Nucleophilic homogeneous hydrogenation by iridium complexes

Article abstract of DOI:10.1055/s-0028-1087513

Catalytic homogeneous hydrogenation of 7-methoxy-3-phenylchromone and other substrates was achieved in the presence of cationic iridium complexes and base as co-catalyst. Contrary to common alkene hydrogenation, which is inactivated by base, the hydrogena

Full text of DOI:10.1055/s-0028-1087513

LETTER
271
Nucleophilic Homogeneous Hydrogenation by Iridium Complexes
N
ucleophilic
H
om
o
ogeneous
H
y
l
droge
n
o
ation by Irid
d
ium Comple
x
es myr Semeniuchenko,*a,b Volodymyr Khilya,b Ulrich Grotha
a
Fachbereich Chemie und Konstanz Research School Chemical Biology, Universität Konstanz, Fach M-720, Universitätsstr. 10,
78457 Konstanz, Germany
E-mail: Ulrich.groth@uni-konstanz.de
National Taras Shevchenko University, 62 Volodymyrska st., Kyiv-33, 01033, Ukraine
E-mail: chem_vova@mail.univ.kiev.ua
b
Received 8 August 2008
Abstract: Catalytic homogeneous hydrogenation of 7-methoxy-3-
phenylchromone and other substrates was achieved in the presence
of cationic iridium complexes and base as co-catalyst. Contrary to
common alkene hydrogenation, which is inactivated by base, the
hydrogenation of the above set of electron-deficient alkenes turned
out to be base-activated.
Key words: hydrogenation, homogenous catalysis, iridium, al-
kenes, complexes
Hydrogenation is one of the most important classes of or-
ganic reactions. The enormous potential of the homoge-
neous metallocomplex-catalyzed hydrogenation is
commonly acknowledged and numerous applications to-
wards preparation of a wide range of organic compounds
are presented in the recently published three-volume set
The Handbook of Homogeneous Hydrogenation.1 It has to
be emphasized that this versatile method is especially
valuable for the enantioselective reduction2 of alkenes,
ketones, imines etc, and is used both in laboratory and in
industry.3
Scheme 1
chromanone 2 was detected after hydrogenation in the
presence of triethylamine (Scheme 1).
In order to explain this unexpected result, we need to take
the structure of both the substrate and catalysts into ac-
count. Iridium complexes of type [Ir(cod)(L)2]X or
[Ir(cod)(L^L)]X (where cod = 1,5-cyclooctadiene, L^L =
chelating ligand, and X = a noncoordinating anion) are
precatalysts. They are activated by hydrogenation of coor-
dinated 1,5-cyclooctadiene to form extremely electro-
philic species.9 These can coordinate to double bond, add
and transfer hydrogen.9 It is also well known that iso-
flavones react with electrophiles to give 8-substituted
chromones as the products of normal electrophilic substi-
tution in the aromatic ring.10 The C=C bond in the
chromone ring of isoflavones is known to be very elec-
tron-deficient. Therefore it is inert towards electrophiles
and rather apt to react with nucleophiles,10 presumably in-
cluding the base-activated catalyst 3, hence the ‘nucleo-
philic homogeneous hydrogenation’.
Base treatment (i.e., Et3N) enhances the catalytical activ-
ity of rhodium complexes forming neutral coordinatively
unsaturated hydrido complexes.4 Unlike rhodium, iridium
is known to be deactivated by organic bases.5,6
Hereby we want to report a completely new type of homo-
geneous hydrogenation, catalyzed by iridium complexes
and activated by base.
Several complexes of rhodium ([Rh(PPh3)3Cl], [Rh(BI-
NAP)(COD)Cl], [Rh(BINAP)(COD)]OTf) and those of
ruthenium ([Ru(PPh3)3Cl2], [Ru(p-cymene)(BINAP)-
Cl]Cl, [Ru(BINAP)(OAc)2]) were found to be inactive to-
wards homogeneous hydrogenation of 7-methoxy-
isoflavone (1) under pressure of hydrogen (100 bar) in
various solvents (toluene, THF, dichloromethane, DMF,
2-propanol, methanol) with or without activation by base
or acid. Transfer hydrogenation, catalyzed by [Ru(p-
cymene)(Tsen)Cl]Cl7 (Tsen = N-Tosylethenylendiamine)
was also ineffective for this substrate.
