Selective partial reduction of quinolines: Hydrosilylation vs. transfer hydrogenation

Article abstract of DOI:10.1016/j.jorganchem.2008.02.004

Two mild and versatile catalytic routes give regioselective hydrogenation of the heterocyclic ring of quinoline derivatives avoiding the high pressures of hydrogen required in the conventional hydrogenation route. Hydrosilylation using H₃SiPh and catalyzed by [Rh(nbd)(PPh₃)₂]PF₆ at room temperature gives dihydroquinoline, a product not obtainable via direct hydrogenation. Hydrosilylation of the C{double bond, long}N bond of PhCH{double bond, long}NPh is also observed under these conditions while PhCH{double bond, long}CHPh is unreactive. Initial in situ disproportionation of phenylsilane to H₂SiPh₂ and SiH₄, catalyzed by the same catalyst, was required for substrate reduction, as SiH₄ proved to be the active reductant. No N-silyl intermediates were ever observed, hydrolysis presumably occurring in situ. This disproportionation reaction is of potential use in gaining access to silane (SiH₄), a material otherwise not readily available. In a separate approach, transfer hydrogenation from isopropanol using [Ir(cod)(NHC)PPh₃]BF₄ (NHC = 1-neopentyl-4-n-butyl triazole-5-ylidene) as catalyst exclusively produces the tetrahydro product.

Full text of DOI:10.1016/j.jorganchem.2008.02.004

Available online at www.sciencedirect.com
Journal of Organometallic Chemistry 693 (2008) 1815–1821
www.elsevier.com/locate/jorganchem
Selective partial reduction of quinolines: Hydrosilylation
vs. transfer hydrogenation
Adelina M. Voutchkova, Dinakar Gnanamgari, Charles E. Jakobsche, Chase Butler,
Scott J. Miller, Jonathan Parr, Robert H. Crabtree *
Chemistry Department, Yale University, 225 Prospect Street, New Haven, CT 06520-8107, United States
Received 17 December 2007; received in revised form 1 February 2008; accepted 1 February 2008
Available online 9 February 2008
Abstract
Two mild and versatile catalytic routes give regioselective hydrogenation of the heterocyclic ring of quinoline derivatives avoiding the
high pressures of hydrogen required in the conventional hydrogenation route. Hydrosilylation using H3SiPh and catalyzed by
[Rh(nbd)(PPh3)2]PF6 at room temperature gives dihydroquinoline, a product not obtainable via direct hydrogenation. Hydrosilylation
of the C@N bond of PhCH@NPh is also observed under these conditions while PhCH@CHPh is unreactive. Initial in situ dispropor-
tionation of phenylsilane to H2SiPh2 and SiH4, catalyzed by the same catalyst, was required for substrate reduction, as SiH4 proved
to be the active reductant. No N-silyl intermediates were ever observed, hydrolysis presumably occurring in situ. This disproportionation
reaction is of potential use in gaining access to silane (SiH4), a material otherwise not readily available. In a separate approach, transfer
hydrogenation from isopropanol using [Ir(cod)(NHC)PPh3]BF4 (NHC = 1-neopentyl-4-n-butyl triazole-5-ylidene) as catalyst exclusively
produces the tetrahydro product.
Ó 2008 Elsevier B.V. All rights reserved.
Keywords: Silane disproportionation; Rhodium; Iridium; Phosphine; 1,2-Dihydroquinoline
1. Introduction
and iridium [10–14]. In addition, direct hydrogenation
always yields the tetrahydro product [15]. The dihydro
product, to the best of our knowledge, has been obtained
from quinoline only by reduction using stoichiometric
amounts of LiAlH4 [16,17]. We now describe the catalytic
reduction strategies that can be selective either for the dihy-
dro or the tetrahydro product.
Catalytic hydrosilylation of alkenes, alkynes, and car-
bonyl compounds is well known [18], but hydrosilylation
of nitrogen-containing compounds has not yet received as
much attention. Hydrosilylation of imines, although still
challenging, has been reported with rhodium [19], zinc
[20], titanium [21], iridium [22], and palladium [23] catalysts.
Reduction of pyridines via hydrosilylation has also been
reported using palladium [24] and titanium [25] catalysts.
We have now screened catalysts for imine hydrosilylation
and then applied the best ones to aromatic heterocycles
such as quinolines. It has been generally assumed that
Regioselective partial reduction of quinoline derivatives
is a convenient route to dihydro and tetrahydroquinolines.
