivatives 43a and 43b that the para-substituent on the
benzyl moiety in the ligands exerts no electronic influence
on the properties of the respective metal complexes.
activity of 34 may be attributed to a lower turnover frequen-
cy rather than a shorter lifetime of the catalytically active
species.
Based on the data presented in Tables 2–6, the following
conclusions can be drawn. Covalent linkage of the negative
moiety at the rearside of the ligand has, as anticipated, no
pronounced influence on the chiral pocket around the cata-
lytically active metal center. This has clearly been demon-
strated by the virtually identical or at least very similar
In contrast, the differences between PHOX betaine 34
and its cationic analogue 51 with respect to both reactivity
Chem. Eur. J. 2011, 17, 4131 – 4144
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4137

A. Pfaltz and A. Franzke
Table 8. Hydrogenations with PHOX complexes 25 and 46 in different
solvents.
and selectivity show that the location of the anchoring point
of the negatively charged subunit on the ligand backbone
has a significant impact on the efficiency of the catalyst,
since an interaction between the cationic metal center and
the sterically demanding anionic subunit is rendered possi-
ble in complex 34.
The generally low conversions and rather variable selec-
tivities of PHIM complexes 20c, 21c, 43c, and 44c might be
attributed to steric overcrowding of the catalytically active
metal center due to the accumulation of bulky substituents
on the imidazoline ring.
Entry
Solvent
Precatalyst
Conversion [%][a]
ee [%][b]
1
2
3
4
5
6
7
8
dichloromethane
dichloromethane
toluene
toluene
hexane
25
46
25
46
25
46
25
46
25
46
25
46
>99
>99
8
97
>99
8
>99
7
>99
12
61
9
96 (R)
98 (R)
93 (R)
97 (R)
78 (R)
85 (R)
77 (R)
94 (R)
79 (R)
89 (R)
78 (R)
88 (R)
hexane
Variation of the reaction conditions: To further evaluate the
properties of the new iridium betaines, compounds 20a and
25 were selected for further studies of the asymmetric re-
duction of stilbene-type olefin 14 in various, predominantly
apolar solvents. The results from these experiments are sum-
marized and compared with the respective data for the cor-
responding cationic complexes 43a and 46 in Tables 7 and 8.
cyclohexane
cyclohexane
pentane
pentane
pentane
9
10
11[c]
12[c]
pentane
[a] Determined by GC. [b] Determined by HPLC on a chiral stationary
phase. [c] 0.3 mol% catalyst.
Table 7. Hydrogenations with PHIM complexes 20a and 43a in different
solvents.
and 46 still gave high enantiomeric excesses of 77–94% in
pure hydrocarbons, although the differences between the re-
spective values of the two structural analogues were in-
creased (Table 8, entries 5–12). In contrast, the selectivities
of the PHIM systems were substantially eroded and even
the direction of the asymmetric induction was inverted
(Table 7, entries 5–12). In toluene and hydrocarbons, the
(R)-product was obtained preferentially, although with very
low enantiomeric excess.
The superior catalytic performances of betaines 20a and
25 in pure hydrocarbons can most likely be attributed to the
enhanced solubilities of these formally uncharged species in
solvents of low polarities, although even with catalyst load-
ings as low as 0.3 mol% the reaction mixtures in pentane re-
mained heterogeneous. Furthermore, the drastic variations
in enantioselectivity seen in the different solvents indicate
that the chiral environment of the iridium center is highly
sensitive to the reaction medium. Finally, it has to be point-
ed out that, regardless of the solvent used, the new iridium
betaines are far more reactive than the established inden-
ide-based system 4, which shows no conversion with trisub-
stituted olefins such as 1-methylcyclohexene.[9c]
Owing to their high reactivities, no pronounced differen-
ces in productivity were detected between the iridium be-
taines and their cationic reference compounds in dichloro-
methane at 1 mol% catalyst loading (Tables 3–6). Therefore,
the kinetic profiles of the selected representatives 21a, 26,
44a, and 47 in this solvent were investigated in more detail
with only 0.02 mol% catalyst loading (Figure 1 and
Table 9).[30] Every data point in Figure 1 represents an inde-
pendent catalytic experiment, which was worked-up and an-
alyzed after the time indicated. Table 9 contains approxi-
mate values for the turnover frequencies and numbers of
these four iridium complexes.
Entry
Solvent
Precatalyst
Conversion [%][a]
ee [%][b]
1
2
3
4
5
6
7
8
dichloromethane
dichloromethane
toluene
toluene
hexane
20a
43a
20a
43a
20a
43a
20a
43a
20a
43a
20a
43a
>99
>99
95
79
>99
45
>99
25
>99
57
93
24
65 (S)
73 (S)
22 (R)
24 (R)
9 (R)
14 (R)
8 (R)
3 (R)
hexane
cyclohexane
cyclohexane
pentane
pentane
pentane
9
13 (R)
14 (R)
8 (R)
10
11[c]
12[c]
pentane
13 (R)
[a] Determined by GC. [b] Determined by HPLC on a chiral stationary
phase. [c] 0.3 mol% catalyst.
Interestingly, zwitterions 20a and 25 maintained their
high levels of activity even in pure hydrocarbons, whereas
the conversions achieved with cationic complexes 43a and
46 decreased sharply. This trend was more pronounced for
the PHOX-derived precatalysts, with betaine 25 showing
about tenfold higher reactivity in hexane, cyclohexane, or
pentane relative to 46 (Table 8, entries 5–10). The turnover
numbers for the zwitterions in pentane after reaction times
of 2 h were 310 (20a, Table 7, entry 11) and 200 (25, Table 8,
entry 11), respectively. The reason for the negligible activity
of complex 25 in toluene (Table 8, entry 3) remains unclear,
but p-complexes of the aromatic solvent with some inter-
mediates in the catalytic cycle might play a role here.
The selectivities of all four precatalysts deteriorated with
decreasing polarities of the solvents. PHOX derivatives 25
The enantioselectivities of the four catalysts remained
almost constant throughout the reactions and were in excel-
4138
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4131 – 4144

Zwitterionic Iridium Complexes with P,N-Ligands
FULL PAPER
crystals from cationic complexes with the BArF anion failed,
hexafluorophosphate salts 53a and 53b were prepared,
which eventually provided single crystals suitable for X-ray
analysis. Figure 2 shows the solid-state structures of precata-
lysts 53a, 53b, and 7c with different substituents on the imi-
dazoline moieties. Selected metric parameters of these com-
plexes are summarized in Table 10.
Figure 1. Kinetic profiles for 21a (blue diamonds), 26 (green triangles),
44a (red squares), and 47 (brown circles) for the hydrogenation of 14 in
dichloromethane with 0.02 mol% catalyst loading at room temperature
and 50 bar hydrogen pressure.
Table 9. Enantioselectivities, turnover frequencies (TOF), and turnover
numbers (TON) for the hydrogenations of 14 shown in Figure 1.
[b]
Entry
Precatalyst
ee [%][a]
TOF [hꢀ1
]
TON[c]
1
2
3
4
21a
44a
26
71–68 (S)
75–73 (S)
98–96 (R)
97–96 (R)
2060
1640
4620
7340
3230
3895
1460
2260
Figure 2. General atomic numbering scheme and crystal structures of cat-
ionic iridium complexes 53a, 53b, and 7c. The counterions, the cycloocta-
diene fragments, all hydrogen atoms, and solvent molecules eventually
present have been omitted for clarity.