In order to optimize the reaction conditions, base- and sol-
vent-screening experiments were performed under stan-
dard conditions11 for hydrogenation as shown in Table 1.
Compound 1 was used as a model substrate and the reac-
tion was run in the presence of 1 mol% of 3.
Complex 3 was not able to perform a homogeneous
hydrogenation of 1 under standard conditions in toluene
or dichloromethane.8 However, the corresponding
Chromone 1 could be heterogenically hydrogenated in di-
oxane using palladium-on-carbon as a catalyst. This reac-
tion is completely suppressed by thiophene, which
normally poisons heterogeneous catalysts. Hydrogenation
SYNLETT 2009, No. 2, pp 0271–0275
Advanced online publication: 15.01.2009
DOI: 10.1055/s-0028-1087513; Art ID: G26608ST
© Georg Thieme Verlag Stuttgart · New York
x
x.
x
x.
2
0
0
8

272
V. Semeniuchenko et al.
LETTER
Table 1 Optimization of the Reaction11
Base
Solvent
toluene
toluene
toluene
toluene
toluene
MeOH
CH2Cl2
DCE
Active Complex
Conversion (%)
Et3N
3
3
3
3
3
3
3
3
3
3
3
3
20a
37a
3a
DIPEA
TMEDA
LHMDS
2,6-di-tert-butylpyridine
DIPEA
0a
0a
37a
0a
DIPEA
DIPEA
0a
DIPEA
CHCl3
THF
0a
DIPEA
tracea
tracea
tracea
DIPEA
MTBE
DIPEA
DIPEA
O
DIPEA
DIPEA
toluene
toluene
22a
10a
Ph2P
N
BARF
Ir
(cod)
O
O
BARF
BARF
Ph2P
N
Ir
(cod)
DIPEA
DIPEA
toluene
10a
Ph2P
N
Ir
(cod)
10a,b
O
toluene
toluene
Ph2P
N
Ir
BARF
(cod)
N
N
DIPEA
8.5a,c
C
C
N
N
Ir
(cod)
BARF
a GC and NMR conversion.
b Preparation of this catalyst is given in ref.12
c Preparation of this catalyst is given in ref.13
catalyzed by 3 readily occurs in the presence of thiophene, the 2,6-di-tert-butylpyridine is known to be a ligand for
thus proving the process’ homogeneity.14 Judging by the palladium,15 gold16 and titanium,17 hence is also a coordi-
results, the coordinating bases LHMDS, TMEDA and 2,6- nating base. Chromone 1 could not be reduced in neat
di-tert-butylpyridine suppress nucleophilic hydrogena- DIPEA due to poor solubility. Two- or ten-fold excess of
tion, while noncoordinating DIPEA activates it more effi- DIPEA (relative to iridium complex) was not sufficient to
ciently than triethylamine. Despite the sterical hindrances, run this reaction. In microscale experiment 0.07 mL of
Synlett 2009, No. 2, 271–275 © Thieme Stuttgart · New York

LETTER
Nucleophilic Homogeneous Hydrogenation by Iridium Complexes
273
Table 2 Reaction of Substrates Bearing Deactivated C=C Bond with Complex 3
Reagent
Product
Catalyst
loading
GC–NMR conver- Synthetic
sion (isolated yield) procedure
7-methoxyisoflavone (1)
7-methoxyisoflavone (1)
7-methoxyisoflavanone (2)
7-methoxyisoflavanone (2)
1 mol%
3 mol%
37% (not isolated)
100% (99%)
experimental11
preparative18
7-methoxy-2-trifluoromethylisoflavone (4) 7-methoxy-2-trifluoromethylisoflavanone 1 mol%
6% (not isolated)
experimental11
preparative18
7-methoxy-2-trifluoromethylisoflavone (4) 7-methoxy-2-trifluoromethylisoflavanone 17 mol% 90% (not isolated)
and overreduced compounds
benzylidenemalononitrile (5)
2-benzylmalononitrile
1 mol%
2 mol%
1 mol%
37% (not isolated)
100% (60%)
experimental11
preparative18
benzylidenemalononitrile (5)
2-benzylmalononitrile
2-(1-phenylethylidene)malononitrile (6)
2-(1-phenylethylidene)malononitrile (6)
3-nitro-benzylidenemalononitrile (7)
3-nitro-benzylidenemalononitrile (7)