Tetrahydroquinolines are important organic synthetic
intermediates for alkaloids, agrochemicals, dyes, and
pharmaceuticals, while dihydroquinolines can be useful
antimicrobials [1] and precursors to biologically active
compounds. The catalytic reduction of quinolines is a chal-
lenging task in organic synthesis and normally requires
high pressures of hydrogen when carried out by homoge-
neous or heterogeneous hydrogenation with Pt [2], Rh [3–
5], Ru [6,7], or Ir [8]. Enantioselective variants are known
with chiral bidentate phosphine complexes of rhodium [9]
*
Corresponding author.
E-mail address: robert.crabtree@yale.edu (R.H. Crabtree).
0022-328X/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2008.02.004

1816
A.M. Voutchkova et al. / Journal of Organometallic Chemistry 693 (2008) 1815–1821
hydrosilylation of imines (Eq. (1)1) involves formation of a
N–Si bond subsequently cleaved during work-up. This is
not the only possibility, however. We also consider two
other mechanisms outlined in Eqs. (2) and (3). In Eq. (2),
protons from water and hydrides from the silane reduce
the substrate with formation of siloxane as side product.
In Eq. (3), one hydrogen from each of 2 equiv. silane
reduces the substrate with formation of disilane as side
product [23].
ð5Þ
Here we use both hydrosilylation and transfer hydrogena-
tion as mild and efficient methods for selectively reducing
quinoline derivatives either to 1,2-dihydroquinoline or to
1,2,3,4-tetrahydroquinoline.
2. Results and discussion
2.1. Hydrosilylation
An initial catalytic screen of selected Rh, Ir, Ru, Pt, and
Pd compounds at 3% loading with diphenylsilane in
reagent-grade toluene with 4 equiv. silane in a closed
10 mL vial revealed very low activity for the reaction. We
did, however, identify the most promising catalyst candi-
date, [Rh(cod)(PPh3)2]PF6 (1). Moving to tricyclohexyl-
phosphine did not have a significant effect on activity.
Analogous compounds with diphenylphosphinoethane
(DPPE) gave significantly lower catalytic activity. Simi-
larly, substituting one or both phosphines with N-heterocy-
clic carbenes (IMe) also proved to be deleterious to
catalytic activity. The analogous iridium phosphine and
IMe catalysts were also tested, but their activities were also
significantly lower. To optimize the silane, we screened
ð1Þ
ð2Þ
ð3Þ
Applying the most widely accepted mechanism of Eq. (1) to
quinoline gives the pathway of Eq. (4) but, as shown below,
the reaction proves to be more complex than this.
ð4Þ
The second route we investigated, transfer hydrogenation,
is of interest due to its use of isopropanol, a greener reduc-
tant, thus avoiding costly reducing agents such as LiAlH4
that also generate undesirable stoichiometric waste byprod-
ucts. A recent review summarizes the mechanistic studies
on transition metal catalyzed hydrogen transfer reactions
[26]. The reaction has been extensively studied for C@O
reduction and has led to important applications such as
the racemization of chiral alcohols [27] and asymmetric
reduction of ketones [28]. Extension to C@N reduction
has proved valuable in the pharmaceuticals industry [29].
Noyori et al. have used formic acid/triethylamine azeo-
trope to reduce imines under mild conditions with high
enantiomeric excesses [30]. Transfer hydrogenation of
quinoline and its derivatives, however, has not been exten-
sively studied [31]. We recently reported a catalytic system
that was active for imine reduction and reductive amina-
tion via hydrogen transfer [32] and we now report the
extension of this system to transfer hydrogenation of het-
eroaromatics (Eq. (5)).
several silanes, as shown in Table 1. Phenylsilane gave
the best activity, with 80% total yield after 8 h to give
two products after base work-up, dihydroquinoline and
tetrahydroquinoline.
The initial activation of the catalyst normally involves
loss of the diene ligand by hydrosilylation and this step is
typically enhanced if cyclooctadiene (cod) is replaced with
norbornadiene (nbd) [35]. The complex [Rh(nbd)(PPh3)2]PF6
(2) indeed proved more active and gave 95% yield under
the same conditions.
