Application of ferrocenylimidazolium salts as catalysts for the transfer hydrogenation of ketones

Article abstract of DOI:10.1002/aoc.2947

Ferrocenylimidazolium salts with methylene and phenyl groups bridging the ferrocenyl and alkylimidazolium moieties were synthesized and characterized by spectroscopic and analytical methods. Crystal structures of two new compounds are also reported. Cyclic voltammetry was used to analyze the influence of the two bridging groups or spacers on electrochemical properties of the salts relative to the shifts in the formal electrode or peak potentials (E⁰ or E_1/2) of the ferrocene/ferrocenium redox couple. Results from this study showed that all the salts exhibited higher electrode potentials relative to ferrocene, which is due to the electron-withdrawing effect of the imidazolium ion on the ferrocenyl moiety. Application of the salts as catalysts in transfer hydrogenation of ketones resulted in high conversion of saturated ketones to corresponding alcohols and turnover numbers as high as 1880. The catalysts were chemoselective towards reduction of the C=C bonds of conjugated 3-penten-2-one and 4-hexen-3-one to yield saturated ketones, while unconjugated 5-hexen-2-one was hydrogenated to an unsaturated alcohol. Copyright

Full text of DOI:10.1002/aoc.2947

Full Paper
Received: 21 August 2012
Revised: 25 September 2012
Accepted: 16 October 2012
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI 10.1002/aoc.2947
Application of ferrocenylimidazolium salts as
catalysts for the transfer hydrogenation
of ketones
Monisola I. Ikhilea, Muhammad D. Balaa*, Vincent O. Nyamoria and
J. Catherine Ngilab
Ferrocenylimidazolium salts with methylene and phenyl groups bridging the ferrocenyl and alkylimidazolium moieties were
synthesized and characterized by spectroscopic and analytical methods. Crystal structures of two new compounds are also
reported. Cyclic voltammetry was used to analyze the influence of the two bridging groups or spacers on electrochemical
properties of the salts relative to the shifts in the formal electrode or peak potentials (E0 or E1/2) of the ferrocene/ferrocenium
redox couple. Results from this study showed that all the salts exhibited higher electrode potentials relative to ferrocene,
which is due to the electron-withdrawing effect of the imidazolium ion on the ferrocenyl moiety. Application of the salts as
catalysts in transfer hydrogenation of ketones resulted in high conversion of saturated ketones to corresponding alcohols
and turnover numbers as high as 1880. The catalysts were chemoselective towards reduction of the C═C bonds of conjugated
3-penten-2-one and 4-hexen-3-one to yield saturated ketones, while unconjugated 5-hexen-2-one was hydrogenated to an
unsaturated alcohol. Copyright © 2013 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this article.
Keywords: ferrocenylimidazolium salts; hydrogen transfer catalysis; cyclic voltammetry; X-ray diffraction
redox couple, the redox potential of which depends on the imidazo-
Introduction
lium ring N-substituents (R in Schemes 1 & 2).[18] Also, the interaction
Transfer hydrogenation of ketones to the corresponding alcohols
has become an important transformation in organometallic
chemistry, especially as applied to the pharmaceutical[1] and fine
chemical industries.[2] Much of the work conducted to date on
catalytic transfer hydrogenation of ketones has utilized effective
but expensive and sometimes toxic transition metals such as irid-
ium,[3] platinum,[4] rhodium,[5] gold[6] and ruthenium.[7] Therefore,
the search for inexpensive, mild and environmentally friendly
catalysts is important in order to expand the scope of available
routes to achieving this important reaction. Iron and its com-
pounds such as ferrocene satisfy these requirements.
The synthesis of compounds based on ferrocene such as
ferrocenylimidazolium salts has thus attracted interest because
of the rich chemistry and applications associated with it.[8] In addi-
tion, the functionalization of ferrocene with imidazolium salts gives
them unique electronic properties, such as the ability to stabilize
reactive carbocations.[9] Also, the powerful donor capacity of
ferrocene is in principle advantageous to additional stabilization
of electron-deficient carbene moieties and metal centres in high
oxidation states.[10,11] This has led to numerous industrial applica-
tions in medicine,[12,13] catalysis,[14–21] as sensors[22] and as compo-
nents of immunoassay reagents.[21,23–28]
between the ferrocene and the imidazolium ring depends on the na-
ture of the bridge between the two moieties.[29] Thomas et al.[30]
have synthesized novel ferrocenyl compounds for use as anion
recognition molecules, where they investigated the electrochemical
properties of the salts by the technique of cyclic voltammetry (CV).
Bai et al.[31] have also synthesized ferrocenylbenzimidazolium salts
and investigated the binding constants of halide (Clꢀ, Brꢀ and Iꢀ)
ions to the cation. Studies by Bildstein et al.[32] on the electrochemi-
cal properties of ferrocenylbenzimidazolium salts, using CV, showed
significant electronic communication between the imidazolium moi-
ety and the N-ferrocenyl substituent. The CV of iron(II) N-heterocyclic
carbene (NHC) complexes[33,34] has also been conducted in
order to establish the influence of the NHC ligand on the catalytic
activity of the iron centre. Hence numerous examples exist in the
literature that have established the popularity and wide application
of ferrocenylimidazolium salts.[10,35–39]
Furthermore, due to their ability to stabilize various metal centres,
their application as ligand precursors in the synthesis of ferrocenyl
*
Correspondence to: Muhammad D. Bala, School of Chemistry and Physics,
University of KwaZulu-Natal, Private Bag X54001, Durban 4000, South Africa.
E-mail: bala@ukzn.ac.za
Most importantly, applications of these salts in electrochemical
processes and as ligand precursors have been noted and, due to
the presence of two electro-active groups, i.e. ferrocene and
imidazole, ferrocenylimidazolium salts have also found use as
molecular recognition species.[29] The presence of ferrocene in
ferrocenylimidazolium salts gives rise to a ferrocene/ferrocenium
a School of Chemistry and Physics, University of KwaZulu-Natal, Durban 4000,
South Africa
b Department of Applied Chemistry, University of Johannesburg, Doornfontein
2028, Johannesburg, South Africa
Appl. Organometal. Chem. 2013, 27, 98–108
Copyright © 2013 John Wiley & Sons, Ltd.

Ferrocenylimidazolium salts as transfer hydrogenation catalysts
H
C
O
LiAlH4
OH
C
Fe
1
Fe
Ether, reflux
H
H
(a)
96%
O
H
C
H
H
C
H
N
N
N
N
OH
N
N
Fe
Fe
(b)
(c)
Dry CH2Cl2, 1 h, reflux
53%
2
H
C
H
H
N
C
H
X-
N
+
N
RX
N
Fe
Fe
R
Reflux, N2, 15 h
72%
84%
92%
3 R = CH3
4 R = C2H5
5 R = C4H9
X = I
X = Br
X = Br
Scheme 1. Synthesis of ferrocenylmethyleneimidazolium salts.
(a)
NaNO2
HBF4
N
NH2
N
N2BF4
N
N
Ferrocene
N
N
Fe
Fe
6
30%
N
(b)
N+
N
X-
DCM
N
RX, 41 h
Fe
R
7
8
9
R = CH3
R = C2H5 X = Br
R = C4H9 X = Br
X = I
86%
64%
91%
Scheme 2. Synthesis of ferrocenylphenylimidazolium salts.
NHC complexes has been shown to be of great importance. For
instance, Bildstein et al.[40] have synthesized metal complexes of
W(0), Pd(II) and Hg(II) with ferrocenylimidazolium salts. Coleman
et al.[8] have also synthesized and structurally characterized a
palladium(II) ferrocenyl NHC complex. In addition, Seo et al.[41] have
successfully obtained chiral ferrocenylimidazolium salts and used
them as precursors in the synthesis of rhodium and iridium com-
plexes. When applied as catalysts in transfer hydrogenation of
ketones, these complexes have shown moderate enantioselectiv-
ities. In 2009, Jiang et al.[42] reported good enantioselectivities
for the transfer hydrogenation of ketones, catalysed by Rh(I)
complexes of planar ferrocene-based chiral NHC ligands. To the
best of our knowledge, this report presents the first direct use of
ferrocenylimidazolium salts as catalysts for the transfer hydrogenation
of a wide variety of ketones under relatively mild reaction conditions.
