Solvent and substituent effects in hydrogenation of aromatic ketones over Ru/polymer catalyst under very mild conditions

Article abstract of DOI:10.1016/j.mcat.2019.04.003

The paper reports on the solvent and the substituent effects in hydrogenation of aromatic ketones (acetophenone and its derivatives) in the presence of Ru catalyst supported on functionalized gel-type methacrylate-styrene resin under very mild conditions

Full text of DOI:10.1016/j.mcat.2019.04.003

Molecular Catalysis 470 (2019) 145–151
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Solvent and substituent effects in hydrogenation of aromatic ketones over
Ru/polymer catalyst under very mild conditions
a,⁎
a
a
a
a
D. Duraczyńska , E.M. Serwicka , A. Drelinkiewicz , R.P. Socha , M. Zimowska ,
b
c
L. Lityńska-Dobrzyńska , A. Bukowska
Jerzy Haber Institute of Catalysis and Surface Chemistry Polish Academy of Sciences, Niezapominajek 8, 30-239, Krakow, Poland
Institute of Metallurgy and Materials Science of Polish Academy of Sciences, Reymonta 25, 30-059, Krakow, Poland
Rzeszow University of Technology, al. Powstancow Warszawy 12, 35-959, Rzeszow, Poland
a
b
c
A R T I C L E I N F O
A B S T R A C T
Keywords:
The paper reports on the solvent and the substituent effects in hydrogenation of aromatic ketones (acetophenone
Hydrogenation of aromatic ketones
Polymer-supported Ru catalyst
Water effect
Solvent effects
Substituent effects
and its derivatives) in the presence of Ru catalyst supported on functionalized gel-type methacrylate-styrene
resin under very mild conditions (1 bar H , 40 °C). The investigated solvents included methanol, 2-propanol,
2
tetrahydrofuran, toluene, isooctane, and cyclohexane, with and without addition of water. The best catalytic
activity was obtained in biphasic solvent system consisting of isooctane/water (1:1). In methanol and 2-propanol
the beneficial role of water was observed. The effects are discussed in terms of solvent protic/aprotic and polar/
apolar properties. The impact of substituents depended on their electron-donating/withdrawing character, and
on the position on the aromatic ring of acetophenone. The presence of electron donating substituents tended to
increase the reaction rate, and the electron withdrawing group slowed down the process. Steric effect was
3
observed for methylacetophenone isomers with CH group at ortho, meta and para positions.
1. Introduction
3
such as −OCH on aromatic ring tends to increase the initial reaction
rate, while the presence of –Cl– substituent (electron withdrawing
group) slows down the process as compared to the hydrogenation of the
parent acetophenone.
The authors also observed a decrease of the initial reaction rate
3
when the −CH group was located at ortho position of aromatic ring,
Catalytic hydrogenation of aromatic ketones constitutes an im-
portant aspect of organic synthesis, enabling the preparation of a great
number of secondary alcohols widely used in pharmaceutical, cosmetic,
food and chemical industries as solvents, intermediates, flavor ad-
ditivities, fragrances, etc. The hydrogenation of acetophenone and its
substituted derivatives are typical examples of this type of reactions.
Since formation of side products is possible during the multistep reac-
tions (for example via hydrogenation of phenyl ring and hydrogenolysis
of produced alcohols) (Scheme 1), the development of highly active and
selective catalysts working preferably under mild conditions is still a
challenging task that requires thorough consideration of various fac-
tors. In addition, the heterogeneous character of a catalyst combined
with its stability would be a very desirable feature of a catalyst of
choice.
There are not many studies directed at the hydrogenation of acet-
ophenone and its derivatives using supported monometallic (Ni, Pt, Pd,
Cu, Ir, Rh) and bimetallic (Ni-Pt, Pt-Sn) catalysts [1–9]. Among them,
only few addressed the effect of substituents, their positions on aro-
matic ring or their electron-donating/withdrawing character. Thus,
Vetere et al. [7] showed that the presence of electron donating group
most likely due to the steric effect. Similar trend was observed by F.