47
[a] Determined by HPLC on a chiral stationary phase. [b] Determined at
t=15 min. [c] Determined at t=10 h (26, 47) or t=48 h (21a, 44a).
lent agreement with the values obtained in the experiments
with 1 mol% catalyst loading (Table 2, entries 4, 5, 12, and
13). In both cases, the cationic complexes were somewhat
more efficient than the corresponding zwitterions, resulting
in about 21–55% higher turnover numbers.
Table 10. Selected bond lengths [ꢂ] and angles [8] for precatalysts 53a,
53b, and 7c.[a]
Complex
53a
53b
7c
anion
PF6
PF6
BArF
ꢀ
Ir1 P1
2.285(1)
2.083(3)
2.201(5)
2.157(3)
2.165(4)
2.116(4)
81.09(9)
48.0
2.280(1)
2.091(3)
2.216(3)
2.199(4)
2.129(4)
2.106(4)
86.63(9)
47.9
2.2717(7)
2.091(3)
2.203(3)
2.177(4)
2.150(4)
2.110(4)
86.08(8)
40.7
However, there are striking differences between the kinet-
ic profiles of the PHOX- and PHIM-based catalysts. Al-
though PHOX complexes 26 and 47 exhibit very high initial
rates, their lifetimes are quite short and maximum conver-
sions are reached within the first hour. In contrast, the turn-
over frequencies of the PHIM complexes are lower, but
they remain active even after prolonged reaction times.
Most interestingly, their TOFs stay more or less constant
during the final 38 h of the transformations (34 hꢀ1 for 21a
and 43 hꢀ1 for 44a). This indicates that deactivation, which
is a well-known problem of iridium complexes,[16,31] is much
slower in the case of the PHIM-based catalysts. After an ini-
tial phase, during which the activity decreases, a steady-state
concentration of catalytically active species is reached and
the TOF value remains virtually constant. Thus, owing to
their enhanced stabilities, the PHIM complexes are more
productive catalysts than the PHOX analogues, even though
their initial reactivities are markedly lower.
ꢀ
Ir1 N1
ꢀ
Ir1 C1 (trans to P)
ꢀ
Ir1 C2 (trans to P)
ꢀ
Ir1 C3 (trans to N)
ꢀ
Ir1 C4 (trans to N)
P1-Ir1-N1
C5-C6-C7-N1[b]
C8-C9-P1-C10[b]
C8-C9-P1-C11[b]
ꢀ117.8
ꢀ102.0
ꢀ93.2
ꢀ7.2
7.7
15.9
[a] Atomic numbering according to Figure 2. [b] Determined with the
help of WebLabViewerLite 3.20.
The crystal structures of the three complexes are very
similar. Their six-membered chelate rings exhibit distorted
ꢀ
ꢀ
boat conformations with almost identical Ir1 P1 and Ir1 N1
bond lengths, although 53a shows a slightly smaller bite
angle P1-Ir1-N1. All torsional angles C5-C6-C7-N1 between
the bridging phenyl ring and the imidazoline subunit are in
ꢀ
the range 41–488. As expected, the Ir C bonds trans to the
ꢀ
ꢀ
X-ray crystallographic studies of PHIM-derived complexes:
To shed light on the molecular basis for the observed trends
in reactivity and enantioselectivity of the new catalysts, crys-
tal structures of representative PHIM-based iridium com-
plexes were determined.[32] Since attempts to obtain suitable
phosphorus donor (Ir1 C1 and Ir1 C2) are somewhat
ꢀ
longer than those trans to the nitrogen atom (Ir1 C3 and
Ir1 C4). Finally, the torsional angles C8-C9-P1-C10 and C8-
C9-P1-C11 in the series 53a, 53b, and 7c change in such a
way that the distances between the P-phenyl groups and the
ꢀ
Chem. Eur. J. 2011, 17, 4131 – 4144
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4139

A. Pfaltz and A. Franzke
progressively more bulky substituents on the imidazoline
rings increase, to minimize unfavorable steric interactions.
The solid-state structure of zwitterion 20c was also deter-
mined. Figure 3 shows one of the two almost identical mole-
cules in the asymmetric unit, together with the crystal struc-
ture of the hexafluorophosphate salt 54. Table 11 lists the
corresponding metric parameters.
Examination of more strongly coordinating counterions: In
the comparative studies of the borate-based zwitterionic
precatalysts and their cationic structural analogues, the gen-
erally very high activities turned out to be a major problem.
To establish the impact of the covalent linkage of the two
opposite charges, the catalyst loadings had to be lowered to
such an extent that disturbing factors such as traces of water
or other impurities might have obscured the effects of inter-
est. Therefore, the corresponding complexes with more co-
ordinating counterions, which were expected to have stron-
ger interactions between the anion and the iridium center
and thus be less active, were also investigated.
After derivatives containing sulfonate moieties had
proven to be almost inactive in the hydrogenation of trisub-
stituted unfunctionalized olefins, we turned our attention to
tetrafluoroborate and aryltrifluoroborates as the immobiliz-
able variants of the former.[16a,33] The preparations of betaine
55 and its cationic counterpart 56, which represent structural
analogues of precatalysts 20a, 21a, 43a, 43b, and 44a with
fluorinated
(Scheme 10).
tetraarylborates,
were
straightforward
Figure 3. Crystal structures of precatalysts 20c and 54. The counterion of
54, the cyclooctadiene fragments, all hydrogen atoms, and solvent mole-
cules eventually present have been omitted for clarity.
Table 11. Selected bond lengths [ꢂ] and angles [8] for betaine 20c and
its cationic structural analogue 54.[a]
Complex
20c
54
ꢀ
Ir1 P1
2.285(1)/2.276(1)
2.100(4)/2.113(4)
2.206(5)/2.221(4)
2.188(5)/2.212(4)
2.141(6)/2.158(5)
2.110(6)/2.124(5)
86.0(1)/85.1(1)
38.0/38.5
2.293(1)
2.101(3)
2.188(4)
2.127(4)
2.171(6)
2.154(5)
85.98(8)
38.9
ꢀ
Ir1 N1
ꢀ
Ir1 C1 (trans to P)
ꢀ
Ir1 C2 (trans to P)
ꢀ
Ir1 C3 (trans to N)
ꢀ
Ir1 C4 (trans to N)
P1-Ir1-N1
C5-C6-C7-N1[b]
C8-C9-P1-C10[b]
C8-C9-P1-C11[b]
ꢀ90.9/ꢀ92.0
ꢀ91.6
19.3/15.9
14.1
[a] Atomic numbering according to Figure 2. [b] Determined with the
help of WebLabViewerLite 3.20.
The coordination geometries of these two structurally
analogous precatalysts are virtually identical. Furthermore,
the respective parameters are very similar to those of the
tert-butyl-substituted PHIM complex 7c (Figure 2 and
Table 10). The similarity of the coordination spheres, togeth-
er with the almost identical enantioselectivities provided by
the zwitterionic derivatives and their cationic analogues, in-
dicates that the covalent linkage of the counterion at the
rearside of the P,N-ligand has hardly any influence on the
chiral environment of the metal center. The cyclooctadiene
ligand of complex 54 is strongly twisted out of the plane of
an ideally square-planar coordination geometry, such that,
against all expectations, the Ir-C2 bond trans to the P atom
is shorter than the respective distances trans to the N atom.
This shows the limitations of assessing the characteristics of
a specific donor with the help of these values.