Table 1
Hydosilylation of quinoline using 3 mol% [Rh(cod)(PPh3)2]PF6 and
4 equiv. silane, in toluene at 90 °C for the time specified
Entry Silane
Product ratio
(dihydro:tetrahydro)a
Time
(h)
Yield
(%)
1
2
3
4
H2SiPh2
H3SiPh
HSiCl3
0:100
10:90
0:100
24
8
24
48
12
80
46
0
HSi[(CH3)3SiO]3 n/a
a
Dihydro = 1,2-dihydroquinoline and tetrahydro = 1,2,3,4-tetrahydro-
quinoline.
1
„SiH@organosilane.

A.M. Voutchkova et al. / Journal of Organometallic Chemistry 693 (2008) 1815–1821
1817
Table 2
dro/tetrahydro, Eq. (6)), while in larger scale reactions
using dried reagent-grade toluene, the dihydro product is
slightly less favored (>95%, 87:13 dihydro:tetrahydro).
Excess silane was found to be necessary for optimal yields
but no direct correlation was found between amount of
silane used and dihydro/tetrahydro ratio.
Influence of solvent in the reaction of quinoline with H3SiPh
Entry Solvent
Time (h) Yield (%) Product (dihydro:tetrahydro)a
1
2
3
4
Toluene 24
98
94
72
0
57:43
45:55
42:58
n/a
THF
DCM
MeCN
24
24
48
Conditions: 3 mol % [Rh(nbd)(PPh3)2] PF6 and room temperature, 24 h.
a
Products as in Table 1; DCM = dichloromethane.
ð6Þ
In no case was a product containing a C–Si bond obtained
in any of the reactions studied. In a competition experi-
ment between trans-stilbene and N-benzylideneaniline un-
der the catalytic conditions of Eq. (7), the only product
observed was the hydrogenated imine; the stilbene was
not consumed under the reaction conditions. Thus our
reaction conditions are not effective for standard hydrosily-
lation of olefins, accounting for the fact that we do not ob-
serve silicon–carbon bonds in the products from quinoline.
In an effort to make the reaction selective for either one of
the two products, we examined variations in solvent, tem-
perature of the reaction and the amount of silane. Table 2
shows that the reaction is strongly favored by non-polar sol-
vents, and that increasing solvent polarity or coordinating
power can be deleterious. For example, toluene and THF
were highly satisfactory, but MeCN failed.
Table 3 shows that low temperature favored the dihydro
product but conversion was low, perhaps because of reduced
ð7Þ
catalyst solubility at ꢀ20 °C. A temperature of ꢀ10 °C pro-
vides a convenient selectivity-activity compromise.
If the PhSiH3 and catalyst are combined in the absence of
substrate, a gas is evolved over a few minutes. If the gas-
eous products of this reaction are released into the open
air, they spontaneously inflame. This, taken with the for-
mation of 1 equiv. Ph2SiH2 strongly implies that the dis-
proportionation of PhSiH3, as shown in Eq. (8), is
catalyzed under these conditions. Indeed, the dispropor-
tionation proceeds essentially to completion over 20–
30 min as judged by the formation of Ph2SiH2. Rh-cata-
lyzed disproportionations of silanes have been reported
previously [33]. Confirmation of the identity of the gaseous
product as SiH4 was obtained from GC–MS data acquired
for the reaction of phenylsilane in toluene solution cata-
lyzed by 2. This gave a set of intense peaks from 28 to 32
m/z. No prior report of the mass spectral characteristics
of SiH4 seems to exist, but we propose that this pattern
(see Section 4) constitutes the signature of SiH4. In a con-
trol experiment involving phenylsilane in the absence of
catalyst but under otherwise identical conditions, no peaks
were observed at 28–32 m/z, but a similar pattern of multi-
ple peaks was seen at 105–108 m/z that we assign to
phenylsilane.