Results and Discussion
Synthesis
Syntheses of the two classes of salts reported herein, one
containing a methylene and the other a phenyl group linking
Appl. Organometal. Chem. 2013, 27, 98–108
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

M. I. Ikhile et al.
ferrocenyl to imidazolium moieties, were conducted via two
different routes. The ferrocenylmethyleneimidazolium salts
were synthesized following adaptation of our earlier reported
procedure.[43a] In this procedure (Scheme 1), ferrocenyl-
methyleneimidazole (2) was prepared from ferrocenylmetha-
nol (1), which in turn was prepared by the reduction of
ferrocenecarboxaldehyde with lithium aluminum hydride. The
target salts (3–5) were obtained through alkylation of 2 with
corresponding alkyl halides and subsequent solvent washing with
diethyl ether to ensure complete removal of any unreacted 2.
Compounds 3–5 were isolated in relatively high yields (72–92%).
Salts 3 and 4 were obtained as yellow powders, while 5 was
isolated as a dark-brown powder.
On the other hand, the synthesis of the ferrocenylphenylimida-
zolium salts was achieved by employing a Gomberg–Bachmann
reaction involving diazotisation of 4-(1H-imidazol-1-yl)aniline fol-
lowed by coupling of ferrocene to afford 6 in an improved yield
as compared to previously reported results for related reactions
(Scheme 2).[39] Alkylation of 6 with the corresponding alkylhalides
(RX) generated the target molecules 7–9 in moderate to high
yields. Salt metathesis (Scheme 3) with sources of counter-anions
yielded compounds 10–12 in high yields (81–96%). The metathe-
sis was conducted in order to widen the variety of counter-anions
and investigate in more detail their effects on the physical proper-
ties of isolated salts.
the ferrocenylphenylimidazolium salts, but when the alkyl length
was further lengthened to butyl a decrease in the melting point
was observed, which is in agreement with reported literature data
where it was attributed to increased degrees of freedom of the alkyl
side chain R.[12]
Spectroscopic Characterization
The NMR data give some indication of the level of electronic
interaction between the cationic centres and the various anions.
A measure of the degree of deshielding of the counter-anions
was observed by monitoring the shifts to higher resonances
(1H NMR) of the imidazole protons in the salts as compared to
the starting neutral compounds 2 and 6. This, as expected, is
proportional to electronegativity of counter-anions (Table 1).
Generally, the deshielding effect was more pronounced for
ferrocenylphenylimidazolium salts, which highlighted the effect
of unsaturation in the phenyl group. The effect of the alkyl chain
length (R) was studied in the two sets of salts by varying its size
from ethyl (4 and 8) to butyl (5 and 9). A similar trend was
observed for both sets of salts, where a shielding effect was
observed with a slight shift to lower frequency from 10.62 and
11.27 ppm to 10.56 and 11.23 ppm respectively, due to the in-
creased electron density from the alkyl groups.
The compounds were also analysed by infrared spectroscopy
and mass spectrometry (MS). They all exhibited similar infrared
spectral patterns. The band observed around 1528–1573 cmꢀ1
corresponds to the N-C-N stretch in neutral imidazole.[11] The
C-H and N-H stretching frequencies were observed between
2934–3079 and 3300–3639 cmꢀ1 respectively. The bands that
are characteristic of the presence of ferrocene in a molecule[11]
were observed around 820–1150 cmꢀ1. All the salts exhibited
characteristic molecular ion peaks (m/z = M+ ꢀ Xꢀ) in MS.
It is worth noting that the chemical nature of the spacer (aliphatic
or aromatic) that linked the ferrocenyl to the imidazolium moiety
had an effect on the physical properties of the salts. As a general
trend, the melting points of the ferrocenylmethyleneimidazolium
compounds 3–5 and 10–11 were lower than those of the
corresponding ferrocenylphenylimidazolium salts 8, 9 and 12, the
only exceptions being compounds 3 and 7 (see Table 1). The ob-
served trend is attributed to larger molecular sizes of the phenyl-
linked salts and extensive network of cooperative pi-pi stacking
interactions due to the phenyl linker in 8, 9 and 12 which provided
additional intermolecular attractive forces and stabilization in the
solid state. Thus compounds 8, 9 and 12 have higher melting
points than the corresponding methylene-linked compounds 3–5
and 10–11. Compound 10 was isolated as oil, which could be
due to a change in the size of the counter-anion. It is well estab-
lished that a good match in relative ionic size (cation-to-anion) is
required for solidification and crystallization of ionic salts.[43,44] In-
creasing the length of the alkyl chain on the imidazole from methyl
to ethyl (12 and 8), did not significantly affect the melting point of
Molecular Structures of Compounds 4 and 11
Compounds 4 and 11 were structurally analysed by X-ray
diffraction studies. Crystals suitable for analysis were obtained
by slow diffusion of hexane into saturated dichloromethane
(DCM) solutions of the respective compounds. ORTEP diagrams
of compounds 4 and 11 are shown in Figs 1 and 2 respectively.
Crystal and experimental refinement data are summarized in
Table 2.
H
H
C
H
+
C
H
+
N
NaY
N
(a)
X-
Y-
N
N
acetone, 24 h
Fe
Fe
R
R
10
11
R = CH3
R = C4H9
X = I
X = Br
Y = Br
Y = PF6
3
5
R = CH3
R = C4H9
81%
96%
(b)
N
N
NaBr
I-
Br-
CH3
+
+
N
N
acetone, 24 h
Fe
Fe
CH3
87%
12
7
Scheme 3. Anion exchange of ferrocenylimidazolium salts.
wileyonlinelibrary.com/journal/aoc
Copyright © 2013 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2013, 27, 98–108

Ferrocenylimidazolium salts as transfer hydrogenation catalysts
Table 1. Structural variations of the ferrocenylimidazolium salts
2–12
Compounds
–R
Xꢀ
m.p. (ꢁC)
1H NMR (ppm)a
2
None
CH3
None
I
80
7.45
9.96
3
145.5
92.3
4
C2H5
C4H9
None
CH3
Br
10.62
10.56
7.84
5
Br
85.5
6
None
I
161.5
140.5
184.5
158.5
Oil
7
10.65
11.27
11.23
9.97
8
C2H5
C4H9
CH3
Br
9
Br
10
11
12
Br
C4H9
CH3
PF6
Br
Paste
179.5
8.62
10.64
[a] chemical shift of imidazole proton
Figure 2. ORTEP representation of compound 11. Thermal ellipsoids are
represented at the 50% probability level. Selected bond lengths (Å): N1-
C11 = 1.484(15); Fe1-C10 = 2.005(13); N2-C15 = 1.481(17); N2-C14 = 1.335
(16); Fe1-C6 = 2.041(13). Selected bond angles (ꢁ): N1-C14-N2 = 107.5(12);
C14-N1-C12 = 110.9(12); C14-N2-C15 = 125.4(12); C10-Fe1-C1 = 120.4(6).
Table 2. Summary of crystal and experimental refinement data for
compounds 4 and 11
Compound
4
11
Formula
Formula weight
Crystal system
Space group
a (Å)
C
16H21BrFeN2O
C18H23F6FeN2P
468.20
393.11
Monoclinic
P21/c
Monoclinic
P21
12.8696(9)
7.5858(5)
16.8379(11)
90
9.7739(13)
16.4278(19)
12.6114(18)
90
b (Å)
c (Å)
a (ꢁ)
b (ꢁ)
94.021(2)
90
93.755(5)
90
g (ꢁ)
Cell volume (Å3)
1639.77(19)
4
2020.6(5)
4
Z
calcd (Mg mꢀ3-
)
1.592
1.539
Figure 1. ORTEP representation of compound 4. Thermal ellipsoids are
represented at the 50% probability level. Selected bond lengths (Å): N1-
C11 = 1.477(3); Fe1-C10 = 2.032(2); N2-C15 = 1.473(3); N2-C12 = 1.329(3);
Fe1-C6 = 2.016(2). Selected bond angles (ꢁ): N1-C12-N2 = 108.4(2); C12-
N1-C11 = 124.2(2); C12-N2-C15 = 124.5(2); C10-Fe1-C6 = 121.12(16).