2-phenylcyclohexenone (8)
2-(1-phenylethyl)malononitrile
2-(1-phenylethyl)malononitrile
2-(3-nitrobenzyl)malononitrile
2-(3-nitrobenzyl)malononitrile
2-phenylcyclohexanone
37% (not isolated)
experimental11
2.4 mol% 90% (90%)
preparative18
1 mol%
90% (not isolated)
experimental11
1.5 mol% 100% (90%)
preparative18
1 mol%
10% (not isolated)
experimental11
2-phenylcyclohexenone (8)
2-phenylcyclohexanone
10 mol% 100% (90%)
preparative18
3-trifluoroacetylchromone (9)
3-trifluoroacetylchromone (9)
3-trifluoroacetylchromanone
mixture of overreduced compounds
1 mol%
5 mol%
20% (not isolated)
experimental11
100% (not isolated) preparative18
base (300–600-fold excess relative to Ir complex, depend- spectral properties of isolated hydrogenated compounds
ing on catalyst load) was directly added to the reaction were in agreement with those reported in literature. In or-
mixture.11
der to achieve full conversion, a higher catalyst load was
necessary. The best conversion of compound 6 was 90%
(2.4 mol% and higher load of 3). Compounds 4 and 9 were
overreduced in the presence of 17 mol% and 5 mol% of 3,
respectively. Unfortunately, in case of under- or over-
reduction, a product is always co-eluted with contami-
nant, hence purification by column chromatography is
impossible. None of these substrates could be hydrogenat-
ed homogeneously by complex 3 in the absence of base.
Compound 9 readily gives the geminate hydrate 10 under
exposure to moisture (Scheme 2).19 In contrast to 9, the
hydrate 10 was hydrogenated by complex 3 in electro-
philic conditions (i.e. without adding base), giving an in-
separable mixture of products. Since the preparation of
pure 9 is not found in literature, we have provided it in the
experimental section.20
We have found that the quality of DIPEA is significant for
the reaction. A very small amount of contaminating diiso-
propylamine could completely inhibit the reaction. Com-
mercially available DIPEA was purified by distillation
over ninhydrine, then twice over lithium hydride. Sodium
hydride was not effective to make diisopropylamine-free
DIPEA.
Chlorinated solvents are widely used6,8 for iridium-cata-
lyzed electrophilic hydrogenation but were found to in-
hibit nucleophilic hydrogenation. The reason behind it is
presumably the base-induced decomposition with the for-
mation of trialkylammonium chloride, which subsequent-
ly reacts either with catalyst 3 or with the catalytically
active intermediate. This assumption was confirmed by
performing hydrogenation experiments in the presence of
tetrabutylammonium chloride. The last compound was
chosen because of the fair solubility in toluene and was
found to inactivate the nucleophilic hydrogenation reac-
tion.
In summary, a new homogeneous hydrogenation reaction,
inherent for iridium complexes, is discovered, and a pre-
liminary study of this reaction is performed. Although the
reduction needs a lot of catalyst, it represents a significant
extension of the chemistry of this type of catalysts. In a
The other five iridium complexes were found to catalyze
nucleophilic hydrogenation of 1 with low conversion
(Table 1).
O
O
P2O5, CH2Cl2, 30 h
H2O
A number of substrates (Table 2), bearing deactivated
C=C bond, were tested in this reaction with complex 3.