Scheme 10. Synthesis of aryltrifluoroborate-based precatalysts 55 and 56:
i) KH, THF, 08C!RT, 2 h then 4-BrC6H4CH2Br, NBu4I, RT, 19 h, 86%;
ii) nBuLi, THF, ꢀ788C, 90 min then
BG
90 min then KHF2, H2O, RT, 2 h; iii) [Ir
2 h then NaHCO3, H2O, RT, 30 min, about 62% over two steps; iv) [Ir-
(COD)Cl]2, CH2Cl2, D, 2 h then K[p-TolBF3], NaHCO3, H2O, RT, 30 min,
53%.
ACHTUNGTRENGNUN(COD)Cl]2, NEtiPr2, CH2Cl2, D,
A
ACHTUNGTRENNUNG
Lithiation of aryl bromide 57 and subsequent transforma-
tion into aryltrifluoroborate 58 provided, as a result of the
acidity of potassium hydrogen difluoride, not the potassium
salt but the protonated imidazoline 58, as indicated by IR
spectroscopy. This was used in the next step without further
purification. Despite substantial experimentation, zwitterion
4140
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4131 – 4144

Zwitterionic Iridium Complexes with P,N-Ligands
FULL PAPER
55 always contained a minor amount of a second species
after recrystallization, which could not be removed. The
latter was presumably a hydrolysis product with an unspeci-
fied counterion.
more strongly coordinating anion. This indicates that the
values presented in Table 12 for betaine 55 actually repre-
sent a lower limit for the conversions that would be ob-
tained with a pure catalyst.
A slight variation of the standard procedure gave the
analogous cationic complex 56 with trifluoro-para-tolyl-
A
Conclusion
fragment in the solid state proved to be virtually identical to
the structure determined for the corresponding hexafluoro-
phosphate derivative 53a.[34]
Notwithstanding the impurity in all samples of complex
55, the two precatalysts 55 and 56 were evaluated for their
efficacies in the asymmetric hydrogenation of selected ole-
fins (Table 12).[23]
Eight zwitterionic iridium complexes comprising chiral P,N-
ligands with imidazoline or oxazoline nitrogen donors have
been prepared in five to six steps in 76–26% overall yield
using our recently developed borate building blocks. They
have been evaluated as precatalysts for the asymmetric hy-
drogenation of unfunctionalized olefins, and the results have
been compared with respective data for the corresponding
cationic structural analogues.
Table 12. Asymmetric reductions with aryltrifluoroborate-based com-
The catalytic experiments showed that anionic functionali-
zation of the complexes at the rearside of the ligands, that
is, pointing away from the catalytically active metal centers,
has virtually no influence on their asymmetric inductions.
The reactivities vary with the anionic moieties and the sol-
vent. Although the tetraarylborate-based betaines show
slightly lower turnover numbers in dichloromethane, they
retain high activities in pure aliphatic hydrocarbons, in
which the cationic analogues are almost inactive. In contrast,
the zwitterion with an anionic aryltrifluoroborate subunit
exhibits up to fourfold higher reactivity even in dichloro-
plexes 55 and 56.
Entry Substrate
Precatalyst Conversion [%][a] ee [%][b]
1
2
55
56
55
56
55
56
45
11
91
28
89
46
52 (S)
59 (S)
54 (S)
59 (S)
16 (S)
22 (S)
3[c]
4[c]
5
AHCTUNGTRENNUNG
6
as anions are superior catalysts in moderately polar solvents
such as dichloromethane, whereas the new iridium betaines
show promise for reactions in nonpolar media such as pure
hydrocarbons.
7
8
55
56
51
13
27 (R)
28 (R)
9
10
55
56
89
56
17 (R)
21 (R)
11
12
55
56
97
99
62 (+)
66 (+)
Experimental Section
13
14
55
56
84
39
50 (S)
49 (S)
General: All reactions were performed in flame-dried glassware under
argon using Schlenk techniques. Solvents, NEt3, NEtiPr2, and TMEDA
were dried according to standard procedures and distilled under nitrogen
or argon.[35] All other commercial reagents were used as received. Deu-
terated solvents for NMR spectroscopy were degassed by three freeze-
pump-thaw cycles, dried over 4 ꢂ molecular sieves, and stored under
argon. Solvents for the work-up and chromatographic purification of air-
sensitive compounds were purged with a stream of argon for at least
15 min prior to use. Catalytic hydrogenations were set up under a nitro-
gen atmosphere in an MBraun Labmaster 130 glovebox using absolute
solvents purchased from Fluka. Chromatographic separations were per-
formed on silica gel 60 (Merck, Darmstadt; 40–63 nm). Pre-coated Ma-
cherey–Nagel Polygram SIL G/UV254 plates were used for TLC analyses,
and the compounds were visualized with the aid of UV light. NaBArF,[36]
[a] Determined by GC. [b] Determined by HPLC on a chiral stationary
phase. [c] 0.2 mol% catalyst loading, 24 h.
With all substrates except allylic alcohol 17, zwitterion 55
gave substantially higher conversions than the cationic ana-
logue 56. In the case of methylstilbene (14), the maximum
TONs were 455 for 55 and 140 for 56, respectively
(Table 12, entries 3 and 4). The enantioselectivities were in
the same range and varied within the same limits as the
values for the structurally related systems 20a, 21a, 43a,
43b, and 44a (Tables 2–6).
Finally, the hydrogenation of olefin 14 using several
batches of 55 with different amounts of the unknown impur-
ity showed that conversion increased with decreasing pro-
portion of the side product, which most likely contains a
Li[B
G
and
(3R,4R)-3,4-diamino-2,2,5,5-tetramethylhexane
(9c)[38] were prepared according to modified literature procedures.