D
Crystal size (mm3)
0.33 ꢂ 0.31 ꢂ 0.06
R1 = 0.0318
wR2 = 0.0748
R1 = 0.0462
wR2 = 0.0791
0.73 ꢂ 0.07 ꢂ 0.02
R1 = 0.0721
wR2 = 0.1442
R1 = 0.1815
wR2 = 0.1839
Final R indices
R indices (all data)
Compound 4 crystallized in the monoclinic space group P21/c
with one molecule in the asymmetric unit. The structure consists
of ferrocenyl moiety with methylene linker to ethylimidazolium
bromide and a molecule of water of crystallization.
Compound 11 crystallised in the monoclinic space group P21
with two molecules in the asymmetric unit. The structure consists
of methylene-linked ferrocenyl and butylimidazolium moieties
counterbalanced by a hexafluorophosphate anion.
substituent on the ferrocenyl moiety was easily evaluated by
comparison of the half-wave potentials (E1/2) of the salts relative
to ferrocene. The data presented in Table 3 show that all the E1/2
values (for compounds 2–12) shifted to more positive potentials
(0.482 to 0.595 V) when compared to ferrocene (0.436 V). This is
attributed to the electron-withdrawing ability of the imidazolium
moiety, which, by draining electron density away from the
ferrocenyl centre, made oxidation relatively more difficult. The
neutral ferrocenylimidazoles (2 and 6) have lower E1/2 values in
comparison to the corresponding salts. The cyclic voltammo-
grams of some of the salts containing halide counter-anions
(e.g. 4, 5 and 12) displayed second redox peaks, which were
attributed to the presence of iodide or bromide counter-ions in
the molecular sphere of the salts. The presence of iodide or
Cyclic Voltammetry
The ferrocenylimidazolium salts were analysed by cyclic voltam-
metry. The voltammograms are presented in Figs 3 and 4, where
the potentials were scanned in the anodic forward direction,
denoted by the arrow (!), from ꢀ1.00 to +1.00 V. A reversible
redox wave similar to unsubstituted ferrocene was observed
for all compounds 2–12. The influence of the imidazolium
Appl. Organometal. Chem. 2013, 27, 98–108
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

M. I. Ikhile et al.
Figure 3. Comparison of the cyclic voltammograms of compounds 2, 4, 5 and 10 with that of ferrocene. Arrows (! or ) indicate scan direction.
Figure 4. Comparison of cyclic voltammograms of ferrocenyl compounds 6, 8, 9 and 12 with that of ferrocene. Arrows (! or ) indicate scan
direction.
bromide counter-ions in ferrocenylimidazolium salts has been
reported to show an additional redox peak due to increased elec-
tro-activity of the imidazolium centre in conjunction with the
halides that are in close proximity.[32,40] It is worth noting that
an increase in the chain length of the imidazolium alkyl substitu-
ent had a direct positive effect on the E1/2 values of the salts, due
to an increase in the positive inductive effect of relatively longer
alkyl chains.
In addition, data in Table 3 clearly showed the difference
between the two sets of salts, i.e. between those with methylene
(2–5 and 10) and those with phenyl (6–9 and 12) linkers. Gener-
ally, the methylene-linked salts exhibited higher formal electrode
potentials than corresponding phenyl-linked ones. This is due to
delocalization of electrons in the phenyl ring, which decreases
the electron-withdrawing effect of the imidazolium moiety that
is responsible for the general shift to more positive potentials
for all the salts as compared to ferrocene (see above). This ren-
ders the E1/2 value in compounds 6–9 and 12 less positive than
those of compounds 2–5 and 10. Batterjee et al. have reported
a similar effect in which the electrode potentials shifted to lower
positive values as a result of the introduction of a double bond in
their compounds.[45]
Furthermore, oxidation became more difficult, which led to a
shift to higher positive potentials as the size of anion increased.
This trend was observed for both methylene-linked (3 and 10)
and phenyl-linked (7 and 12) salts on changing counter-ions
from bromide to iodide. A possible explanation could be that,
as the anion size increases, there is a relatively larger charge dis-
tribution and the electron-withdrawing ability of the imidazolium
moiety increases. As a result, a lower electron density is experienced
wileyonlinelibrary.com/journal/aoc
Copyright © 2013 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2013, 27, 98–108

Ferrocenylimidazolium salts as transfer hydrogenation catalysts
Table 3. Cyclic voltammetry data for ferrocenylimidazolium salts, showing oxidation (anodic) and reduction (cathodic) peak potentials (Epa and Epc
respectively) and peak currents (ipa and ipc, respectively)
,
Compound
Epa (V)
Epc (V)
E1/2 (V)
E(Epc ꢀ Epa) (V)
ipc (mA)
ipa (mA)
ipc/ipa
Ferrocene
0.479
0.577
0.631
0.621
0.637
0.518
0.565
0.536
0.547
0.601
0.619
0.524
0.393
0.506
0.553
0.540
0.553
0.446
0.494
0.458
0.458
0.512
0.547
0.458
0.436
0.542
0.592
0.581
0.595
0.482
0.529
0.497
0.503
0.557
0.583
0.491
0.086
0.071
0.078
0.081
0.084
0.072
0.071
0.078
0.089
0.089
0.072
0.066
10.7
6.53
8.93
4.39
8.90
4.33
6.86
0.79
0.39
2.47
5.59
0.94
8.88
5.77
10.5
2.68
8.10
3.74
7.93
0.80
0.37
2.29
3.63
1.55
1.20
1.13
0.85
1.64
1.10
1.16
0.87
0.99
1.06
1.08
1.54
0.61
2
3
4
5
6
7
8
9
10
11
12
by the ferrocenyl centre, which makes oxidation relatively difficult.
By using H NMR chemical shifts, we have in the past probed
Table 4. Transfer hydrogenation of acetophenone catalysed by
compounds 2–12
1
Conversion (%)a
TOFb
TONc
the effect of increasing the size of the anion on the acidity of
imidazole protons and found that, as the anionic size decreased,
Compound
1
the chemical shifts of resonances in the H NMR spectra shifted
downfield.[43a] This confirmed that the size of the anion had an
influence on electrostatic interactions between the imidazolium
moiety and the ferrocenyl centre.
2
63
72
70
80
57
68
69
94
66
74
61
105
120
117
133
95
1260
1440
1400
1600
1140
1356
1380
1880
1320
1480
1220
Catalytic Transfer Hydrogenation
3
Application of the salts as potential catalysts was investigated in
transfer hydrogenation of both saturated and unsaturated
ketones. Catalytic transfer hydrogenation of ketones to alcohols
by the use of simple ionizable inorganic salts MOH (M = Li,
Na, K) is well established.[46,47] In this study we have used transfer
hydrogenation as a benchmark for testing structure–activity rela-
tionships. The study began with a set of exploratory experiments
that utilized acetophenone as substrate and 0.05 mol% of the
ferrocenylimidazolium salts as catalysts. The solvent, propan-2-ol,
was the H-transfer agent and the reaction was conducted at
82 ꢁC for 12 h in a basic KOH environment. The result of this study
is presented in Table 4. All compounds 2–12 showed high activi-
ties toward conversion of acetophenone to the desired alcohol.