Zaccheria et al. [4]. On the other hand, Hess et al. [6] noticed an op-
posite tendency, i.e., the relative rates were higher in the presence of
electron-withdrawing substituents and lower with electron-donating
groups. Moreover, the authors pointed out to the influence of chiral
modifier on hydrogenation rates. However, all these studies were per-
formed under different conditions, with temperature ranging from
ambient to 90 °C, hydrogen pressure from 1 to 10 bar, using various
solvents and catalysts. Therefore, it is difficult to determine which
factors have a decisive influence on the observed reaction rates and
trends. To shed some light on this phenomenon we have decided to
investigate first the impact of solvent on the catalytic activity and se-
lectivity of the studied catalyst, 2 wt.%Ru supported on functionalized
gel-type methacrylate-styrene resin (referred to as FCN, where FCN
means “functionalized”), and then the impact of electron-donating/
withdrawing character of substituents and their positions on aromatic
⁎
Corresponding author.
E-mail address: ncduracz@cyf-kr.edu.pl (D. Duraczyńska).
https://doi.org/10.1016/j.mcat.2019.04.003
Received 21 December 2018; Received in revised form 29 March 2019; Accepted 2 April 2019
2468-8231/ © 2019 Elsevier B.V. All rights reserved.

D. Duraczyńska, et al.
Molecular Catalysis 470 (2019) 145–151
Scheme 1. Hydrogenation of acetophenone and its derivatives.
ring of acetophenone. Our previous studies [10,11] revealed that ru-
thenium catalysts (1–4 wt% Ru) supported on gel-type methacrylate-
styrene resin functionalized with C]O, −OH, –NH, and –NH groups
2
structurally diverse array of ketones related to acetophenone, differing
by the position of methyl substituent on the aromatic ring, the electron-
donating/withdrawing character of substituents, and the length of alkyl
chain connected to carbonyl group, carried out in the presence of 2 wt.
showed a very promising catalytic behavior in hydrogenation of acet-
ophenone. Specific properties of this polymer, e.g. hydrophobic/hy-
drophilic character, the ability to coordinate to Ru via functional
groups and the swelling ability, make this polymer very attractive as a
carrier. FCN resin contains both apolar (styrene) and polar (O- and N-
containing groups) constituents. The ability to swell is a very important
feature since it effectively separates polymeric chains thus greatly en-
hancing the accessibility of active centers inside the polymer matrix
%Ru/FCN catalyst under very mild conditions (1 bar H
. Experimental
.1. Materials and methods
This catalyst was synthesized according to the method described by
2
, 40 °C).
2
2
[
12]. This support was also successfully used for Pd catalyst in hydro-
Duraczynska et al. [10,11]. Briefly, the appropriate amount of func-
genation of 2-butyne-1,4-diol and phenylacetylene [13]. In addition,
ruthenium catalysts are known to be active and selective in hydro-
genation of C]O group containing compounds. For example, the hy-
drogenation of levulinic acid and alkyl-levulinates have been thor-
oughly investigated in the presence of a range of Ru supported catalysts
tionalized gel-type resin FCN [21] was allow to swell for 30 min in THF.
Then the aqueous solution of RuCl ∙x H O (x < 1) (Aldrich) corre-
3 2
sponding to 2 wt% Ru was added to the resin suspension. The whole
was gently shaken until the complete discoloration of initially black
solution. The resulting black polymer beads were separated by filtra-
tion, washed with acetone and dried in air. Subsequently the solid was
[
14,15]. Ru/Y zeolite catalyst was successfully applied in the hydro-
genation of cinnamaldehyde [16]. Ruthenium catalysts supported on
metal oxides exhibited high activity and selectivity in the stereo-
selective hydrogenation of paracetamol [17].
It is well known that the solvent nature may influence both the
activity and the selectivity of the catalyst, and the solvent effect must be
considered when optimizing reaction conditions. Bertero et al. [18], as
well as Trasarti et al. [19], studied this aspect for the liquid-phase
reduced with excess of methanol solution of NaBH
until the wash liquid gave no reaction with AgNO
in air at ambient conditions.