In reactions catalyzed by [Rh(cod)(PPh3)2]PF6 the frac-
tion of dihydro product was somewhat smaller than with
[Rh(nbd)(PPh3)2]PF6. This would imply that changing the
olefin ligands affects not only the activation, but presum-
ably somewhat different active catalysts may be formed in
the two cases. This could be due to the fact that unlike
the more reactive norbornadiene, cyclooctadiene is not
completely removed from the metal by reaction with the
silane. These reactions must be performed under anhy-
drous conditions to give optimal yields presumably due
to sensitivity of the silane to catalyzed hydrolysis. The
dihydro/tetrahydro ratio is somewhat variable depending
on the purity of the silane and solvent used, but the yield
always remains above 95%. The formation of dihydro
product is enhanced in NMR scale reactions with fresh
silane and anhydrous deuterated toluene (>98%, 97:3 dihy-
Table 3
Variation of temperature in the reaction of 2-methyl quinoline with
phenylsilane
Temperature (°C)
% Conversion
Product (dihydro:tetrahydro)a
H3SiPh ca!talyst H2SiPh2 þ H4Si
ð8Þ
ꢀ20
ꢀ10
0
25
90
14
70
78
98
70
93:8
84:16
75:25
57:43
28:72
A possible mechanism may involve initial hydrosilyla-
tion of the C@N bond, followed by isomerization of the
C@C bond to the imine position. The presence of excess si-
lane is thus necessary for the second hydrosilylation. The
stoichiometry of the reaction involves two moles of PhSiH3
for the dihydro product, suggesting one mole of SiH4 is
Conditions: 3 mol% [Rh(nbd)(PPh3)2] PF6, 24 h.
a
Products as in Table 1.

1818
A.M. Voutchkova et al. / Journal of Organometallic Chemistry 693 (2008) 1815–1821
a much faster rate than the rate of hydrogenation. In an
attempt to identify the reductant, we conducted a reaction
in which a solution of PhSiH3 and rhodium catalyst was
placed in a small vessel enclosed within a larger container
holding the substrate and catalyst in toluene solution
(Scheme 1). This ensured that essentially only the volatile
SiH4 was able to diffuse into the larger closed vessel and
react with the catalyst and substrate. Within 12 h the
hydrogenated product was indeed formed in the outer ves-
sel, albeit in a slightly reduced yield (68%). We also tested
whether the rhodium catalyst is involved in the reaction
with substrate, or if its role is limited to producing reactive
silane gas which attacks the quinoline directly. We thus
repeated the reaction described above using the same two
vessels, but omitting the catalyst from the outer vessel so
that the gas produced was allowed to react with the sub-
strate alone. These reactions did not show any product for-
mation, indicating that the rhodium catalyst acts first to
disproportionate phenylsilane, and then also to catalyze
the reaction between the substrate and the gas produced
from this reaction. Diphenylsilane, as noted earlier, was
Scheme 1.
sufficient for this step, so each SiH4 provides two hydro-
gens in the final product. The stoichiometry of the hydro-
silylation of the imine is therefore most consistent with
the mechanism of Eq. (1) where 1 mole of SiH4 is required.
The disproportionation raises uncertainty about the
active reductant in these reactions. PhSiH3 is an unlikely
candidate because it is depleted by disproportionation at
Table 4
Regioselective reduction of N-heteroaromatic compounds by (a) hydrosilylation using 3 mol% 1 and 4 equiv. silane, in toluene at room temperature and
(b) transfer hydrogenation using 1 mol% 3 in isopropanol 0.5 equiv. K2CO3 at 82 °C for the time specified
Entry
Substrate
Product
Hydrosilylation
Yield (%)
>95
Transfer hydrogenation
Time (h)
6
Yield (%)
–
Time (h)
–
1
–
–
67
24
2
>95
–
6
–
–
–
50
24
3
4
10
12
100
24
5
12
12
0
0
0
87
24
5
6
93
82
12
12
<5
<5
24
24

A.M. Voutchkova et al. / Journal of Organometallic Chemistry 693 (2008) 1815–1821
1819
found to be inactive for this reaction. In a further control,
catalyst was omitted from the inner vessel to check that
PhSiH3 alone is not equally capable of carrying out the
reaction by diffusion into the outer vessel; a much lower
yield of hydrogenation product was obtained (14%), result-
ing from the insufficient volatility of this silane.
rhodium analogue also gave similar results, but the use of
the iridium catalyst is preferable due to its lower cost. We
achieved a TON of 67 which is comparable to a rhodium
bis-bipyridine complex (TON = 54, T = 82 °C) [36]. Cata-
lyst 1, [Rh(cod)(PPh3)2]PF6, was not active under these con-
ditions (<5% conversion to 1,2,3,4-tetrahydroquinoline).
No extensive mechanistic information is yet available
but some further observations were made. No N-silyl inter-
mediates were ever observed, so if the direct hydrosilylation
mechanism of Eq. (1) applies, spontaneous hydrolysis of
the N–Si bond must occur under the reaction conditions.