NMR experiments were performed on Bruker Avance 400, 500, or 600
spectrometers. 1H and 13C spectra were referenced relative to SiMe4
using the solvent residual peaks and the solvent signals, respectively, as
internal standards.[39,40] 31P, 19F, and 11B spectra were calibrated using
H3PO4 (85%), CFCl3, and BF3·OEt2 as external standards. Only multip-
lets other than singlets are specified explicitly in the 13C, 31P, and
11B NMR data. Mass spectra were measured on VG70–250, Finnigan
MAT 95Q (EI), Finnigan MAT 312, Finnigan MAR 8400 (FAB), or Fin-
Chem. Eur. J. 2011, 17, 4131 – 4144
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4141

A. Pfaltz and A. Franzke
nigan MAT LCQ spectrometers (ESI). Elemental analyses were per-
formed by the staff of the Micro Analysis Laboratory at the University
of Basel. IR spectra were measured on a Perkin–Elmer 1600 FTIR spec-
trometer. Melting points were determined on a Bꢃchi 535 apparatus and
are uncorrected. Specific rotations were measured on a Perkin–Elmer
314 polarimeter. HPLC analyses were performed on a Shimadzu system,
GC measurements on equipment from Carlo Erba Instruments. The ab-
trated under reduced pressure. Purification of the crude product by
column chromatography under argon (silica gel, 4ꢄ23 cm; CH2Cl2) gave
precatalyst 43a as a red solid (253 mg, 82%). RF =0.65 (CH2Cl2, tailing);
[a]2D0 =+209 (c=0.230 in CHCl3); 1H NMR (500.1 MHz, CDCl3, 295 K):
d=1.05 (mc, 1H; CH(N)CH2CHH), 1.17–1.47 (m, 4H; CH(N)CHHCHH
and COD-CHH), 1.56 (mc, 1H; COD-CHH), 1.70 (d, J=11.4 Hz, 1H;
CH(N)CHH), 1.77 (d, J=13.7 Hz, 1H; CH(N)CH2CHH), 1.87 (d, J=
12.9 Hz, 1H; CH(N)CH2CHH), 1.90–2.04 (m, 3H; COD-CHH and Im-
7a-H), 2.21–2.48 (m, 4H; COD-CH2), 2.65 (d, J=11.1 Hz, 1H;
CH(N)CHH), 3.09 (mc, 1H; COD-CH), 3.30 (t, J=12.6 Hz, 1H; Im-3a-
H), 3.88 (d, J=15.5 Hz, 1H; Bn-CHH), 3.93 (brs, 1H; COD-CH), 4.46
(d, J=15.6 Hz, 1H; Bn-CHH), 4.52 (quint, J=7.1 Hz, 1H; COD-CH),
5.35 (brs, 1H; COD-CH), 7.13 (dd, J=10.7, 7.3 Hz, 2H; PPh2-o-H),
7.44–7.68 (m, 11H; Ar-H), 7.70 ppm (dd, J=7.5, 3.4 Hz, 1H; Ar-6’’-H);
13C{1H} NMR (125.8 MHz, CDCl3, 295 K): d=24.3 (CH(N)CH2CH2),
25.1 (CH(N)CH2CH2), 25.7 (COD-CH2), 28.1 (COD-CH2), 29.9
(CH(N)CH2), 33.8 (CH(N)CH2), 34.6 (COD-CH2), 37.2 (d, J=5 Hz;
COD-CH2), 41.6 (Bn-CH2), 63.0 (COD-CH), 65.7 (COD-CH), 70.2 (Im-
3a-CH), 72.4 (Im-7a-CH), 86.0 (d, J=16 Hz; COD-CH), 95.7 (d, J=
9 Hz; COD-CH), 108.4 (t, J=19 Hz; Bn-i-C), 121.5 (d, J=52 Hz; PPh2-i-
C), 124.1 (br; C6F5-i-C), 125.6 (d, J=49 Hz; Ar-2’-C), 129.2 (d, J=10 Hz;
Ar-CH), 129.5 (d, J=52 Hz; PPh2-i-C), 129.5 (d, J=11 Hz; Ar-CH),
130.0 (brd, J=7 Hz; Ar-6’-CH), 131.2 (Ar-CH), 132.2–132.3 (several Ar-
CH and Ar-1’-C), 132.7 (d, J=7 Hz; Ar-CH), 132.8 (d, J=2 Hz; Ar-CH),
133.7 (d, J=10 Hz; PPh2-o-CH), 136.3 (dmc, J=244 Hz; C6F5-m-C), 137.7
(dmc, J=253 Hz; Bn-m-C), 138.3 (dmc, J=244 Hz; C6F5-p-C), 141.7 (dmc,
J=257 Hz; Bn-p-C), 145.3 (dmc, J=250 Hz; Bn-o-C), 148.3 (dmc, J=
241 Hz; C6F5-o-C), 165.6 ppm (d, J=6 Hz; Im-2-C); 19F{1H} NMR
(376.5 MHz, CDCl3, 300 K): d=ꢀ167.2 (brs, 8F; C6F5-m-F), ꢀ163.5 (t,
J=20 Hz, 4F; C6F5-p-F), ꢀ160.0 (mc, 2F; Bn-m-F), ꢀ151.4 (t, J=21 Hz,
1F; Bn-p-F), ꢀ141.3 (mc, 2F; Bn-o-F), ꢀ132.9 ppm (brs, 8F; C6F5-o-F);
31P{1H} NMR (202.5 MHz, CDCl3, 295 K): d=23.8 ppm; IR (KBr): n˜ =
3062, 2946, 2884, 1644, 1513, 1463, 1366, 1306, 1272, 1090, 1021, 980, 774,
753, 695, 663, 606, 571, 539, 492 cmꢀ1; MS (ESI+, CH2Cl2): m/z (%)=
breviation BArF refers to the tetrakisACTHNUTRGENUGN[3,5-bis(trifluoromethyl)phenyl]bo-
rate anion, whilst ArF denotes the 3,5-bis(trifluoromethyl)phenyl sub-
stituent in general and ArF any arbitrary fluorinated aryl moiety. The
term dmc refers to a doublet of centered multiplets.
General procedure for the synthesis of zwitterionic complexes
Compound 20a: Absolute NEtiPr2 (153 mL, 0.900 mmol) followed by
ligand 22a (318 mg, 0.300 mmol) in absolute CH2Cl2 (3 mL) were added
dropwise to a solution of [IrACHTUNTGRENUNG(COD)Cl]2 (111 mg, 0.165 mmol) in absolute
CH2Cl2 (5 mL) at room temperature. After the red mixture had been
stirred in a closed vessel for 2 h at 508C, it was cooled to room tempera-
ture, whereupon saturated aqueous NaHCO3 (6 mL) and H2O (2 mL)
were added. The resulting two-phase system was vigorously stirred for
30 min at room temperature, the phases were allowed to separate, and
the aqueous layer was extracted with CH2Cl2 (3ꢄ10 mL). The combined
organic phases were dried over MgSO4, filtered, and concentrated under
reduced pressure. Purification of the crude product by column chroma-
tography under argon (silica gel, 4ꢄ22 cm; CH2Cl2) yielded precatalyst
20a as a red, foamy solid (397 mg, 97%). RF =0.77 (CH2Cl2); [a]2D0 =+
233 (c=0.250 in CHCl3); 1H NMR (500.1 MHz, CDCl3, 295 K): d=1.02
(mc, 1H; CH(N)CH2CHH), 1.15–1.34 (m, 3H; CH(N)CHHCHH), 1.39
(mc, 1H; COD-CHH), 1.54 (mc, 1H; COD-CHH), 1.71–1.89 (m, 3H;
CH(N)CHHCHH), 1.90–2.04 (m, 2H; COD-CHH), 2.08 (t, J=12.8 Hz,
1H; Im-7a’-H), 2.22–2.50 (m, 4H; COD-CH2), 2.58 (brd, J=6.8 Hz, 1H;
CH(N)CHH), 3.07 (brs, 1H; COD-CH), 3.26 (t, J=11.8 Hz, 1H; Im-3a’-
H), 3.84 (brs, 1H; COD-CH), 4.01 (d, J=15.3 Hz, 1H; Bn-CHH), 4.33
(d, J=15.4 Hz, 1H; Bn-CHH), 4.54 (mc, 1H; COD-CH), 5.29 (brs, 1H;
COD-CH), 7.11 (t, J=9.0 Hz, 2H; PPh2-o-H), 7.40–7.73 ppm (m, 12H;
865 (100) [MꢀB
(C6F5)4]+; elemental analysis calcd (%) for
ACHTUNGTRENNUNG
Ar-H);
13C{1H} NMR
(125.8 MHz,
CDCl3,
295 K):
d=24.3
C64H38BF25IrN2P (1543.96): C 49.79, H 2.48, N 1.81; found: C 49.68, H
2.62, N 1.84.