4
, washed with water
3
solution, and dried
The infrared spectra were registered in the middle infrared (MIR)
−1
4
000–400 cm region with the use of Nicolet 380 FTIR spectrometer.
The sample (1 mg) was mixed with 200 mg of spectroscopically pure
KBr and pressed into disks before the spectra were recorded. Spectra
hydrogenation of acetophenone over SiO
catalysts in the presence of numerous solvents of different nature
protic, aprotic polar, and apolar solvents). They found that the highest
2
supported Ni, Co and Cu
−1
were obtained by co-addition of 64 scans at a resolution of 4 cm and
processed using the Thermo Scientific OMNIC™ software package.
X-ray photoelectron spectroscopy (XPS) spectra were obtained with
a hemispherical analyzer (SES R4000, Gammadata Scienta, pass energy
(
catalytic activities were achieved when C2-C3 alcohols were used due
to the solvent interactions with metal surface and the appropriate po-
larization/activation of C]O bond of acetophenone. Similar results
were observed by Cheng et al. [20] regarding hydrogenation of acet-
1
00 eV). The Al Kα X-ray source (1486.6 eV) was applied to generate
core excitation. The system was calibrated according to ISO
5,472:2010. The analysis area of the sample (powder pressed into
1
ophenone in the presence of phosphine modified Ru/γ-Al
2
O
3
catalyst.
2
indium foil) was about 3 mm . The electron binding energy scale (BE)
was calibrated for maximum of C 1s core excitation at 285.0 eV. The
spectra were fitted with the Casa XPS2.3.12 software, using Gaussian/
Lorentzian functional (70:30) and Shirley-type background.
The highest conversion was detected in 2-propanol.
In order to establish the most efficient solvent system enabling to
achieve high activity and selectivity to desired phenyl alcohols we
tested several solvents of different chemical nature. Thus, methanol
Transmission electron microscopic (TEM) studies were performed
(
MeOH) and 2-propanol (IPA) belong to polar group of solvents, to-
with FEI Tecnai G2 transmission electron microscope at 200 kV
luene, cyclohexane, and isooctane (IO) are apolar, while tetra-
hydrofuran (THF) is considered polar but its properties are close to the
polar/apolar borderline. We observed a series of non-trivial effects,
which are reported in the present study. The phenomena are discussed
equipped with EDX and HAADF/STEM detectors.
2.2. Catalytic tests
by considering the solvent polarity, its protic/aprotic character, H
solubility, the solvent-catalyst and the reactants-catalyst interactions, as
well as the role of water.
2
Hydrogenation experiments were carried out in an agitated batch
glass reactor (PARR 5100) at constant pressure of hydrogen (1 bar) and
temperature 40 °C. 25 mL of the appropriate solvent or water (in the
case of biphasic condition) and 0.2 g of the catalyst were placed in the
Moreover, in this paper we report the results of hydrogenation of
146

D. Duraczyńska, et al.
Molecular Catalysis 470 (2019) 145–151
−5
reactor. Then 25 mL of the reaction mixture containing 300 × 10
Table 1
mole of ketone dissolved in the solvent was added. The whole was
vigorously stirred (800 rpm) and the reaction time was set for 6 h. A gas
chromatograph (Clarus 500, Perkin Elmer) equipped with Elite-5MS
column was used for analyzing the composition of reaction mixture.
After catalytic test the samples from both phases (aqueous and organic)
were taken and the contents of the individual components were de-
termined by comparison with calibration curves. From the data of GC
analysis ketone conversions and selectivities to the desired products
were calculated. Catalytic results in the monophasic systems were not
affected by an increase of the agitation speed, which pointed to the
absence of external mass transport limitations. In biphasic systems, the
selected rpm ensured both complete and uniform dispersion of one li-
quid in the other.
In biphasic solvent system the distribution of all the reagents in both
solvent phases was determined by GC method. The known amounts of
individual reagents were added into 20 mL biphasic solvent system (1:1
by volume). The whole lot was vigorously mixed with a mechanical
stirrer for 45 min. The samples of liquids from both phases were
withdrawn immediately after the completion of shaking, and after 5,
Binding energies (BE) and contributions (%) of individual Ru, N and C peak
components.