The pathway of Eq. (2) involving silane hydrolysis could
not be definitively eliminated, however careful drying of
the solvents did not lower the efficiency of the reaction
and addition of water was deleterious.
Attempts to extend this methodology to other heteroar-
omatic compounds showed that the method is not yet gen-
eral (Table 4). Only one other substrate, pyrazine, showed
complete conversion. On the hypothesis that the nitrogen
lone pairs of the substrate were blocking the catalytic sites,
two of the heterocyclic substrates pyridine and 2,20-bipyri-
dine were N-methylated to give 1-methylpyridinium iodide
and 1-methyl-2,20-bipyridinium iodide respectively. These
modified substrates, however, did not react. Future work
may identify the right conditions to extend the generality
of this procedure.
2.2. Transfer hydrogenation
Prior work in our group has shown that catalyst 3 is par-
ticularly active for the reduction of imines under transfer
hydrogenation conditions [32]. To test this catalyst for
transfer hydrogenation of nitrogen heterocycles, we initially
took quinoline as substrate. Treatment of quinoline
(1 mmol) with 0.5 equiv. K2CO3 and 10 mL of i-PrOH
under refluxing conditions for 24 h with 1 mol% of iridium
catalyst 3 yielded 67% of 1,2,3,4-tetrahydroquinoline. The
Similar results were obtained using a different hydrogen
source (HCOOH) and no base. Treating 1 mmol of quino-
line with 1 equiv. HCOOH and 1 mol% catalyst 3 yielded
Fig. 1. Mass spectra of the vapor phase from (a) the reaction of phenylsilane in toluene solution catalyzed by 2 and (b) of phenylsilane in toluene in the
absence of catalyst 2.

1820
A.M. Voutchkova et al. / Journal of Organometallic Chemistry 693 (2008) 1815–1821
40% of 1,2,3,4-tetrahydroquinoline at 80 °C. Increasing the
quantity of HCOOH to 5 equiv. increased the yield to 60%.
No further improvement was seen with higher amounts
and this method was equally limited in its scope (Eq. (9)).
In no case was the dihydroquinoline ever seen.
previously described [32]. Silane (SiH4) is a potentially haz-
ardous, pyrophoric material implying that particular care is
needed in carrying out these reactions.
4.2. Hydrosilylation reactions
4.2.1. Example of general procedure: preparation of 1,2-
dihydroquinaldine
ð9Þ
A
flame-dried Schlenk tube was charged with
[Rh(cod)(PPh3)2]PF6(10.0 mg, 10.2 lmol), 1,3,5-tri-tert-
butylbenzene as internal standard, a stir bar and placed
under a nitrogen atmosphere. Toluene (1.00 mL) was
added via syringe, followed by quinaldine (50.0 lL) and
then phenylsilane (200 lL) in quick succession. After stir-
ring at room temperature for 6 h, the flask was flushed with
nitrogen gas, then carefully opened. 1 mL of a 1:15 solution
of 50% aqueous NaOH/methanol was added dropwise over
5 min. After 10 min of additional stirring, the solvent was
removed and the residue was dissolved in 2 mL dichloro-
methane and extracted with water. The organic fractions
was dried over MgSO4 and purified by flash chromatogra-
phy on silica gel. 1H and 13C NMR spectra were compared
with literature data for all products.
2.3. Generality/selectivity of methods
A variety of heterocycles were surveyed as substrates for
reduction using both methods. For hydrosilylation, the
quinolines, acridine and indole all gave the dihydro prod-
ucts in satisfactory yield. Isoquinoline gave a mostly the
tetrahydro product. Pyrazine was inefficiently reduced to
the tetrahydro compound, however. Transfer hydrogena-
tion showed less generality, only the quinolines and pyra-
zine being efficiently reduced to the tetrahydro products.
7,8-benzoquinoline gave no products under either condi-
tions, perhaps because this substrate cyclometallates very
easily. For a fuller list of substrates, see Supplementary
Material.
4.2.2. Mass spectral identification of SiH4
A 1-mL vial fitted with a screw cap with a septum was
charged with a stir bar and [Rh(nbd)(PPh3)2]PF6 (2)
(10.0 mg, 10.2 lmol) under an inert atmosphere. Toluene
(0.10 mL) was added, and the catalyst was allowed to dis-
solve with stirring. Phenylsilane (300 lL) was added and
reaction was stirred for additional 10 min. The evolution
of a gas was observed. The gas from the headspace of the
vial was injected into a GC–MS, and two peaks were
observed – one, with retention time 1.62 min, with a mass
spectrum of 28–32 m/z (Fig. 1a) and another at 2.71 min,
which gave a major set of peaks at 105–108 m/z (Fig. 1b).