(CH(N)CH2CH2), 25.3 (CH(N)CH2CH2), 25.8 (COD-CH2), 28.2 (COD-
CH2), 29.7 (CH(N)CH2), 34.0 (CH(N)CH2), 34.6 (COD-CH2), 37.2 (d,
J=4 Hz; COD-CH2), 41.4 (Bn-CH2), 61.9 (COD-CH), 64.5 (COD-CH),
69.9 (Im-3a’-CH), 72.0 (Im-7a’-CH), 85.5 (d, J=16 Hz; COD-CH), 94.8
(d, J=9 Hz; COD-CH), 107.7 (t, J=16 Hz; Bn-p-C), 122.1 (d, J=52 Hz;
PPh2-i-C), 123.8 (br; C6F5-i-C and Bn-i-C), 125.0 (d, J=50 Hz; Ar-2’’-C),
128.9 (d, J=53 Hz; PPh2-i-C), 129.0 (d, J=10 Hz; Ar-CH), 129.6 (br; Ar-
CH), 130.4 (br; Ar-CH), 130.9 (Ar-CH), 132.0 (Ar-CH), 132.3 (d, J=
8 Hz; Ar-CH), 132.7 (d, J=13 Hz; Ar-1’’-C), 132.9 (Ar-CH), 133.7 (d, J=
10 Hz; PPh2-o-CH), 136.4 (dmc, J=249 Hz; C6F5-m-C), 138.4 (dmc, J=
250 Hz; C6F5-p-C), 143.7 (dmc, J=248 Hz; Bn-m-C), 148.3 (dmc, J=
241 Hz; C6F5-o-C and Bn-o-C), 165.2 ppm (d, J=5 Hz; Im-2’-C);
19F{1H} NMR (376.5 MHz, CDCl3, 300 K): d=ꢀ167.4 (brs, 2F; C6F5-m-
F), ꢀ167.2 (mc, 2F; C6F5-m-F), ꢀ166.8 (brs, 2F; C6F5-m-F), ꢀ163.3 (t,
J=21 Hz, 1F; C6F5-p-F), ꢀ163.2 (t, J=21 Hz, 2F; C6F5-p-F), ꢀ146.7 (mc,
2F; Bn-m-F), ꢀ133.2 (brs, 2F; ArF-o-F), ꢀ132.8 (brs, 2F; ArF-o-F),
ꢀ132.6 (brs, 2F; ArF-o-F), ꢀ131.8 (brs, 1F; ArF-o-F), ꢀ131.3 ppm (brs,
1F; ArF-o-F); 31P{1H} NMR (202.5 MHz, CDCl3, 295 K): d=24.1 ppm;
11B{1H} NMR (160.5 MHz, CDCl3, 295 K): d=ꢀ16.9 ppm; IR (KBr): n˜ =
3061, 2941, 2883, 1643, 1513, 1462, 1364, 1260, 1090, 978, 894, 770, 695,
664, 539, 464 cmꢀ1; elemental analysis calcd (%) for C58H38BF19IrN2P
(1357.91): C 51.30, H 2.82, N 2.06; found: C 51.41, H 2.93, N 2.21.
General procedure for the preparation of ligands by heteroatom alkyla-
tion
Compound 45a: PHIM 12a (577 mg, 1.50 mmol) was added to a suspen-
sion of KH (66.2 mg, 1.65 mmol) in absolute THF (15 mL) at 08C. After
the mixture had been stirred at room temperature until no further gas
evolution was detected (about 2 h), pentafluorobenzyl bromide (272 mL,
1.80 mmol) was added dropwise to the now yellow solution at 08C. The
resulting yellow suspension was stirred for 2.5 h at 08C and then for 18 h
at room temperature. After the addition of aqueous Na2S2O3 (5%,
20 mL), the mixture was extracted with CH2Cl2 (4ꢄ20 mL). The com-
bined organic phases were dried over MgSO4, filtered, and concentrated
under reduced pressure. Purification of the remaining yellow oil by
column chromatography under argon (silica gel, 4ꢄ19 cm; hexanes/
EtOAc (2:1) + 5 vol% NEt3) yielded ligand 45a as a colorless, foamy
solid (369 mg, 44%). RF =0.31 (hexanes/EtOAc (2:1) + 5 vol% NEt3);
[a]2D0 =ꢀ13.2 (c=0.555 in CHCl3); 1H NMR (400.1 MHz, CDCl3, 300 K):
d=1.13–1.39 (m, 4H; CH(N)CHHCHH), 1.65–1.86 (brm, 3H;
CH(N)CHHCHH), 2.18 (brd, J=10.0 Hz, 1H; CH(N)CHH), 2.68 (brs,
1H; Im-7a-H), 2.88 (brs, 1H; Im-3a-H), 3.93 (brd, J=11.7 Hz, 1H; Bn-
CHH), 4.33 (d, J=14.7 Hz, 1H; Bn-CHH), 7.06 (dd, J=7.1, 4.0 Hz, 1H;
Ar-3’-H), 7.27–7.42 ppm (m, 13H; Ar-H); 13C{1H} NMR (100.6 MHz,
CDCl3, 300 K): d=24.8 (CH(N)CH2CH2), 25.5 (CH(N)CH2CH2), 29.7
(CH(N)CH2), 31.0 (CH(N)CH2), 39.5 (br; Bn-CH2), 71.9 (br; Im-CH),
112.4 (br; Bn-i-C), 128.6–129.0 (several Ar-CH), 129.5 (br; Ar-CH), 132.1
(br; Ar-C), 133.5 (Ar-3’-CH), 134.3 (d, J=20 Hz; PPh2-o-CH), 136.0–
138.7 (several Ar-C), 145.5 (dmc, J=251 Hz; Bn-o-C), 166.6 ppm (br; Im-
2-C); 19F{1H} NMR (376.5 MHz, CDCl3, 300 K): d=ꢀ162.4 (brs, 2F; Bn-
m-F), ꢀ155.2 (brs, 1F; Bn-p-F), ꢀ141.7 ppm (brs, 2F; Bn-o-F);
31P{1H} NMR (162.0 MHz, CDCl3, 300 K): d=ꢀ10.6 ppm; IR (KBr): n˜ =
3056, 2935, 2860, 1962, 1891, 1825, 1654, 1607, 1578, 1505, 1436, 1330,
1303, 1250, 1122, 1015, 963, 938, 745, 696, 668, 543, 507, 418 cmꢀ1; MS
(FAB, NBA): m/z (%)=565 (84) [M+H]+, 487 (12) [MꢀPh]+, 383 (100)
General procedure for the synthesis of cationic precatalysts
Compound 43a: A solution of ligand 45a (113 mg, 0.200 mmol) in abso-
lute CH2Cl2 (2 mL) was added dropwise at room temperature to a solu-
tion of [Ir
After the resulting red solution had been stirred in a closed vessel for 2 h
at 508C, the mixture was cooled to room temperature and Li[B(C6F5)4]
ACHTUNGTRENNUNG(COD)Cl]2 (73.9 mg, 0.110 mmol) in absolute CH2Cl2 (3 mL).
AHCTUNGTRENNUNG
(192 mg, 0.280 mmol) was added. The slightly turbid solution was stirred
for 5 min and then H2O (5 mL) was added. After the mixture had been
vigorously stirred for 30 min at room temperature, the phases were sepa-
rated and the aqueous layer was extracted with CH2Cl2 (3ꢄ10 mL). The
combined organic phases were dried over MgSO4, filtered, and concen-
4142
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4131 – 4144

Zwitterionic Iridium Complexes with P,N-Ligands
FULL PAPER
[MꢀC6F5CH2]+, 278 (45), 181 (56) [C6F5CH2+], 81 (30); elemental analy-
sis calcd (%) for C32H26F5N2P (564.53): C 68.08, H 4.64, N 4.96; found: C
68.35, H 5.00, N 4.52.
Acknowledgements
A.F. thanks the Fonds der Chemischen Industrie (Frankfurt) and the
German Federal Ministry for Science and Technology (BMBF) for a
Kekulꢅ Fellowship. Financial support by the Swiss National Science
Foundation is gratefully acknowledged. We also thank Markus Neubur-
ger, Dr. Sylvia Schaffner, and Dr. Stefan Kaiser for the crystal structure
analyses. Finally, we are grateful to Denise Rageot and Marcel Mꢃri for
preparing batches of PHOX derivative 31 and diamine 9c.