N 1s
Ru 3p3/2
BE (eV)
C 1s
BE (eV)
Share (%)
Share (%)
BE (eV)
Share (%)
3
3
97.5
99.8
38.6
61.4
458.1
461.6
464.1
10.4
68.3
21.3
283.0
285.1
286.9
26.8
51.4
12.8
9.0
2
88.3
[
10], and O–Ru, and N–Ru [22] vibrations, respectively. Those spectral
differences suggest bonding of Ru to polymer matrix via coordination to
N– or/and O–containing functional groups.
High resolution XP spectroscopy was used to characterize surface
oxidation states of the catalyst. The XPS spectrum of 2 wt.%Ru/FCN
showed three ruthenium states characterized by 3 p3/2 energy of 458.1,
4
3
61.6 and 464.1 eV, respectively (Table 1). The one with the lowest Ru
3/2 energy of 458.1 eV may be attributed to carbon or oxygen bound
p
metallic ruthenium. The binding energy of the second, the most abun-
dant, ruthenium state with Ru 3p3/2 energy of 461.6 eV corresponds to
metallic ruthenium [23–25]. The highest energy state with Ru 3p3/2
energy of 464.1 eV suggests the presence of oxidized Ru species such as
15, 30, and 60 min. The concentrations of individual components in
both organic and aqueous phases were determined based on calibration
curves. The measurement aimed at determination of time required for
the biphasic system to reach equilibrium. This served to minimize the
error related to quantification of substrate/product concentrations. The
experiments have shown that the equilibrium concentrations were es-
tablished just after stopping the mixing. It has been checked that dis-
tribution coefficients measured at the end of each activity test were
equal to the values estimated through equilibria determination.
4+
3+
Ru
bonds is supported by the presence of carbon state with C 1s energy of
83.0 eV. The other energy states of C 1s component correspond to the
and/or Ru
[23,24]. The formation of C–Ru and/or C–O–Ru
2
presence of CeC and C]O bonds as well as carbon bound to oxygen- or
oxygen and nitrogen-containing organic groups. In the N 1s XPS spec-
trum two nitrogen states appear. The one with binding energy of 397.5
eV is attributed to the formation of N–C–Ru and/or N–Ru bonds. The
second nitrogen state (399.8 eV) corresponds to the presence of NeC
bonds (in an amine). The results obtained by the XPS technique are
consistent with the FT-IR spectra indicating the participation of func-
tional groups of polymer matrix in the bonding of Ru.
High resolution transmission electron micrograph of 2 wt.% Ru/
FCN reveals particles of morphology close to spherical, uniformly dis-
tributed on the polymer support (Fig. 2). The size of ruthenium particles
is within wide range; their diameters do not exceed few nm.
3. Results and discussion
3.1. Characterization of the catalyst
Ru/FCN catalysts of the type used in this study have been thor-
oughly characterized in our earlier reports [10,11]. For this reason in
this work we present only selected data with focus on the nature of Ru/
polymer interaction and verification of the nanosize character of Ru
species. Fourier transform IR spectra of FCN carrier and 2 wt.%Ru/FCN
are presented in Fig. 1. The comparison of both spectra reveals that the
most noticeable changes occur within the range characteristic of func-
3.2. Solvent effect
−
1
−1
−1
tional groups of the support, e.g. 1717 cm , 1644 cm , 1384 cm
,
−1
−1
670 cm
and 530 cm
ascribed to CeO (C]O group), NeH, NeC
The type of solvent may strongly affect both the activity and the
Fig. 1. FT-IR spectra of FCN carrier and 2 wt.% Ru/FCN.
147

D. Duraczyńska, et al.
Molecular Catalysis 470 (2019) 145–151
Fig. 2. TEM (left) and HRTEM (right) images of 2 wt.% Ru/FCN and fast Fourier transform (FFT) from the marked area (insert).
Table 2
Solvent effect.