CF3
CF3
OH
O
O
O
O
The substrates 1, 5, 7 and 8 were hydrogenated with full
conversion, according to the general preparative proce-
dure,18 and the corresponding products were isolated. The
H
9
10
Scheme 2
Synlett 2009, No. 2, 271–275 © Thieme Stuttgart · New York

274
V. Semeniuchenko et al.
LETTER
subjected to gas chromatography with mass-spectral
analyser (GCMS). GC conversion was calculated from
correspondent peak areas and was not corrected with an
external standard. If needed, the reaction mixture was
evaproated in vacuo, the residue dissolved in CDCl3 or
DMSO-d6 and subjected to 1H NMR analysis, which showed
the same conversion.
foreseen full paper we are going to describe the scope and
limitations of this reaction and make an approach to a sup-
posed mechanism.
Acknowledgment
S.V. thanks the Herbert-Quandt foundation and the German Acade-
mic Exchange Service (DAAD) for doctoral fellowships. S.V. again
thanks the partnership program between the University of Konstanz
and the NTSU Kiew for having the opportunity to perform his doc-
toral studies in Konstanz. The authors are thankful to Olga Tsubrik
(Tartu University) and to Milena Quentin for helping to prepare the
manuscript. We are grateful to Umicore, MCAT, Bayer AG, Merck
KGaA and Wacker AG for generous gifts of reagents.
(12) To a solution of 8-hydroxyquinoline (23.4 mg, 0.162 mmol)
in Et2O (1 mL) n-BuLi (0.1 mL, 0.16 mmol, 1.6 M hexane
solution) was added. After stirring for 10 min Ph2PCl (36.5
mg, 0.162 mmol) was added and stirred for 2 h. The Et2O
was evaporated, and [IR (cod)Cl]2 (54.2 mg, 0.081 mmol) in
CH2Cl2 (2 mL) added. The mixture obtained was refluxed
for 2 h, cooled, and sodium tetrakis[3,5-bis(trifluoromethyl)-
phenyl]borate (NaBARF; 143.2 mg, 0.162 mmol) was
added. After overnight stirring the iridium [8-(diphenyl-
phosphinooxy)quinoline](1,5-cyclooctadiene) tetrakis-
[3,5-bis(trifluoromethyl)phenyl]borate was separated by
silica gel column chromatography (Rf 0.9; CH2Cl2). This
complex decomposed after its formation, and could not be
isolated in pure form at r.t. This product was characterized
by 31P NMR, having only one peak at d = +107 ppm. During
1 d it could catalyze nucleophilic or electrophilic
References and Notes
(1) The Handbook of Homogeneous Hydrogenation; de Vries,
J. G.; Elsevier, C. J., Eds.; Wiley-VCH: Weinheim, 2007.
(2) (a) Comprehensive Asymmetric Catalysis I; Jacobsen, E. N.;
Pfaltz, A., Eds.; Springer: Berlin, 1999. (b) Jacobsen, E. N.;
Pfaltz, A. Comprehensive Asymmetric Catalysis II;
Jacobsen, E. N.; Pfaltz, A., Eds.; Springer: Berlin, 1999.
(c) Comprehensive Asymmetric Catalysis III; Jacobsen,
E. N.; Pfaltz, A., Eds.; Springer: Berlin, 1999.
hydrogenation. In the latter case we checked the
hydrogenation of stilbene in CH2Cl2.
(13) To a solution of [IR(cod)Cl]2 (30.1 g, 0.044 mmol) and
NaBARF (77.4 mg, 0.088 mmol) in CH2Cl2 (1.5 mL) freshly
distilled 1.5-cyclooctadiene (0.1 mL) was added. The
mixture was stirred for 30 min, then Na2SO4 was added to
adsorb the formed NaCl. The mixture was filtrated,
thoroughly washed by CH2Cl2, evaporated to 0.25 mL and
pentane (10 mL) was added. The formed crystalline iridium
bis(1,5-cyclooctadiene)tetrakis [3,5-bis(trifluoromethyl)-
phenyl]borate ([IR(cod)2]BARF) was collected by filtration,
washed from the filter by CH2Cl2 and dried under vacuum.