Turnover numbers (TON) up to 1880 were observed, which are
comparable to some of the precious metal-catalyzed reactions
highlighted in this report.[3] Further exploratory work was con-
ducted in order to establish the roles of the various components
in the catalysis and to clarify the identity of the active species, since
it has been well established that inorganic salts are capable of
independent activation in this type of reactions. Hence a blank test
run was conducted in which only the catalyst salt was added and,
as expected, no catalytic activity was observed. The next test run
involved only KOH base monitored over a 48 h period and the re-
sult showed a maximum conversion of 20% to the desired alcohol
product after 12 h, with little or no improvement with time. Varia-
tion in the relative concentration of the base showed insignificant
change in the catalytic result. These observations support our
proposed mechanism[48] involving the formation of a K-NHC inter-
mediate (formed via coordination of potassium from KOH to the
reactive carbene moiety from deprotonation of the catalyst salt)
as the active species for this set of catalytic reactions. Moreover,
in order to clarify whether the K-NHC complex was the only
4
5
6
7
113
115
157
110
123
102
8
9
10
11
12
aConversion was determined by gas chromatographic analysis.
bTurnover frequency (TOF) = mol product/(mol catalyst ꢂ time),
determined after time t.
cTurnover number (TON) = mol product/mol catalyst.
active species or it was just catalyzed by the base, potassium
tert-butoxide (tBuOꢀK+) and an organic base, triethylamine, were
screened under similar conditions. In the absence of the catalyst
salt, 30% conversion of acetophenone to 1-phenylethanol was
observed for tBuOꢀK+ and no reaction was observed with triethy-
lamine. Addition of compound 11 improved the activity of the
tBuOꢀK+-initiated reaction to 65%, still with no sign of activity in
the triethylamine-catalyzed reaction, indicating that the organic
base was unsuitable for this reaction. Further evidence is provided
(see supporting information) from H and 13C NMR spectroscopic
1
monitoring of a solution of KOH and compound 11. 1H NMR results
showed deprotonation of the salt due to the disappearance of the
imidazolium proton signal at 8.59 ppm. Coordination of potassium
to the deprotonated reactive carbene species was confirmed by
the appearance of a downfield signal at 152.46 ppm which corre-
sponded to a carbene–K bond. This is similar to values recorded
Appl. Organometal. Chem. 2013, 27, 98–108
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

M. I. Ikhile et al.
for related metal–NHC bonds.[33,34,49] An important work by Arnold
et al.[50] has provided clear evidence on the isolation and structural
characterisation of K-NHC compounds.
comparative analysis of the catalytic and CV data revealed a
direct correlation between the observed high catalytic activities
and the tendency of the formal electrode potentials (E1/2
)
Having established optimum conditions for the reaction and
roles of major components, the study was extended to other
ketones, aimed at establishing its scope and limitations. For the
extended study, the more active compounds (5 and 9) were used
since they gave highest catalytic TON for their respective series. It
is interesting to note that 5 and 9 had potentials that shifted the
most to positive values in the CV study and are structurally similar
except for difference in linker, i.e. methylene vs. phenyl, respec-
tively. Table 5 presents results of this study. By introduction of
an electron-donating group to the ortho position of cyclohexa-
none, the recorded conversion was drastically reduced (Table 5,
entry 2). However, a better conversion was obtained when the
same electron-donating group was at the para position. This
may be attributed to steric interference that resulted in limited
access to the reactive site in the o-substituted substrate. Likewise,
the introduction of electron-withdrawing groups to acetophe-
none brought about significant drops in conversion (entries 8
and 9). In general, the cyclic aliphatic ketones gave better
conversions than their straight-chain counterparts, which may
be attributed to accessibility of the C═O bond in the relatively
rigid cyclic compounds as compared to the more sterically
hindered reactive site in the straight-chain ketones that led to
diminished conversions to the alcohols.
The study was further extended to the investigation of three
aliphatic a,b-unsaturated ketones. Contrary to expectation, two
of the substrates were not converted to the corresponding
alcohols; instead they were reduced to saturated ketones at
moderate to high yields (entries 10 and 11). In two of the cases
investigated, chemoselectivity of the catalyst towards hydroge-
nation of the C═C bond over the C═O bond was facilitated by
conjugated keto-ene resonance stabilization. Examples of this
selectivity abound; for example, Bond[51] earlier reported on the
relative selectivity of hydrogen addition to a conjugated C═C
bond over the C═O bond. Recently, Ide et al.[52] reported the
preferred hydrogenation of C═C bond over C═O bond in their
catalytic system, which was attributed to higher activation bar-
riers for the C═O bond. However, unconjugated 5-hexen-2-one
(entry 12) was selectively reduced to an unsaturated alcohol. This
can be attributed to the ethyl spacer between the C═C and C═O
bonds of the 5-hexen-2-one which disrupted the keto-ene conju-
gation and resonance stabilization.
towards higher positive values. Hence this set of results may
serve as a benchmark in future catalyst design for NHC com-
plexes derived from ferrocenylimidazolium salts.
Experimental
General Procedure
All manipulations involving air and moisture sensitive com-
pounds were performed through the use of standard Schlenk
techniques under an atmosphere of dry nitrogen. All solvents
were dried and purified by standard procedures prior to use.
Glassware was oven dried at 110 ꢁC. All NMR experiments were
conducted on a 400 MHz Bruker Ultrashield spectrometer and
samples were dissolved in deuterated chloroform. Infrared spec-
tra were recorded with a PerkinElmer Universal ATR Spectrum
100 FT-IR spectrometer. All low-resolution mass data were run
on a Thermo Finnigan linear ion trap mass spectrometer using
electrospray ionisation in positive mode. Accurate mass data
were obtained on a Thermo Electron DFS dual-focusing magnetic
sector instrument using electrospray ionization (ESI) in positive
mode. Polyethylenimine was used as reference solution. Cyclic
voltammograms were obtained with a METROHM 797 VA Com-
putrace at a scan rate of 0.05 V sꢀ1 in acetonitrile solution con-
taining 0.1 mol Lꢀ1 of tetrabutylammonium tetrafluoroborate
(n-Bu4NBF4) as the supporting electrolyte. An Ag/AgCl couple
was used as the reference electrode to determine the potential
with Pt working and auxiliary electrodes. All reagents and solvents
were purchased from Aldrich or Merck. Reagents were used as re-
ceived without further purification. Compounds 1,[43d] 2,[43a–c]
3,[43a] 6[39] and 7[53] were characterized by comparing their
analytical and spectroscopic data with those of related compounds
found in the literature.
Synthesis of 1-(ferrocenyl)methanol (1)
Ferrocenecarboxaldehyde (10 g, 0.047 mol) was dissolved in
mimimal anhydrous diethyl ether and transferred to a pressure-
equalizing dropping funnel. An ethereal solution of lithium
aluminium hydride, (LAH, 1.80 g, 0.047 mol) was prepared in a
three-necked round-bottomed flask. The aldehyde solution was
added to the ethereal solution dropwise while stirring under
nitrogen atmosphere. The solution was maintained at reflux for
2 h (45 ꢁC), while being monitored by thin-layer chromatography
(TLC). After completion it was allowed to cool, followed by the
addition of 60 ml diethyl ether. Excess LAH was destroyed by
dropwise addition of cold ethyl acetate, followed by addition of
ice-water slurry. The organic layer was separated, washed with
water (3 ꢂ 100 ml) and then dried over anhydrous magnesium
sulfate before concentration using a rotary evaporator, after
which it was dried under vacuum to give a yellow powder. Yield
7.2 g, 96%; m.p. 76–78 ꢁC; IR (attenuated total reflectance (ATR)
cmꢀ1) 3920, 3219, 2932, 1656, 1379, 1350, 1235, 1189, 987, 807,
498, 476; 1H NMR (400 MHz, CDCl3): d 4.29 (s, 2H, CH2), 4.25
(s, 2H, C5H4), 4.19 (s, 7H, C5H4, C5H5); 13C NMR (100 MHz, CDCl3):
d 88.58, 68.51, 68.48, 68.08 (ferrocenyl moiety), 60.97 (CH2OH);
m/z (ESI): 199 (M+ ꢀ OH, 92%), 216 (M+, 42%). Anal. calcd for
C11H12FeO 216.02538. (M+); found 216.02375 (M+).
Overall, compound 9 gave the best conversion and better
catalytic activity as compared to 5. The high activity exhibited
by 9 was due to its molecular structure, which also contributed
to its high melting point, shift to higher resonance of the imidaz-
1
ole proton in H NMR analysis and also a shift to more positive
potentials in the CV study.
Conclusions
In this study, two series of ferrocenylimidazolium salts comprising
12 compounds were synthesized and fully characterized. Applica-
tion of the salts as catalysts in the transfer hydrogenation of
ketones resulted in high efficiencies and turnover numbers as
high as 1880. Electrochemical studies by cyclic voltammetry
showed that all the salts exhibited positive electrode potentials
relative to ferrocene, which is due to the electron-withdrawing
effect of the imidazolium ion on the ferrocenyl moiety. A
wileyonlinelibrary.com/journal/aoc
Copyright © 2013 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2013, 27, 98–108

Ferrocenylimidazolium salts as transfer hydrogenation catalysts
Table 5. Transfer hydrogenation of saturated and unsaturated ketones catalysed by compounds 5 and 9
Entry
1
Ketones
% Conversiona (5)
% Conversiona (9)
TONb (5)
TONb (9)
76
83
1520
1660
2
3
4
50
62
9
20
93
6
1000
1240
180
400
1860
120
5
6
7
9
24
11
5
180
269
200
480
220
100
13
10
8
21
58
420
1160
9
31
68
89
62
57
85
620
1360
1780
1240
1140
1700
10c
11c
12
30
48
600
960
aConversion was determined by gas chromatographic analysis.
bTurnover number (TON) = mol product/mol catalyst.
cConverted to saturated ketone.