In a control experiment, identical to the one above but
without catalyst 2, only one peak was observed in the
GC trace at 2.71 min retention time, with a mass spectrum
identical to that of Fig. 1b.
3. Conclusions
We have shown two mild and versatile routes the regio-
selective hydrogenation of quinoline derivatives without
the use of high pressures of hydrogen gas. Hydrosilylation
offers a selective reduction of the quinoline to dihydroquin-
oline, which is not possible with direct hydrogenation,
while transfer hydrogenation uses only isopropanol as a
hydrogen source. The reaction is very selective for particu-
lar substrates. The disproportionation of phenylsilane by
[Rh(cod)(PPh3)2]PF6 observed is of potential use in gaining
access to silane (SiH4), otherwise not readily available.
4. Experimental
4.3. Transfer hydrogenation of quinoline
4.1. General considerations
4.3.1. General procedure for reduction of heterocycles with
isopropanol
All the operations were carried out under a nitrogen
atmosphere using standard Schlenk techniques. Toluene
was distilled from sodium-benzophenone. All starting
materials and reagents were obtained from commercial
sources and used as received unless otherwise noted. All
An oven-dried flask was charged with quinoline (118 lL,
1.00 mmol), K2CO3 (69 mg, 0.50 mmol), 1,3,5-tri-tert-
butylbenzene as internal standard and the catalyst
[Ir(cod)(NHC)PPh3]BF4(3) (8.4 mg, 9.95 lmol) at 1.0%
loading. The flask was evacuated and filled with N2,
10.0 mL of dry degassed i-PrOH was added, and the mix-
ture refluxed for 24 h. An aliquot (0.50 mL) of the mixture
was quenched with 2 mL of pentane and the resulting solu-
tion was filtered through Celite to remove insoluble inor-
ganic material. The volatiles were then removed under
1
glassware was dried overnight prior to use. H, 13C and
31P NMR spectra were obtained using a Bruker spectrom-
eter operating at 400 MHz or 500 MHz. Chemical shifts
are reported in ppm with the residual solvent peak as an
internal reference. [Rh(nbd)Cl]2 was prepared by the
method of Abel [34], while [Rh(cod)(PPh3)2]PF6 and
[Rh(nbd)(PPh3)2]PF6 were prepared by the method of
Schrock [35]. Preparation of compound 3 has also been
1
vacuum and the conversion was checked with H NMR.
The conversion is based on moles of substrate and compar-

A.M. Voutchkova et al. / Journal of Organometallic Chemistry 693 (2008) 1815–1821
1821
[6] R.A. Sanchez-Delgado, E. Gonzalez, Polyhedron 8 (1989) 1431.
[7] R. Kuwano, M. Kashiwabara, Org. Lett. 8 (2006) 2653.
[8] C.S. Chin, Y. Park, B. Lee, Catal. Lett. 31 (1995) 239.
ison to internal standard. The data reported are based on
an average of two catalytic runs.
[9] R. Kuwano, M. Kashiwabara, K. Sato, T. Ito, K. Kaneda, Y. Ito,
Tetrahedron-Asymmetry 17 (2006) 521.
[10] W.-B. Wang, S.-M. Lu, P.-Y. Yang, X.-W. Han, Y.-G. Zhou, J. Am.
Chem. Soc. 125 (2003) 10536.
[11] K.H. Lam, L. Xu, L. Feng, Q.H. Fan, F.L. Lam, W. Lo, A.S.C.
Chan, Adv. Synth. Catal. 3 (2005).
[12] R. Kuwano, K. Kaneda, T. Ito, K. Sato, T. Kurokawa, Y. Ito, Org.
Lett. 6 (2004) 2213.
[13] Y.J. Zhao, Y.Q. Wang, Y.G. Zhou, Chin. J. Catal. 26 (2005)
737.
[14] S.M. Lu, X.W. Han, Y.G. Zhou, Adv. Synth. Catal. 346 (2004) 909.
[15] E. Baralt, S.J. Smith, J. Hurwitz, I.T. Horvath, R.H. Fish, J. Am.