Compound 22a: In analogy to the synthesis of 45a, PHIM 12a (384 mg,
1.00 mmol) was reacted with KH (44.1 mg, 1.10 mmol) and borate 19a
(897 mg, 0.900 mmol) in absolute THF (10 mL) for 18 h at room temper-
ature. Purification of the yellowish crude product by column chromatog-
raphy under argon (silica gel, 4ꢄ19 cm; CH2Cl2) yielded the protonated
ligand 22a as a colorless, foamy solid (918 mg, 96%). According to
31P NMR spectroscopy, 22a is in equilibrium between two conformers in
a ratio of 2:1 at 272 K when it is dissolved in [D6]acetone. RF =0.54
(CH2Cl2); [a]2D0 =+32.1 (c=1.05 in CHCl3); 1H NMR (500.1 MHz,
[D6]acetone, 295 K, major conformer): d=1.36–1.62 (m, 3H;
CH(N)CHHCHH), 1.71 (mc, 1H; CH(N)CHH), 1.85–1.99 (m, 2H;
CH(N)CH2CHH), 2.20 (brd, J=10.9 Hz, 1H; CH(N)CHH), 2.38 (brd,
J=9.4 Hz, 1H; CH(N)CHH), 3.94 (brs, 2H; Im-3a’-H and Im-7a’-H),
4.73 (d, J=15.4 Hz, 1H; Bn-CHH), 4.94 (d, J=15.4 Hz, 1H; Bn-CHH),
7.31 (brdd, J=7.2, 3.8 Hz, 1H; Ar-6’’-H), 7.36–7.57 (m, 12H; PPh2-H,
Ar-3’’-H and Ar-4’’-H), 7.68 (t, J=7.6 Hz, 1H; Ar-5’’-H), 10.02 ppm (brs,
1H; NH+); 1H NMR (500.1 MHz, [D6]acetone, 295 K, characteristic sig-
nals of the minor conformer): d=3.37 (brt, J=12.9 Hz, 1H; Im-7a’-H),
3.67 (brt, J=12.9 Hz, 1H; Im-3a’-H), 4.45 (d, J=15.7 Hz, 1H; Bn-CHH),
4.60 ppm (d, J=15.2 Hz, 1H; Bn-CHH); 13C{1H} NMR (500.1 MHz,
[D6]acetone, 295 K, major conformer): d=24.4 (CH(N)CH2CH2), 24.5
(CH(N)CH2CH2), 28.3 (CH(N)CH2), 29.3 (CH(N)CH2), 39.3 (Bn-CH2),
65.2 (Im-7a’-CH), 71.9 (Im-3a’-CH), 129.6–131.1 (several Ar-CH and Ar-
C), 133.7 (Ar-CH), 134.3–134.7 (several Ar-CH), 135.4 (Ar-CH), 135.5
(Ar-CH), 137.1 (dmc, J=250 Hz; C6F5-m-C), 139.0 (dmc, J=244 Hz;
C6F5-p-C), 139.1–139.2 (several Ar-C), 149.0 (dmc, J=239 Hz; C6F5-o-C
and Bn-o-C), 171.0 ppm (Im-2’-C) (despite prolonged data acquisition,
the missing signals were not detected); 19F{1H} NMR (376.5 MHz,
[D6]acetone, 300 K): d=ꢀ167.9 to ꢀ167.4 (m, 6F; C6F5-m-F), ꢀ163.7 to
ꢀ163.4 (m, 3F; C6F5-p-F), ꢀ146.8 to ꢀ146.2 (m, 2F; Bn-m-F), ꢀ132.6 to
ꢀ131.5 ppm (m, 8F; C6F5-o-F and Bn-o-F); 31P{1H} NMR (202.5 MHz,
[1] Selected examples: a) Y. Chen, G. Wu, G. C. Bazan, Angew. Chem.
2005, 117, 1132–1136; Angew. Chem. Int. Ed. 2005, 44, 1108–1112;
b) J. W. Strauch, G. Erker, G. Kehr, R. Frçhlich, Angew. Chem.
2002, 114, 2662–2664; Angew. Chem. Int. Ed. 2002, 41, 2543–2546.
[2] Reviews: a) G. Erker, G. Kehr, R. Frçhlich, J. Organomet. Chem.
2005, 690, 6254–6262; b) G. Erker, Chem. Commun. 2003, 1469–
1476; c) G. Erker, Acc. Chem. Res. 2001, 34, 309–317; d) W. E.
Piers, Y. Sun, L. W. M. Lee, Top. Catal. 1999, 7, 133–143; e) M.
Bochmann, Top. Catal. 1999, 7, 9–22; f) W. E. Piers, Chem. Eur. J.
1998, 4, 13–18.
[3] Reviews: a) D. Sinou, Adv. Synth. Catal. 2002, 344, 221–237; b) F.
Joꢆ, ꢇ. Kathꢆ, J. Mol. Catal. A 1997, 116, 3–26; c) W. A. Herrmann,
C. W. Kohlpaintner, Angew. Chem. 1993, 105, 1588–1609; Angew.
Chem. Int. Ed. Engl. 1993, 32, 1524–1544.
[4] Review about betaines of platinum group metals in general: M.
Stradiotto, K. D. Hesp, R. J. Lundgren, Angew. Chem. 2010, 122,
504–523; Angew. Chem. Int. Ed. 2010, 49, 494–512.
[5] R. R. Schrock, J. A. Osborn, Inorg. Chem. 1970, 9, 2339–2343.
[6] a) Z. Zhou, B. R. James, H. Alper, Organometallics 1995, 14, 4209–
4212; b) I. Amer, H. Alper, J. Am. Chem. Soc. 1990, 112, 3674–
3676.
[7] a) C. Dai, E. G. Robins, A. J. Scott, W. Clegg, D. S. Yufit, J. A. K.
Howard, T. B. Marder, Chem. Commun. 1998, 1983–1984; b) S. A.
Westcott, H. P. Blom, T. B. Marder, R. T. Baker, J. Am. Chem. Soc.
1992, 114, 8863–8869.
[D6]acetone, 272 K): d=ꢀ11.6 and ꢀ11.4 ppm (in
a ratio of 2:1);
31P{1H} NMR (202.5 MHz, [D6]acetone, 315 K): d=ꢀ10.7 ppm;
11B{1H} NMR (160.5 MHz, [D6]acetone, 295 K): d=ꢀ15.9 ppm; IR
(KBr): n˜ =3680, 3418, 3193, 3063, 2952, 2874, 1644, 1515, 1462, 1372,
1311, 1266, 1162, 1090, 977, 759, 694, 663, 606, 574, 544, 505 cmꢀ1; MS
(FAB, NBA): m/z (%)=1059 (6) [M+H]+, 891 (51) [MꢀC6F5]+, 743 (30)
[8] T. A. Betley, J. C. Peters, Angew. Chem. 2003, 115, 2487–2491;
Angew. Chem. Int. Ed. 2003, 42, 2385–2389.
[9] a) R. J. Lundgren, M. A. Rankin, R. McDonald, G. Schatte, M. Stra-
diotto, Angew. Chem. 2007, 119, 4816–4819; Angew. Chem. Int. Ed.
2007, 46, 4732–4735; b) J. Cipot, R. McDonald, M. J. Ferguson, G.
Schatte, M. Stradiotto, Organometallics 2007, 26, 594–608; c) J.