Appl. Organometal. Chem. 2013, 27, 98–108
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

M. I. Ikhile et al.
Synthesis of 1-(ferrocenylmethylene)imidazole (2)
(s, 1H, CH), 7.03 (m, 2H, NCH, NCH), 5.36 (s, 2H, NCH2), 4.41
(s, 2H, C5H4), 4.26 (m, 9H, C5H4, C5H5, NCH2), 1.87 (m, 2H, CH2),
1.36 (m, 2H, CH2), 0.95 (t, J 7.2, 3H, CH3); 13C NMR (100 MHz,
CDCl3), 136.42 (imd-C), 120.80 (NCH), 120.66 (NCH), 78.83, 77.33,
69.84, 69.56, 69.30 (ferrocenyl moiety), 49.98 (NCH2), 32.12
(CH2), 19.53 (CH2), 13.44 (CH3); m/z (ESI): 199 (M+ ꢀ imd ꢀ Br-,
6%), 323 (M+ ꢀ Br- , 100%), 324 (M+ ꢀ Br- + 1, 22%); Anal. calcd
for C18H23N2FeBr 323.12107 (M+ ꢀ Brꢀ); found 323.12104
(M+ ꢀ Br-).
Ferrocenylmethanol, 1 (0.5 g, 2.3 mmol) and an excess of N,N0-
carbonyldiimidazole (0.52 g, 3.2 mmol) were dissolved in a mini-
mum of amount DCM (10 ml). The mixture was brought to reflux
and the temperature maintained at 45 ꢁC. TLC was used to mon-
itor the reaction, which was completed after 1 h. The reaction was
quenched and allowed to cool to room temperature. Diethyl
ether (30 ml) was then added to the product mixture, which
was stirred for 5 min at room temperature. Solvent was removed
in vacuo and the resulting product was subjected to column
chromatography on silica gel. Diethyl ether was used to flash
down any unreacted starting material. Ethyl acetate (100%) was
later used to obtain the product, as yellow powder. Yield 0.32 g,
53%; m.p. 80 ꢁC; IR (ATR cmꢀ1) 3392, 3117, 1638, 1509, 1437,
1237, 1218, 1102, 1077, 809, 740, 658, 479; 1H NMR (400 MHz,
CDCl3): d 7.46 (s, 1H, CH), 7.00 (s, 1H, NCH), 6.89 (s, 1H, NCH),
4.84 (s, 2H, CH2), 4.16 (m, 9H, C5H4, C5H5); 13C NMR (100 MHz,
CDCl3): d 136.47 (imd-C), 129.00 (NCH), 118.60 (NCH), 82.51,
68.58, 68.31 (ferrocenyl moiety), 46.52 (CH2); m/z (ESI): 199
(M+ ꢀ imd [imd = imidazole], 6%), 267 (M+ , 100%), 268 (M+ + 1,
18%). Anal. calcd for C14H14N2Fe 267.05847 (M+); found
267.05748 (M+).
Synthesis of 1-(4-ferrocenylphenyl)imidazole (6)
A solution of sodium nitrite (0.088 g, 1.26mmol) in water (2ml) was
added dropwise to an ice-cold solution of 4-(1H-imidazol-1-yl)
aniline (0.2 g, 1.26 mmol) and aqueous tetrafluoroboric acid (50%,
0.4 ml, 6.2 mmol) in water (4ml), whereupon a yellowish precipi-
tate was observed. After complete addition of the sodium nitrite
solution, the mixture was stirred for another 5 min, followed by
the dropwise addition of a solution of ferrocene (0.24 g, 1.26 mmol)
prepared from a mixture of CH3CN and CH2Cl2 (1:2) 3 ml, resulting
in a greenish-brown suspension. The cooling bath was removed
and the mixture was further stirred for 1 h. Subsequently, H2O
(40 cm3) was added and the reaction mixture extracted with
CH2Cl2 (2ꢂ 14ml). The red organic layer was washed with dilute
Na2SO4; the solvent was removed in vacuo, resulting in a brown
residue, which was washed with n-hexane (3ꢂ 6 cm3) to remove
residual ferrocene. A brownish-orange powder was obtained. Yield
0.12 g, 30%; m.p. 161.5ꢁC; IR (ATR cmꢀ1) 3096, 1660, 1531, 1488,
1302, 1249, 1103, 1055, 1031, 1002, 885, 813, 725, 655, 531, 509,
Synthesis of 1-(ferrocenylmethylene)-3-methylimidazolium iodide (3)
In a two-necked flask, methyl iodide (0.3 ml, 8.1 mmol) was added
to ferrocenylmethylene imidazole (0.05 g, 0.19 mmol) and was
allowed to reflux gently at 50 ꢁC under an atmosphere of nitrogen
for 15 h. The mixture was then allowed to cool to room tempera-
ture, washed with anhydrous diethyl ether (5 ꢂ 3 ml) until diethyl
ether remained clear after washing. A yellow powder was
obtained. Yield 0.056 g, 72%; m.p. 145.5 ꢁC; IR (ATR cmꢀ1) 3173,
1566, 1331, 1243, 1331, 1174, 1150, 811, 754, 710, 619, 553, 480;
1H NMR (400 MHz, CDCl3): d 9.92 (s, 1H, CH), 7.28 (s, 1H, NCH),
7.26 (s, 1H, NCH), 5.34 (s, 2H, CH2), 4.47 (s, 2H, C5H4), 4.26
(m, 7H, C5H4, C5H5), 4.04 (s, 3H, CH3); 13C NMR (100 MHz, CDCl3):
d 135.96 (imd-C), 123.27 (NCH), 121.55 (NCH), 78.46, 77.30,
69.86, 69.70, 69.24 (ferrocenyl moiety), 50.05 (NCH3); m/z (ESI):
199 (M+ ꢀ imd ꢀ Iꢀ, 6%), 281 (M+ ꢀ Iꢀ , 100%), 282 (M+ ꢀ Iꢀ + 1,
18%). Anal. calcd for C15H17N2FeI 281.07412 (M+ ꢀ Iꢀ); found
281.07350 (M+ ꢀ Iꢀ).
1
482, 456; H NMR (400 MHz, CDCl3) d 7.85 (s, 1H, CH), 7.54 (d, J
8.4, 2H, C6H4), 7.29 (m, 3H, C6H4, NCH), 7.19 (s, 1H, NCH), 4.64 (t, J
1.6, 2H, C5H4), 4.35 (t, J 1.6, 2H, C5H4), 4.04 (s, 5H, C5H5), 13C NMR
(100MHz, CDCl3) d 139.24 (imd-C), 135.57 (Ar-C), 135.10 (Ar-C),
130.35 (Ar-CH), 127.24 (Ar-CH), 121.57 (NCH), 118.24 (NCH), 83.88,
69.73, 69.38, 66.95 (ferrocenyl moiety); m/z (ESI): 262 (M+ ꢀ imd,
17%), 329 (M+ , 100%), 330 (M+ + 1, 22%); Anal. calcd for
C19H16N2Fe 329.07412 (M+); found 329.07411 (M+).
General Procedure for the Preparation of Compounds 7–9
In a two-neck flask, 1-(4-ferrocenylphenyl) imidazole (1 molar
equiv.) was reacted with the respective alkylhalide (5 molar
equiv.) in CH2Cl2 (15 ml). The mixture was stirred for 41 h at room
temperature. After solvent removal, the mixture was treated with
Et2O (10 cm3), filtered, washed with Et2O (3 ꢂ 15 ml) and dried
in vacuo. Crystals suitable for X-ray analysis were obtained by
slow diffusion of hexane into a CH2Cl2 solution.