1H NMR (400 MHz, CDCl3/TMS): d = 2.26 (m, 8 H, CH2),
2.41 (m, 8 H CH2), 5.00 (s, 8 H, CHCOD), 7.56 (s, 4 H,
4-HBARF), 7.71 (s, 8 H, 2-HBARF and 6-HBARF).
(d) Comprehensive Asymmetric Catalysis, Supplement;
Jacobsen, E. N.; Pfaltz, A.; Yamamoto, H., Eds.; Springer:
Berlin, New York, 2004.
(3) Asymmetric Catalysis on Industrial Scale: Challenges,
Approaches and Solutions; Blaser, H. U.; Schmidt, E., Eds.;
Wiley-VCH: Weinheim, 2004, 454.
(4) (a) Oro, L. A.; Carmona, D. Rhodium, In The Handbook of
Homogeneous Hydrogenation, Vol. 1; de Vries, J. G.;
Elsevier, C. J., Eds.; Wiley-VCH: Weinheim, 2007, 3–30.
(b) Schrock, R. R.; Osborn, J. A. J. Am. Chem. Soc. 1976, 98,
2134. (c) Schrock, R. R.; Osborn, J. A. J. Am. Chem. Soc.
1976, 98, 2143. (d) Schrock, R. R.; Osborn, J. A. J. Am.
Chem. Soc. 1976, 98, 4450.
(5) (a) Crabtree, R. H. Iridium, In The Handbook of
Homogeneous Hydrogenation, Vol. 1; de Vries, J. G.;
Elsevier, C. J., Eds.; Wiley-VCH: Weinheim, 2007, 31–44.
(b) Crabtree, R. H.; Felkin, H.; Morris, G. E. J. Organomet.
Chem. 1977, 141, 205.
(6) Blackmond, D. G.; Lightfoot, A.; Pfaltz, A.; Rosner, T.;
Schnider, P.; Zimmermann, N. Chirality 2000, 12, 442.
(7) Xue, D.; Chen, Y.-C.; Cui, X.; Wang, Q.-W.; Zhu, J.; Deng,
J.-G. J. Org. Chem. 2005, 70, 3584.
Under nitrogen to a suspension of 1,3-dimethyl-1H-
imidazol-3-ium iodide (23.1 mg, 0.103 mmol) in THF (5
mL) n-BuLi (0.064 mL, 0.103 mmol, 1.6 M solution in
hexane) was added and stirred for 2 h. [IR(cod)2]BARF
(65.5 Mg, 0.0516 mmol) was added and stirred for 28 h ar r.t.
Reaction was quenched by MeOH and adsorbed on silica
gel. Column chromatography on silica gel (9 g; under air,
eluent: CH2Cl2; Rf 0.9) gave orange crystalline iridium
bis(imidazol-3-ene)(1,5-cyclooctadiene)tetrakis[3,5-
bis(trifluoromethyl)phenyl]borate (30 mg, 42%). 1H NMR
(400 MHz, CDCl3/TMS): d = 1.95 (m, 4 H, ch2cod), 2.21 (m,
4 H, ch2cod), 3.76 (s, 4 H, chcod), 3.79 (s, 12 H, Me), 6.73 (s,
4 H, CHimidazolene), 7.52 (s, 4 H, 4-HBARF), 7.71 (s, 8 H,
2-HBARF, 5-HBARF). 13C NMR (100 MHz, CDCl3/TMS): d =
31.14 (CH2), 37.49 (Me), 76.78 (chcod), 117.50 (br s,
4-CHBARF), 122.60 (CHimidazolene), 124.53 (q, J = 272.3 Hz,
CF3), 128.9 (br q, J = 30.0 Hz, 3-CCF3), 134.80 (br s,
2-CHBARF, 6-CHBARF), 161.74 (q 1:1:1:1, JCB = 50.5 Hz,
CBBARF), 177.70 (Ccarbene). 11B NMR (128 MHz, CDCl3/
BF3Et2O): d = –6.9. 11F NMR (376 MHz, CDCl3/CFCl3): d =
–62.8.