Cipot, R. McDonald, M. Stradiotto, Chem. Commun. 2005, 4932–
4934; examples of other transformations catalyzed by similar zwit-
terions: d) R. J. Lundgren, M. A. Rankin, R. McDonald, M. Stra-
diotto, Organometallics 2008, 27, 254–258; e) J. Cipot, C. M. Vogels,
R. McDonald, S. A. Westcott, M. Stradiotto, Organometallics 2006,
25, 5965–5968; f) M. Stradiotto, J. Cipot, R. McDonald, J. Am.
Chem. Soc. 2003, 125, 5618–5619.
[10] a) V. Kçhler, C. Mazet, A. Toussaint, K. Kulicke, D. Hꢈussinger, M.
Neuburger, S. Schaffner, S. Kaiser, A. Pfaltz, Chem. Eur. J. 2008, 14,
8530–8539; b) A. Toussaint, A. Pfaltz, Eur. J. Org. Chem. 2008,
4591–4597; c) S. J. Roseblade, A. Pfaltz, Synthesis 2007, 3751–3753;
d) B. Ramalingam, M. Neuburger, A. Pfaltz, Synthesis 2007, 572–
582; e) C. Mazet, S. Roseblade, V. Kçhler, A. Pfaltz, Org. Lett. 2006,
8, 1879–1882; f) C. Mazet, V. Kçhler, S. Roseblade, A. Toussaint, A.
Pfaltz, Chimia 2006, 60, 195–198; g) C. Mazet, V. Kçhler, A. Pfaltz,
Angew. Chem. 2005, 117, 4966–4969; Angew. Chem. Int. Ed. 2005,
44, 4888–4891.
[MꢀC6F5ꢀC6F4]+, 383 (53) [Mꢀ (C6F4CH2)]+, 184 (57), 81 (100);
ACHUTNGTREN(NUG C6F5)3BACHUTNGTRENNGUN
elemental analysis calcd (%) for C50H27BF19N2P (1058.52): C 56.73, H
2.57, N 2.65; found: C 56.50, H 2.77, N 2.57.
General procedure for enantioselective hydrogenations: The precatalyst
(usually 1.0 mmol) and substrate (usually 100 mmol) were weighed in a
2 mL screw-cap glass vial equipped with a magnetic stirrer bar and the
desired solvent was added (usually 0.5 mL of absolute CH2Cl2). Alterna-
tively, stock solutions of the iridium complex and alkene were sometimes
used. Four vessels were placed in a 60 mL autoclave (Premex), which
was closed under an inert atmosphere. After pressurizing the autoclave
with hydrogen (usually 50 bar), the transformation was initiated by
switching on the stirrer (700 minꢀ1). After the required reaction time, the
hydrogen was released and hexanes (2 mL) was added. The resulting sus-
pension was filtered through a pad of silica gel, which was washed with
Et2O/hexanes (1:1). The eluate was concentrated under reduced pressure,
the residue was redissolved in heptane (3 mL), and the conversion and
enantioselectivity were directly determined by GC and HPLC
analyses.[11a,41,42]
Crystal structure analyses: CCDC-798427 (7c), 798428 (20c), 798429
(53a), 798430 (53b), 798431 (54) and 798432 (56) contain the supplemen-
tary 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.
[11] a) A. Lightfoot, P. Schnider, A. Pfaltz, Angew. Chem. 1998, 110,
3047–3050; Angew. Chem. Int. Ed. 1998, 37, 2897–2899; reviews:
b) S. J. Roseblade, A. Pfaltz, Acc. Chem. Res. 2007, 40, 1402–1411;
c) X. Cui, K. Burgess, Chem. Rev. 2005, 105, 3272–3296; d) A.
Pfaltz, J. Blankenstein, R. Hilgraf, E. Hçrmann, S. McIntyre, F.
Menges, M. Schçnleber, S. P. Smidt, B. Wꢃstenberg, N. Zimmer-
mann, Adv. Synth. Catal. 2003, 345, 33–43.
[12] a) Y. Zhu, A. Loudet, K. Burgess, Org. Lett. 2010, 12, 4392–4395;
b) K. Tiefenbacher, L. Trçndlin, J. Mulzer, A. Pfaltz, Tetrahedron
2010, 66, 6508–6513; c) T. Yoshinari, K. Ohmori, M. G. Schrems, A.
Chem. Eur. J. 2011, 17, 4131 – 4144
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4143

A. Pfaltz and A. Franzke
Pfaltz, K. Suzuki, Angew. Chem. 2010, 122, 893–897; Angew. Chem.
Int. Ed. 2010, 49, 881–885; d) M. G. Schrems, A. Pfaltz, Chem.
Commun. 2009, 6210–6212; e) G. G. Bianco, H. M. C. Ferraz, A. M.
Costa, L. V. Costa-Lotufo, C. Pessoa, M. O. de Moraes, M. G.
Schrems, A. Pfaltz, L. F. Silva, Jr., J. Org. Chem. 2009, 74, 2561–
2566; f) J. Zhao, K. Burgess, Org. Lett. 2009, 11, 2053–2056; g) Y.
Zhu, K. Burgess, J. Am. Chem. Soc. 2008, 130, 8894–8895; h) A.
Wang, B. Wꢃstenberg, A. Pfaltz, Angew. Chem. 2008, 120, 2330–
2332; Angew. Chem. Int. Ed. 2008, 47, 2298–2300; i) J. Zhou, K.
Burgess, Angew. Chem. 2007, 119, 1147–1149; Angew. Chem. Int.
Ed. 2007, 46, 1129–1131; j) S. Bell, B. Wꢃstenberg, S. Kaiser, F.
Menges, T. Netscher, A. Pfaltz, Science 2006, 311, 642–644; k) M.
Harmata, X. Hong, Org. Lett. 2005, 7, 3581–3583.
5195; b) C. A. Busacca, D. Grossbach, R. C. So, E. M. OꢀBrien,
E. M. Spinelli, Org. Lett. 2003, 5, 595–598.
[21] S. J. Coote, G. J. Dawson, C. G. Frost, J. M. J. Williams, Synlett 1993,
509–510.
[22] N. A. Boland, M. Casey, S. J. Hynes, J. W. Matthews, M. P. Smith, J.
Org. Chem. 2002, 67, 3919–3922.
[23] For additional asymmetric hydrogenation reactions, see the Support-
ing Information.
[24] F. Menges, Dissertation, University of Basel (Switzerland), 2004.
[25] Selected spectroscopic data of the reaction mixture: 31P{1H} NMR
(202.5 MHz, CD2Cl2, 295 K): d=ꢀ2.2 (brs) and 19.2 ppm (d, J=
5 Hz) (in a ratio of 5:4); 1H NMR (500.1 MHz, CD2Cl2, 295 K, hy-
dride region): d=ꢀ27.05 (brs, 1H; major), ꢀ16.38 (brs, 1H; major),
ꢀ15.98 (d, J=8.6 Hz, 1H; minor), ꢀ12.91 ppm (d, J=16.2 Hz, 1H;
minor).
[26] A. Franzke, A. Pfaltz, Synthesis 2008, 245–252.
[27] a) B. Fernꢉndez, I. Perillo, S. Lamdan, J. Chem. Soc. Perkin Trans. 2
1973, 1371–1374; b) J. Elguero, ꢊ. Gonzalez, J.-L. Imbach, R. Jacqu-
ier, Bull. Soc. Chim. Fr. 1969, 11, 4075–4077.
[28] M. Peer, J. C. de Jong, M. Kiefer, T. Langer, H. Rieck, H. Schell, P.
Sennhenn, J. Sprinz, H. Steinhagen, B. Wiese, G. Helmchen, Tetrahe-
dron 1996, 52, 7547–7583.