Synthesis of 1-(ferrocenylmethylene)-3-ethylimidazolium bromide (4)
Synthesis similar to 3 using ethyl bromide (0.6 ml, 8.1 mmol), to
give a yellow powder. Yield 0.06 g, 84%; m.p. 92.3 ꢁC; IR (ATR
cmꢀ1) 3432, 3386, 3138, 3065, 2065, 1626, 1565, 1558, 1463,
1412, 1347, 1319, 1236, 1157, 1027, 1007, 848, 807, 774, 708,
554, 506, 485, 447; 1H NMR (400 MHz, CDCl3): d 10.62 (s, 1H,
CH), 7.16 (s, 1H, NCH), 7.11 (s, 1H, NCH), 5.34 (s, 2H, CH2), 4.42
(s, 2H, C5H4), 4.36 (q, J 7.3, 2H, NCH2,), 4.23 (s, 7H, C5H4, C5H5),
1.56 (t, J 7.3, 3H, CH3); 13C NMR (100 MHz, CDCl3) d 136.94
(imd-C), 120.18 (NCH), 120.99 (NCH), 77.58, 70.09, 69.85, 69.54,
50.16 (ferrocenyl moiety), 45.56 (NCH2), 15.76 (CH3); m/z (ESI):
199 (M+ ꢀ imd ꢀ Brꢀ, 6%), 295 (M+ ꢀ Br- , 100%), 296 (M+ ꢀ Br- + 1,
18%). Anal. calcd for C16H19N2FeBr 295.08977 (M+ ꢀ Brꢀ); found
295.08962 (M+ ꢀ Brꢀ).
Synthesis of 1-(4-ferrocenylphenyl)-3-methylimidazolium iodide (7)
This compound was prepared from 1-(4-ferrocenylphenyl)imidazole
(0.1 g, 0.3 mmol) and methyl iodide (0.1 ml, 1.5 mmol). The crude
product was obtained as an orange powder. Yield 0.12 g, 86%;
m.p. 140.5 ꢁC; IR (ATR cmꢀ1) 3396, 3079, 1552, 1528, 1220, 1071,
1
820, 749, 620, 542, 504, 459; H NMR (400 MHz, CDCl3) d 10.65
(s, 1H, CH), 7.64 (m, 4H, C6H4), 7.52 (s, 1H, NCH), 7.40 (s, 1H,
NCH), 4.66 (t, J 1.7, 2H, C5H4), 4.39 (t, J 1.7, 2H, C5H4), 4.27 (s, 3H,
NCH3), 4.04 (s, 5H, C5H5); 13C NMR (100 MHz, CDCl3) d 143.15
(imd-C), 135.93 (Ar-C), 131.59 (Ar-C), 127.67 (Ar-CH), 123.99
(Ar-CH), 122.04 (NCH), 120.37 (NCH), 82.37, 69.88, 67.89, 66.79
(ferrocenyl moiety), 37.63 (NCH3); m/z (ESI): 262 (M+ ꢀ imd ꢀ Iꢀ,
7%), 343 (M+ ꢀ Iꢀ , 100%), 344 (M+ ꢀ Iꢀ + 1, 22%); Anal. calcd for
C20H19N2FeI 343.08977 (M+ ꢀ Iꢀ); found 343.08906 (M+ ꢀ Iꢀ).
Synthesis of 1-(ferrocenylmethylene)-3-butylimidazolium bromide (5)
Synthesis similar to 3 using butyl bromide (2.6 ml, 8.1 mmol). A
dark-brown powder was obtained. Yield 0.07 g, 92%; m.p.
85.5 ꢁC; IR (ATR cmꢀ1) 3571, 3393, 3168, 2965, 1563, 1448, 1155,
1
1105, 812, 656, 631, 553, 478; H NMR (400 MHz, CDCl3) d 10.78
wileyonlinelibrary.com/journal/aoc
Copyright © 2013 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2013, 27, 98–108

Ferrocenylimidazolium salts as transfer hydrogenation catalysts
1448, 1154, 820, 775, 555, 501, 478; 1H NMR (400 MHz, CDCl3)
d 8.50 (s, 1H, CH), 7.18 (s, 1H, NCH), 7.17 (s, 1H, NCH), 5.15 (s, 2H,
NCH2), 4.38 (t, J 1.6, 2H, C5H4), 4.23 (t, J 1.6, 2H, C5H4), 4.20 (s, 5H,
C5H5), 4.08 (t, J 5.7, 2H, NCH2), 1.82 (m, 2H, CH2), 1.32 (m, 2H,
CH2), 0.93 (t, J 7.2, 3H, CH3); 13C NMR (100 MHz, CDCl3) 134.49
(imd-C), 121.86 (NCH), 121.72 (NCH), 78.38, 77.35, 69.89, 69.54,
69.23 (ferrocenyl moiety), 49.97 (NCH2), 31.79 (CH2), 19.39 (CH2),
Synthesis of 1-(4-ferrocenylphenyl)-3-ethylimidazolium bromide (8)
This compound was prepared from 1-(4-ferrocenylphenyl)imidazole
(0.056 g, 0.17 mmol) and ethyl bromide (0.06 ml, 0.85 mmol). The
crude product was obtained as an orange powder. Yield 0.025 g,
64%; m.p. 184.5ꢁC; IR (ATR cmꢀ1) 3378, 3096, 1608, 1530, 1492,
1300, 1249, 1058, 885, 815, 741, 658, 531, 501, 455; 1H NMR
(400 MHz, CDCl3) d 11.25 (s, 1H, CH), 7.64 (m, 4H, C6H4), 7.53
(s, 1H, NCH), 7.19 (s, 1H, NCH), 4.66 (q, J 7.3, 2H, NCH2), 4.65 (t, J
1.6, 2H, C5H4), 4.38 (t, J 1.6, 2H, C5H4), 4.04 (s, 5H, C5H5), 1.67 (t, J
7.3, 3H, CH3); 13C NMR (100 MHz, CDCl3) d 142.92 (imd-C), 135.56
(Ar-C), 130.36 (Ar-C), 127.70 (Ar-CH), 121.50 (Ar-CH), 121.55 (NCH),
120.05 (NCH), 83.87, 69.72, 66.82, 66.58 (ferrocenyl moiety), 45.84
(NCH2), 15.73 (CH3); m/z (ESI): 262 (M+ ꢀ imd ꢀ Brꢀ, 8%), 329 (M+ ꢀ
C2H5 ꢀ Br- , 100%), 357 (M+ ꢀ Br-, 80%), 358 (M+ ꢀ Br- + 1, 18%);
Anal. calcd for C21H21N2FeBr 357.10542 (M+ ꢀ Brꢀ); found
357.10388 (M+ ꢀ Brꢀ).
13.30 (CH3); m/z (ESI): 199 (M+ ꢀ imd ꢀ PF6ꢀ, 7%), 323 (M+ ꢀ PFꢀ6
,
100%), 324 (M+ ꢀ PF6ꢀ + 1, 22%). Anal. Calcd for C18H23N2FeBr
323.12107 (M+ ꢀ PF6ꢀ); found 323.12105 (M+ ꢀ PFꢀ6 ).
Synthesis of 1-(4-ferrocenylphenyl)-3-methylimidazolium bromide (12)
In a two-necked flask was added sodium bromide (0.019 g,
0.18 mmol) to an acetone solution of 1-(3-methyl-(4-ferrocenyl-
phenyl)imidazolium iodide, 7 (0.05 g, 0.11 mmol). The mixture
was stirred at room temperature for 24 h. The resultant mixture
was then filtered through a plug of celite and concentrated
in vacuo to give a yellowish-brown powder. Yield 0.035 g, 87%;
m.p. 179.5 ꢁC; IR (ATR cmꢀ1) 3452, 3397, 3079, 1706, 1553, 1423,
1220, 1003, 821, 749, 621, 519. 1H NMR (400 MHz, CDCl3) d
10.64 (s, 1H, CH), 7.64 (m, 4H, C6H4), 7.54 (s, 1H, NCH), 7.45
(s, 1H, NCH), 4.65 (t, J 1.6, 2H, C5H4), 4.38 (t, J 1.6, 2H, C5H4), 4.26
(s, 3H, NCH3), 4.04 (s, 5H, C5H5); 13C NMR (100 MHz, CDCl3) d
143.26 (imd-C), 136.27 (Ar-C), 131.73 (Ar-C), 127.81 (Ar-CH),
124.07 (Ar-CH), 122.15 (NCH), 120.45 (NCH), 82.54, 70.07, 70.01,
66.95 (ferrocenyl moiety), 37.70 (NCH3); m/z (ESI): 262 (M+ ꢀ imd
Brꢀ, 8%), 343 (M+ ꢀ Br-, 100%), 344 (M+ ꢀ Br- + 1, 22%). Anal.
calcd for C20H19N2FeBr 343.08977 (M+ ꢀ Brꢀ); found 343.08885
(M+ ꢀ Brꢀ).