(8) Lightfoot, A.; Schnider, P.; Pfaltz, A. Angew. Chem. 1998,
37, 2897.
(9) Kallstrom, K.; Munslow, I.; Andersson, P. G. Chem. Eur. J.
2006, 12, 3194.
(10) (a) Frasinyuk, M. S.; Khilya, V. P. Chem. Heterocycl.
Compd. (Engl. Transl.) 1999, 35, 20. (b) Kazakov, A. L.;
Khilya, V. P.; Mezheritskii, V. V.; Litkei, Y. Natural and
Modified Isoflavonoids (in Russian); Izdat. Rostovsk
University: Rostov na Donu, 1985, 184.
(11) Typical Experimental Procedure: Catalyst (1–2 mg) was
weighed into a Teflon vessel, then a calculated amount of
substrate (catalyst/substrate = 1:100), a few drops of
thiophene (to suppress possible heterogeneous process),
base (0.07 mL) and degassed solvent (5 mL, distilled under
an N2 atmosphere) were added. All operations were quickly
carried out under air and the vessel was sealed up in stainless
steel autoclave. The autoclave was first purged with H2 (3 ×)
and then charged to 100 bar. After 8 h of hydrogenation at
r.t., the autoclave was unsealed and the reaction mixture was
(14) Ugo, R. Aspects of Homogeneous Catalysis: A Series of
Advances; Manfredi: Milano, 1970.
(15) (a) Jiang, Z.; Sen, A. J. Am. Chem. Soc. 1990, 112, 9655.
(b) Jiang, Z.; Sen, A. Organometallics 1993, 12, 1406.
(16) Munakata, M.; Yan, S.-G.; Maekawa, M.; Akiyama, M.;
Kitagawa, S. J. Chem. Soc., Dalton Trans. 1997, 4257.
(17) Barsan, F.; Karam, A. R.; Parent, M. A.; Baird, M. C.
Macromolecules 1998, 31, 8439.
Synlett 2009, No. 2, 271–275 © Thieme Stuttgart · New York

LETTER
Nucleophilic Homogeneous Hydrogenation by Iridium Complexes
275
(18) Typical Preparative Procedure: DIPEA was distilled from
ninhydrine, then twice from LiH. Substrate (50 mg) was
weighed into a Teflon vessel, then a calculated amount of
catalyst (to achieve the desired catalyst/substrate ratio),
DIPEA (500 equiv relative to Ir complex) and degassed
toluene (15 mL, distilled under an N2 atmosphere) were
added. Further operations were similar to the experimental
procedure given above. Products were purified by silica gel
column chromatography, eluting with Et2O.
(19) Sosnovskikh, V. Y.; Irgashev, R. A.; Barabanov, M. A.
Synthesis 2006, 2707.
(20) 3-Trifluoroacetylchromone, mixture of ketone and hydrate,
was obtained as a gift from A. Kotljatrov and V. Iaroshenko,
or synthesized by the known procedure.21 This mixture (0.8
g) was stirred with P2O5 (4.7 g) in CH2Cl2 (50 mL) for 30 h
(control by 19F NMR), and then filtrated under nitrogen. The
filtrate was evaporated and dried under high vacuum, to give
10 (0.3 g). 1H NMR data were similar to those published.19
13C NMR (100 MHz, CDCl3/TMS): d = 115.73 (q, J = 289.4
Hz, CF3), 118.34 (8-CH), (119.14, 3-CH), 124.87 (10-C),
126.65 (5-CH), 127.09 (6-CH), 135.04 (7-CH), 155.551
(9-C), 163.07 (q, J = 2.2 Hz, 2-CH), 172.65 (4-C), 179.38 (q,
J = 39.2 Hz, COCF3). 19F NMR (376 MHz, CDCl3/CFCl3):
d = –74.81.
(21) Yokoe, I.; Maruyama, K.; Sugita, Y.; Harashida, T.;
Shirataki, Y. Chem. Pharm. Bull. 1994, 42, 1697.
Synlett 2009, No. 2, 271–275 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.