[29] H.-J. Frohn, H. Franke, P. Fritzen, V. V. Bardin, J. Organomet.
Chem. 2000, 598, 127–135.
[13] a) I. Krossing, Chem. Eur. J. 2001, 7, 490–502; b) S. M. Ivanova,
B. G. Nolan, Y. Kobayashi, S. M. Miller, O. P. Anderson, S. H.
Strauss, Chem. Eur. J. 2001, 7, 503–510; c) review about weakly co-
ordinating anions: I. Krossing, I. Raabe, Angew. Chem. 2004, 116,
2116–2142; Angew. Chem. Int. Ed. 2004, 43, 2066–2090.
[14] A. G. Massey, A. J. Park, J. Organomet. Chem. 1964, 2, 245–250.
[15] H. Nishida, N. Takada, M. Yoshimura, T. Sonoda, H. Kobayashi,
Bull. Chem. Soc. Jpn. 1984, 57, 2600–2604.
[16] a) S. P. Smidt, N. Zimmermann, M. Studer, A. Pfaltz, Chem. Eur. J.
2004, 10, 4685–4693; b) D. Drago, P. S. Pregosin, A. Pfaltz, Chem.
Commun. 2002, 286–287; c) D. G. Blackmond, A. Lightfoot, A.
Pfaltz, T. Rosner, P. Schnider, N. Zimmermann, Chirality 2000, 12,
442–449.
[30] For the entire data of the kinetic studies, see the Supporting Infor-
mation.
[31] a) S. P. Smidt, A. Pfaltz, E. Martꢋnez-Viviente, P. S. Pregosin, A. Al-
binati, Organometallics 2003, 22, 1000–1009; b) H.-H. Wang, A. L.
Casalnuovo, B. J. Wang, A. M. Mueting, L. H. Pignolet, Inorg.
Chem. 1988, 27, 325–331; c) H. H. Wang, L. H. Pignolet, Inorg.
Chem. 1980, 19, 1470–1480; d) D. F. Chodosh, R. H. Crabtree, H.
Felkin, G. E. Morris, J. Organomet. Chem. 1978, 161, C67–C70.
[32] For other crystallographically characterized iridium PHIM com-
plexes, see refs. [19a,b].
[33] Reviews about the synthesis and application of organotrifluorobo-
rates: a) G. A. Molander, N. Ellis, Acc. Chem. Res. 2007, 40, 275–
286; b) S. Darses, J.-P. Genet, Eur. J. Org. Chem. 2003, 4313–4327.
[34] For selected bond lengths and angles as well as a graphical represen-
tation of the structure, see the Supporting Information.
[35] D. D. Perrin, W. L. F. Armarego (Eds.), Purification of Laboratory
Chemicals, 3rd ed., Pergamon Press, Oxford, 1988.
[17] a) X. Cui, Y. Fan, M. B. Hall, K. Burgess, Chem. Eur. J. 2005, 11,
6859–6868; b) Y. Fan, X. Cui, K. Burgess, M. B. Hall, J. Am. Chem.
Soc. 2004, 126, 16688–16689; c) C. Mazet, S. P. Smidt, M. Meuwly,
A. Pfaltz, J. Am. Chem. Soc. 2004, 126, 14176–14181; d) R. Dietiker,
P. Chen, Angew. Chem. 2004, 116, 5629–5632; Angew. Chem. Int.
Ed. 2004, 43, 5513–5516; e) P. Brandt, C. Hedberg, P. G. Andersson,
Chem. Eur. J. 2003, 9, 339–347.
[18] Further representative examples: a) S. Li, S.-F. Zhu, J.-H. Xie, S.
Song, C.-M. Zhang, Q.-L. Zhou, J. Am. Chem. Soc. 2010, 132, 1172–
1179; b) Z. Han, Z. Wang, X. Zhang, K. Ding, Angew. Chem. 2009,
121, 5449–5453; Angew. Chem. Int. Ed. 2009, 48, 5345–5349;
c) M. G. Schrems, E. Neumann, A. Pfaltz, Angew. Chem. 2007, 119,
8422–8424; Angew. Chem. Int. Ed. 2007, 46, 8274–8276; d) D. Liu,
Q. Dai, X. Zhang, Tetrahedron 2005, 61, 6460–6471; e) D. Liu, W.
Tang, X. Zhang, Org. Lett. 2004, 6, 513–516; f) P. G. Cozzi, F.
Menges, S. Kaiser, Synlett 2003, 833–836; g) W. Tang, W. Wang, X.
Zhang, Angew. Chem. 2003, 115, 973–976; Angew. Chem. Int. Ed.
2003, 42, 943–946; h) D.-R. Hou, J. Reibenspiess, T. J. Colacot, K.
Burgess, Chem. Eur. J. 2001, 7, 5391–5400.
[19] a) E. Guiu, C. Claver, J. Benet-Buchholz, S. Castillꢆn, Tetrahedron:
Asymmetry 2004, 15, 3365–3373; b) F. Menges, M. Neuburger, A.
Pfaltz, Org. Lett. 2002, 4, 4713–4716; analogous rhodium catalysts:
c) J. C. Lorenz, C. A. Busacca, X. W. Feng, N. Grinberg, N. Haddad,
J. Johnson, S. Kapadia, H. Lee, A. Saha, M. Sarvestani, E. M. Spine-
lli, R. Varsolona, X. Wei, X. Zeng, C. H. Senanayake, J. Org. Chem.
2010, 75, 1155–1161; d) C. A. Busacca, J. C. Lorenz, N. Grinberg, N.
Haddad, H. Lee, Z. Li, M. Liang, D. Reeves, A. Saha, R. Varsolona,
C. H. Senanayake, Org. Lett. 2008, 10, 341–344.
[20] a) C. A. Busacca, D. Grossbach, S. J. Campbell, Y. Dong, M. C.
Eriksson, R. E. Harris, P.-J. Jones, J.-Y. Kim, J. C. Lorenz, K. B.
McKellop, E. M. OꢀBrien, F. Qiu, R. D. Simpson, L. Smith, R. C. So,
E. M. Spinelli, J. Vitous, C. Zavattaro, J. Org. Chem. 2004, 69, 5187–
[36] D. L. Reger, T. D. Wright, C. A. Little, J. J. S. Lamba, M. D. Smith,
Inorg. Chem. 2001, 40, 3810–3814.
[37] D. Alberti, K.-R. Pçrschke, Organometallics 2004, 23, 1459–1460.
[38] S. Roland, P. Mangeney, A. Alexakis, Synthesis 1999, 228–230.
[39] H. E. Gottlieb, V. Kotylar, A. Nudelman, J. Org. Chem. 1997, 62,
7512–7515.
[40] For CD2Cl2 the solvent shifts d=5.32 (1 H) and 54.0 ppm (13C), re-
spectively, were used to calibrate the spectra.
[41] a) J. Blankenstein, A. Pfaltz, Angew. Chem. 2001, 113, 4577–4579;
Angew. Chem. Int. Ed. 2001, 40, 4445–4447; b) P. Schnider, G. Koch,
R. Prꢅtꢌt, G. Wang, F. M. Bohnen, C. Krꢃger, A. Pfaltz, Chem. Eur.
J. 1997, 3, 887–892.
[42] The conversions of 17 and 60 were determined by GC (Restek Rtx
1701, 60 kPa He, 1008C/2 min/78C minꢀ1/2508C/10 min): tR =14.2
(product), 16.1 min (substrate) for 17 and tR =13.5 (product),
14.2 min (substrate) for 60.
Received: November 17, 2010
Published online: March 4, 2011
4144
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4131 – 4144