Solvent
ε
ACT
Conversion
TOF [h−1] PhEtol
Selectivity
H
2
solubility x
i
*104 Solubility
in H
ACT %
ACT %
PhEtol %
PhEtol %
2
O
org. phase aq. phase org. phase aq. phase
[
%]
[%]
MeOH
32.7
19.92
6.15
2.38
17
24
13
23
0
7
12
4
69
75
41
67
0
1.61
2.66
2.74
3.82
3.80
7.96
Miscible
Miscible
Miscible
n.a
n.a
n.a
n.a
n.a
n.a
1
n.a
n.a
n.a
89
n.a
n.a
n.a
11
MeOH (10%)
IPA
IPA (10%)
THF (10%)
11
˜ 0
˜ 0
˜ 0
20
˜ 0
32
˜ 0
11
35
THF
1
0
100
0
Toluene
Toluene/H
Cyclohexane
Cyclohexane/H
0.52 g/L (20 °C) 99
2
O (1:1)
2.38/80.1 29
2.02
O (1:1) 2.02/80.1 58
1.94
77
100
79
60
86
77
1
Immiscible
Immiscible
96
92
4
66
34
2
IO
2
8
48
52
IO/H
IO/H
2
O (4:1)
O (1:1)
1.94/80.1 23
61
2
Catalytic conditions: temperature: 40 °C, H
2
pressure:1 bar, reaction time: 6 h. Conversions determined by GC analysis of crude reaction mixtures. TOF calculated
from initial hydrogenation rate (below 10% conversion) as moles of converted substrate per mole of Ru per hour.
selectivity of the catalyst as well as determine the mechanism path itself
26,27]. Therefore the search for the proper solvent is of critical im-
and IPA are protic due to the presence of labile proton bound to oxygen.
[
Toluene, cyclohexane, and isooctane are apolar and aprotic and hence
cannot act as H-bond donors. THF, being a polar aprotic solvent, cannot
participate in H-bond formation. Comparison of both protic solvents,
MeOH and IPA, shows that ACT conversion grows with the solvent
polarity and is higher on MeOH. The effect is opposite to that reported
portance. The results of our studies devoted to the determination of
solvent effect are presented in Table 2. The hydrogenation of un-
substituted acetophenone (ACT) has served as a test reaction. In this
case two alternative pathways are observed (Scheme 1). One involves
hydrogenation of C]O group to give 1-phenylethanol (PhEtol), the
other hydrogenation of aromatic ring of ACT to form cyclohexyl methyl
ketone. 1-phenylethanol, the most desired product, found an applica-
tion as a fragrance or an intermediate in synthesis of various pharma-
ceuticals. Both products, 1-phenylethanol and cyclohexyl methyl ke-
tone, may be hydrogenated even further to 1-cyclohexylethanol.
Hydrogenolysis of 1-phenylethanol and 1-cyclohexylethanol leads to
ethylbenzene and ethylcyclohexane, respectively. Due to the possibility
of side reactions, achieving high selectivity to 1-phenylethanol is dif-
ficult. In our case we observed the formation of all mentioned products
in various amounts.
We found that, in general, the 2 wt.%Ru/FCN catalyst exhibits
better activity in polar solvents (MeOH and IPA) as compared to apolar
solvents (toluene, cyclohexane, isooctane), although a clear relation-
ship regarding the impact of solvent polarity, indicated by the value of
its dielectric constant (ε) [28], on the observed conversions, cannot be
established. The following order of catalysts activity is observed: to-
luene (ε = 2.38) = THF (ε = 6.15) ≈ Cyclohexane (ε = 2.02) ≈ IO
2
for Ni/SiO catalyst by Bertero et al. [18]. The authors argued that the
more polar MeOH solvates ACT molecules more efficiently than IPA,
thereby hindering the reactant adsorption on the metallic surface. Ob-
viously, in the case of Ru/FCN catalysts it is not the ACT solvation that
plays the decisive role. This aspect is addressed in more detail further.