Chem. Soc. 114 (1992) 5187.
[16] T.P. Forrest, D.A. Dauphinee, S.A. Deraniyagala, Can. J. Chem. 63
(1985) 412.
[17] B. Ferdinand, Chem. Berichte 85 (1952) 390.
4.3.2. General procedure for reduction of quinoline with
HCOOH
An oven-dried flask was charged with quinoline (118 lL,
1.00 mmol), 1,3,5-tri-tert-butylbenzene as internal stan-
dard, HCOOH acid (185 lL) (5 equiv.) and the catalyst
[Ir(cod)(NHC)PPh3]BF4 (3) (8.4 mg, 9.95 lmol) at 1.0%
loading. The flask was then evacuated and filled with N2.
The mixture was then placed in an oil bath at 82 °C to
reflux for 24 h. The mixture was then quenched with pen-
tane (2 mL) and the resulting solution filtered through Cel-
ite to remove insoluble inorganic material. The volatiles
were then removed under vacuum and the conversion
1
checked with H NMR. The conversion is based on moles
[18] K.A. Horn, Chem. Rev. 95 (1995) 1317.
of substrate and comparison to internal standard. The data
reported are based on an average of two catalytic runs.
[19] I. Ojima, T. Kogure, Y. Nagai, Tetrahedron Lett. 14 (1973) 2475.
[20] B.-M. Park, S. Mun, J. Yun, Adv. Synth. Catal. 348 (2006) 1029.
[21] J. Yun, S.L. Buchwald, J. Org. Chem. 65 (2000) 767.
[22] I. Takei, Y. Nishibayashi, Y. Arikawa, S. Uemura, M. Hidai,
Organometallics 18 (1999) 2271.
Acknowledgements
[23] P.K. Mandal, J.S. McMurray, J. Org. Chem. 72 (2007) 6599.
[24] N.C. Cook, J.E. Lyons, J. Am. Chem. Soc. 88 (1966) 3396.
[25] L. Hao, J.F. Harrod, A.-M. Lebuis, Y. Mu, R. Shu, E. Samuel,
H.-G. Woo, Angew. Chem., Int. Ed. 37 (1998) 3126.
[26] J.S.M. Samec, J.-E. Ba¨ckvall, P.G. Andersson, P. Brandt, Chem.
We thank the US Department of Energy for funding.
Appendix A. Supplementary data
Soc. Rev. 35 (2006) 237.
˚
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.jorganchem.
2008.02.004.
´
[27] B. Martın-Matute, J.B. Aberg, M. Edin, J.-E. Ba¨ckvall, Chem. Eur.
J. 13 (2007) 6063.
[28] S. Gladiali, E. Alberico, Chem. Soc. Rev. 35 (2006) 226.
[29] S. Kobayashi, H. Ishitani, Chem. Rev. 99 (1999) 1069.
[30] R. Noyori, S. Hashiguchi, Acc. Chem. Res. 30 (1997) 97.
[31] K. Fujita, C. Kitatsuji, S. Furukawa, R. Yamaguchi, Tetrahedron
Lett. 45 (2004) 3215.
References
[1] I. Abou-Nemeh, PCT Int. Appl. (2007), Patent #: WO2007044398.
[2] M.V. Honel, F.W. Vierhapper, J. Chem. Soc., Perkin Trans. 1 (1980)
1933.
[32] D. Gnanamgari, A. Moores, E. Rajaseelan, R.H. Crabtree, Organo-
metallics 26 (2007) 1226.
[33] I. Ojima, S. Inaba, T. Kogure, J. Organomet. Chem. 55 (1973) C7.
[34] E.W. Abel, M.W. Bennett, G.J. Wilkinson, Chem. Soc. (1959) 3178.
[35] R.R. Schrock, J.A.J. Osborn, J. Am. Chem. Soc. 93 (1971) 2397.
[36] P. Frediani, L. Rosi, L. Cetarini, M. Ferdiani, Inorg. Chim. Acta 359
(2006) 2650.
[3] M. Rosales, R. Vallejo, J. Soto, G. Chacon, A. Gonzalez, B.
Gonzalez, Catal. Lett. 106 (2006) 101.
[4] F. Fache, Synlett 15 (2004) 2827.
[5] S. Murahashi, Y. Imada, Y. Hirai, Bull. Chem. Soc. Japan 62 (1989)
2968.