Synthesis of 1-(4-ferrocenylphenyl)-3-butylimidazolium bromide (9)
This compound was prepared from 1-(4-ferrocenylphenyl)imidazole
(0.024 g, 0.0732 mmol) and butyl bromide (0.04 ml, 0.366 mmol).
The crude product was obtained as a brownish-orange powder.
Yield 0.03 g, 91%; m.p. 158.5 ꢁC; IR (ATR cmꢀ1) 3076, 1527, 1304,
1057, 962, 887, 820, 885, 729, 656, 536, 507, 458; 1H NMR
(400 MHz, CDCl3) d 11.24 (s, 1H, CH), 7.64 (d, J 8.4, 2H, C6H4), 7.53
(m, 3H, C6H4, NCH), 7.34 (s, 1H, NCH), 4.66 (m, 4H, C5H4, NCH2),
4.38 (t, J 1.7, 2H, C5H4), 4.04 (s, 5H, C5H5), 1.98 (m, 2H, CH2), 1.23
(m, 2H, CH2), 1.01 (t, J 7.4, 3H, CH3); 13C NMR (100 MHz, CDCl3) d
142.84 (imd-C), 135.84 (Ar-C), 131.70 (Ar-C), 127.68 (Ar-CH), 127.19
(Ar-CH), 121.65 (NCH), 119.76 (NCH), 82.46, 69.84, 66.77, 65.53
(ferrocenyl moiety), 42.71 (NCH2), 32.27 (CH2), 19.54 (CH2), 13.51
(CH3); m/z (ESI): 262 (M+ ꢀ imd ꢀ Brꢀ, 8%), 329 (M+ ꢀ C4H9 ꢀ Brꢀ
,
General Procedure for Transfer Hydrogenation
100%), 385 (M+ ꢀ Brꢀ, 78%), 386 (M+ ꢀ Br- + 1, 22%); Anal. calcd
for C23H25N2FeBr 385.13672 (M+ ꢀ Brꢀ); found 385.13641
(M+ ꢀ Brꢀ).
Ferrocenylimidazolium salts as catalysts, 0.05 mol%, ketones
(2.1mmol) and KOH (0.112 g, 10 ml, 2 mmol, 0.2M in propan-2-ol)
were introduced into a round-bottomed flask fitted with a con-
denser and refluxed at 82ꢁC. The transfer hydrogenation reaction
was monitored with an Agilent capillary gas chromatograph,
model 6820, fitted with a DB wax polyethylene column (0.25 mm
diameter, 30m length) and a flame ionization detector. Nitrogen
gas was used as carrier gas at a flow rate of 2 mL minꢀ1. The oven
temperature for the aromatic (except 4-fluoroacetophenone) and
the cyclic (except cyclobutanone) ketones was 70ꢁC, while for
the remaining ketones the oven temperature was 50ꢁC. Samples
(0.1ml) were injected at 260 ꢁC front inlet temperature for the
aromatic (except 4-fluoroacetophenone) and the cyclic ketones
(except cyclobutanone). For the remaining ketones the front inlet
temperature was 180 ꢁC. The identity of the alcohols was assessed
by comparing their retention times with commercially available
(Aldrich Chemical Co.) pure samples. Conversions obtained were
calculated from the integration values of the gas chromatographic
peaks, which were related to residual unreacted ketone.
Synthesis of 1-(ferrocenylmethylene)-3-methylimidazolium bromide (10)
In a two-necked flask was added sodium bromide (0.043 g,
0.42 mmol) to an acetone solution of 1-(ferrocenylmethylene)-3-
methylimidazolium iodide (3) (0.1 g, 0.25 mmol). The mixture
was stirred at room temperature for 24 h. The reaction mixture
was then filtered through a plug of celite and concentrated in
vacuo to give an orange oil. Yield 0.073 g, 81%; IR (ATR cmꢀ1
)
3438, 3080, 1614, 1572, 1557, 1454, 1425, 1221, 1153, 1105,
1
1037, 1003, 822, 749, 673, 621, 543, 503, 462; H NMR (400 MHz,
CDCl3): d 9.97 (s, 1H, CH), 7.17 (s, 1H, NCH), 7.15 (s, 1H, NCH),
5.33 (s, 2H, CH2), 4.43 (s, 2H, C5H4), 4.24 (s, 7H, C5H4, C5H5), 4.02
(s, 3H, CH3); 13C NMR (100 MHz, CDCl3): d 136.64 (imd-C), 122.83
(NCH), 121.18 (NCH), 78.46, 77.34, 69.87, 69.63, 69.25 (ferrocenyl
moiety), 50.15 (CH3); m/z (ESI): 199 (M+ ꢀ imd ꢀ Brꢀ, 6%), 281
(M+ ꢀ Brꢀ, 100%), 282 (M+ ꢀ Br- + 1, 22%); Anal. calcd for
C15H17N2FeBr 281.07412 (M+ ꢀ Brꢀ); found 281.07381 (M+ ꢀ Brꢀ).
X-ray crystal determination
Synthesis of 1-(ferrocenylmethylene)-3-butylimidazolium hexafluorophosphate
(11)
Intensity data were collected on a Bruker APEX II CCD area detec-
tor diffractometer with graphite monochromated Mo Ka radiation
(50 kV, 30 mA) using APEX 2[54] data collection software. The col-
lection method involved o-scans of width 0.5 ꢁ and 512 ꢂ 512-bit
data frames. Data reduction was carried out using the program
SAINT+[54] and face-indexed absorption corrections were made
using XPREP.[54] The crystal structure was solved by direct meth-
ods using SHELXTL. Non-hydrogen atoms were first refined
isotropically, followed by anisotropic refinement by full matrix
In a two-necked flask, sodium hexafluorophosphate (0.035g,
0.21 mmol) was added to an acetone solution of (ferrocenylmethy-
lene)-3-butylimidazolium bromide (5) (0.05 g, 0.13 mmol). The mix-
ture was stirred under a nitrogen atmosphere for 24h at room
temperature. The reaction mixture was then filtered through a
plug of celite, concentrated in vacuo to afford an orange-brown
paste. Yield 0.1 g, 96%. IR (ATR cmꢀ1): 3639, 3167, 2934, 1562,
Appl. Organometal. Chem. 2013, 27, 98–108
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

M. I. Ikhile et al.
least-squares calculations based on F2 using SHELXTL. Hydrogen
atoms were first located in the difference map, then positioned
geometrically and allowed to ride on their respective parent
atoms. Diagrams and publication material were generated using
SHELXTL, PLATON[55] and ORTEP-3.[56]
[21] N. J. Forrow, N. C. Foulds, J. E. Frew, J. T. Law, Bioconjugate Chem.
2004, 15, 134.
[22] C. L. Ferreira, C. B. Ewart, C. A. Barta, S. Little, V. Yardley, C. Martins,
E. Polishchuk, P. J. Smith, J. R. Moss, M. Merkel, M. J. Adam, C. Orvig,
Inorg. Chem. 2006, 45, 8414.
[23] C. E. Immoos, S. J. Lee, M. W. Grinstaff, J. Am. Chem. Soc. 2004, 126,
10814.
[24] J. Liu, S. Tian, L. Tiefenauer, P. E. Nilesen, W. Knoll, Anal. Chem. 2005,
77, 2756.
Supporting Information
[25] W. Tarraga, P. Molina, J. L. Lopez, M. D. Velasco, D. Bautisla,
Supporting information may be found in the online version of
this article. Crystallographic data in cif format for the structural
analysis has been deposited with the Cambridge Crystallographic
Data Centre, with numbers CCDC 876270 and 844511 for com-
pounds 4 and 11, respectively. Copies of this information may
be obtained free of charge from: The Director, CCDC, 12 Union
Road, Cambridge, CB2 1EZ, UK. Fax: +44(1223)336-033, e-mail:
deposit@ccdc.cam.ac.uk, or http:// www.ccdc.cam.ac.uk.