Solvent effect is sometimes related to the variations in hydrogen
solubility, as better solubility is expected to afford higher concentration
of hydrogen on the catalyst surface [18]. From this point of view, sol-
vents with lower dielectric constant, characterized by higher hydrogen
solubility should be preferred. Analysis of the data in Table 2 shows
that there is no correlation between the ACT conversion and H
bility in the employed solvents [29]. In particular, the best performance
is observed in MeOH, with the lowest H solubility, while only very low
2
solu-
2
activity is displayed in IO, possessing the highest capacity for dissolving
hydrogen.
It should be noted, that even in polar solvents the obtained ACT
conversions are relatively low, 17% in methanol, and 13% in 2-pro-
panol. The results change dramatically when water is introduced to the
reaction mixture. In all instances but THF, much higher conversions are
observed. Noteworthy, the addition of water to MeOH, IPA and THF (all
(
ε = 1.94) < IPA (ε = 19.92) < MeOH (ε = 32.70). Surprisingly, no
catalytic activity is observed when the hydrogenation reaction is per-
formed in tetrahydrofuran (THF), whose polarity is distinctly higher
than that of hydrocarbon solvents. This suggests that, beside polarity,
yet another factor, determining the solvent properties, is of importance.
It appears, that this factor may be the ability of the solvent molecules to
act as hydrogen bond donors. Of the investigated solvents, only MeOH
2
miscible with H O) is limited by the low solubility of acetophenone in
water. The addition of larger amount of water leads to an incomplete
dissolution of acetophenone in the reaction medium (visible as tiny,
individual drops of the second liquid phase dispersed in the solvent).
The largest impact of water is observed with apolar, water-immiscible
148

D. Duraczyńska, et al.
Molecular Catalysis 470 (2019) 145–151
solvents, such as toluene, cyclohexane and isooctane. The conversions
jump from 0% to 29% for toluene, from 1% to 58% for cyclohexane and
from 2% to 61% for isooctane upon the addition of water to the ap-
propriate solvent (1:1).
In an attempt to rationalize the experimental data for polar solvents,
we propose to consider the impact of solvent not only on ACT substrate
it has been pointed out that accumulation of 1-phenylethanol at the
catalyst surface has a detrimental effect on catalysis, due to blocking of
the active Ru sites, but water acts as an excellent scavenger of adsorbed
reaction product and ensures increase of the catalytic activity. Thus, in
the case of biphasic systems one should consider solvent-reactant in-
teractions such as the susceptibility of acetophenone and the desired
product, 1-phenylethanol, to reside in organic and/or aqueous phases.
Of particular importance is the solubility of 1-phenylethanol in aqueous
phase, facilitating recovery of active Ru surface. For this reason we
have determined the appropriate distribution coefficients and the data
[
18], but also on the reaction product, 1-phenylethanol. The efficiency
of solvation is critically dependent on the protic/aprotic character of
the solvent [30]. Thus, polar protic solvents, i.e. MeOH and IPA, are
capable of solvating reagents both by dipole-dipole interactions, and by
formation of hydrogen bonds. The distinctly higher polarity of MeOH
results in stronger solvation ability. If the adsorption of ACT on the
catalyst surface were the rate determining step, the solvation effects,
hindering ACT interaction with the Ru surface, would be detrimental
for the catalyst activity, and a solvent with the lower polarity, i.e. IPA,
would be preferred. This is clearly not the case. In contrast, if the re-
action rate depended on 1-phenylethanol desorption from the catalyst,
the solvation effects should facilitate the removal of the reaction pro-
duct and enhance the ACT conversion by preventing blockage of the
active sites. In that case, the more polar protic solvent would secure a
higher catalytic activity. Given the observed dependencies, it is rea-
sonable to assume that the better performance in MeOH, as compared
to the less polar IPA, is due to the more efficient solvation and fa-
cilitated removal of the reaction product, which makes room for ad-
sorption of ACT molecules. Solvating ability of THF towards the re-
agents is much lower, because the solvent is not only less polar, but
also, as aprotic, not capable of H-bond donation. In view of this, we
attribute the lack of activity in THF to the fact that this solvent fails to
clean the catalytically active Ru phase from accumulated reaction
product. The investigated apolar solvents, toluene, cyclohexane and
isooctane, also lack the ability to interact strongly with 1-pheny-
lethanol, hence no meaningful activity is observed when these solvents
are used as a reaction medium.