P. G. Jones, Organometallics 2002, 21, 2055.
[26] F. Tompatanaget, T. Tuntulani, O. Chailapakul, Org. Lett. 2003, 5,
1539.
[27] I. Willner, E. Katz, Angew. Chem. Int. Ed. 2000, 39, 1180.
[28] C. J. Yu, H. Wang, Y. Yowanta, J. C. Kim, L. H. Donilon, C. Tao,
M. Strong, Y. Chong, J. Org. Chem. 2001, 66, 2937.
[29] O. B. Sutcliffe, M. R. Bryce, A. S. Batsanov, J. Organomet. Chem. 2002,
656, 211.
[30] J. -L. Thomas, J. Howarth, A. M. Kennedy, Molecules 2002, 7, 861.
[31] Y. Bai, B.-G. Zhang, C.-Y. Duan, D.-B. Dang, Q.-J. Meng, New J. Chem.
2006, 30, 266.
[32] B. Bildstein, M. Malaun, H. Kopacka, K.–. H. Ongania, K. Wurst,
Acknowledgements
J. Organomet. Chem. 1999, 572, 177.
[33] L. Mercs, G. Labat, A. Neels, A. Ehlers, M. Albrecht, Organometallics
2006, 25, 5648.
[34] P. Buchgraber, L. Toupet, V. Guerchais, Organometallics 2003, 22,
5144.
[35] J. Thomas, J. Howarth, K. Hanlon, D. McGuirk, Tetrahedron Lett. 2009,
41, 413.
The authors wish to thank the University of KwaZulu-Natal and
the National Research Foundation (NRF) for financial support.
Dr Manuel Fernandes is also thanked for crystallographic data
collection and solution.
[36] W. Xuchum, W. Jieying, Z. Hongping, T. Yupeng, L. Lin, Y. Jiaxiang,
J. Baokang, Z. Shengyi, Sci. China Ser B Chem. 2009, 52, 930.
[37] Y. Miura, F. Shimizu, T. Mochida, Inorg. Chem. 2001, 49, 10032.
[38] J. Howarth, K. Hanlon, Tetrahedron Lett. 2001, 42, 751.
[39] U. E. I. Horvath, G. Bentivoglio, M. Hummel, H. Schottenberger,
K. Wurst, M. J. Nell, C. E. J. van Rensburg, S. Cronje, H. G. Raubenheimer,
New J. Chem. 2008, 32, 533.
References
[1] R. Di Santo, A. Tafi, R. Costi, M. Botta, M. Artico, F. Corelli, M. Forte,
F. Caporuscio, L. Angiolella, A. T. Palamara, J. Med. Chem. 2005, 48,
5140.
[2] J. Hongbing, Y. Huang, Y. Qian, T. Wang, M. Zhang, Chinese J. Chem.
Eng. 2006, 14, 118.
[40] B. Bildstein, M. Malaun, H. Kopacka, K. Ongania, K. Wurst,
J. Organomet. Chem. 1998, 552, 45.
[3] D. Gnanamgari, E. L. O. Sauer, N. D. Schley, C. Butler, C. D. Incarvito,
R. H. Crabtree, Organometallics 2009, 28, 321.
[4] C. Exner, A. Pfaltz, M. Studer, H. Blaser, Adv. Synth. Catal. 2003, 345,
1253.
[5] A. Aupoix, C. Bournaud, G. Vo-Thanh, Eur. J. Org. Chem. 2011, 2772.
[6] Y. Zhu, H. Qian, B. A. Drake, R. Jin, Angew. Chem. Int. Ed. 2010, 49,
1295.
[7] S. Enthaler, R. Jackstell, B. Hagemann, K. Junge, G. Erre, M. Beller,
J. Organomet. Chem. 2006, 691, 4652.
[8] K. S. Coleman, S. Turberville, S. I. Pascu, M. L. H. Green, J. Organomet.
Chem. 2005, 690, 653.
[9] B. Bildstein, J. Organomet. Chem. 2001, 28, 617.
[10] R. C. J. Atkinson, V. C. Gibson, N. J. Long, Chem. Soc. Rev. 2004, 33,
313.
[41] H. Seo, B. Y. Kim, J. H. Lee, H. Park, S. U. Son, Y. K. Chung, Organome-
tallics 2003, 22, 4783.
[42] R. Jiang, X. Sun, W. He, H. Chen, Y. Kuang, Appl. Organomet. Chem.
2009, 23, 179.
[43] a) V. O. Nyamori, M. Gumede, M. D. Bala, J. Organomet. Chem. 2010,
695, 1126; b) V. O. Nyamori, M. D. Bala, Acta Cryst. 2008, E64, m1451;
c) A. A. Simenel, E. A. Morozova, Y. V. Kuzmenko, L. V. Snegur,
J. Organomet. Chem. 2003, 665, 13; d) W. L. Davis, R. F. Shago,
E. G. H. Langer, J. S. Swarts, Polyhedron 2005, 24, 1611.
[44] P. Wasserscheid, T. Welton, Ionic Liquids in Synthesis, Wiley-VCH,
Weinheim, 2003.
[45] S. M. Batterjee, M. I. Marzouk, M. E. Aazab, M. A. El-Hashash, Appl.
Organomet. Chem. 2003, 17, 291.
[11] X. Wang, J. Wu, H. Zhou, Y. Tian, L. Li, J. Yang, B. Jin, S. Zhang, Sci.
China Ser B Chem. 2009, 52, 930.
[46] A. Ouali, J.-P. Majoral, A.-M. Caminade, M. Taillefer, ChemCatChem
2009, 1, 504.
[12] S. Ling, Z. Xin, H. Yan, J. Zhong, L. Jianbing, F. Jian, Chinese Chem.
[47] V. Polshettiwar, R. S. Varma, Green Chem. 2009, 11, 1313.
[48] M. I. Ikhile, V. O. Nyamori, M. D. Bala, Tetrahedron Lett. 2012, 53, 4925.
[49] A. Labande, J. C. Daran, N. J. Long, A. J. P. White, R. Poli, New J. Chem.
2011, 35, 2162.
[50] P. L. Arnold, M. Rodden, C. Wilson, Chem. Commun. 2005, 1743.
[51] G. C. Bond, Catalysis of Metals, Academic Press, London, 1962.
[52] M. S. Ide, B. Hao, M. Neurock, R. J. Davis, ACS Catal. 2012, 2, 671.
[53] D. Onyancha, C. McCleland, T. Gerber, E. Hosten, P. Mayer, Acta Cryst.
2010, E66, m49.
Lett. 2006, 36, 325.
[13] D. R. Van Staveren, N. Metzler-Nolte, Chem. Rev. 2004, 104, 5931.
[14] A. Albinati, P. S. Pregosin, K. Wick, Organometallics 1996, 15, 2419.
[15] L. X. Dai, S. L. You, W. P. Deng, W. L. Hou, Acc. Chem. Res. 2003, 36, 659.
[16] A. A. Nazarov, C. G. Hartinger, V. B. Arion, G. Giester, B. K. Keppler,
Tetrahedron 2002, 58, 8489.
[17] J. G. Seitzberg, C. Dissing, I. Sotofte, P. O. Norrby, M. Johannsen,
J. Org. Chem. 2005, 70, 8332.
[18] A. Togni, T. Hayashi, Ferrocenes: Homogeneous Catalysis, Organic
Synthesis, Materials Science, VCH, Weinheim, 1995.
[19] A. Togni, Angew. Chem. Int. Ed Engl. 1996, 35, 1475.
[20] S. J. Choi, B. G. Choi, S. M. Park, Anal. Chem. 2002, 74, 1998.
[54] Bruker AXS Inc., APEX2, SAINT and XPREP, Bruker, Madison, WI, USA,
2005.
[55] A. L. Spek, Acta Cryst. 2009, D65, 148.
[56] L. J. Farrugia, J. Appl. Cryst. 1997, 30, 565.
wileyonlinelibrary.com/journal/aoc
Copyright © 2013 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2013, 27, 98–108