2
are given in Table 2. For IO/H O solvent system, where the highest
conversion (61%) is observed, 92% of ACT stays in organic phase and
8% in water. But most (52%) of 1-phenylethanol remains in aqueous
2
phase. Regarding cyclohexane/H O solvent system (second best con-
version of 58%) 96% of ACT resides in cyclohexane and 4% in water.
Smaller amount of 1-phenylethanol (34%) stays in water as compared
to organic phase (66%). For the system with lowest conversion (29%),
2
i.e. toluene/H O, almost all ACT (99%) occupies organic phase, and
only 11% of 1-phenylethanol “prefers” water. In view of this, one can
notice a clear trend: the higher the affinity of the product to the water
phase, the higher the activity. The effect confirms the conclusion on the
importance of removal of the reaction product from the Ru catalyst
surface. However, when analyzing the data in Table 2, one can notice
that, when passing from cyclohexane to isooctane, only a very modest
enhancement of activity is observed, despite a considerable increase of
organic reagents solubility in water. This suggests that yet another
factor may play a role in determining the catalytic performance in the
systems involving two immiscible solvents. An important parameter
governing the behavior of biphasic liquid systems is the interfacial
tension, which influences the degree of dispersion of one phase in the
other [31]. The lower is the surface tension between water and the
organic liquid, the better the mutual dispersion of both components and
the higher the interfacial area. The water/cyclohexane interfacial ten-
sion is ca. 49 mN/m [32], while that of water/isooctane system equals
59 mN/m [33]. Therefore, it is expected that, for the same mixing
speed, the interface area generated in the case of isooctane will be
lower than for cyclohexane. As indicated above, the catalyst resides at
the interface, so the development of the contact area between both
phases is essential for the efficient catalysis. In view of this, we propose
that suppression of the interface area for water/isooctane system, as
compared to the cyclohexane/water counterpart, is the reason for the
very limited enhancement of the catalytic performance in the former
mixture, despite increase of the reagents solubility in water. Note-
worthy, the biphasic isooctane/water system for acetophenone hydro-
Similar arguments may be employed for explanation of the effect of
water addition to the investigated solvents. In MeOH/H
2
O and IPA/
H
2
O systems, addition of a more polar component capable of strong H-
bonding renders the solvent even more efficient in solvating the reac-
tion product and accelerating its desorption, which results in an im-
proved activity. The effect is negligible in the case of THF, which shows
that 10% of water, forming single phase with the main solvent, is not
enough to provide sufficient supply of polar and protic molecules to the
catalyst surface and promote desorption of the product in a meaningful
way. A different situation occurs with “biphasic” solvent systems (to-
2 2 2
luene/H O, cyclohexane/H O and IO/H O). It should be noted that in
contrast to single solvent case (Fig. 3a) the catalyst remains at the in-
terface between two phases (Fig. 3b) in the static conditions, both be-
fore and after the catalytic test. This is understandable, since the hy-
drophobic part of the polymer support (styrene chain) has affinity to
the organic phase and the hydrophilic part (functional groups, to which
Ru is bonded) tends towards the aqueous phase.
Thus, it is reasonable to assume that also during the catalytic test
the Ru-bearing hydrophilic part would tend to remain in contact with
water. In the paragraph describing the phenomena in the polar solvents,
2
genation (50 °C, 1 bar H ) has also been employed by Tundo et al. [9].
Using a Pt/C catalyst modified with cynchonidine, the authors reported
conversions in the range 25–78%, depending on the amount of modi-
fier.
The above reasoning accounts for the effect of water addition on
catalytic activity, but there are other features that make water a par-
ticularly advantageous solvent, due to the possible impact on Ru cata-
lyst selectivity. As pointed out in the review by Michel and Gallezot
[34] one should consider different mechanisms occurring at water/
2
Fig. 3. The location of 2 wt.% Ru/FCN in a) IO b) IO/H O solvent system.
149

D. Duraczyńska, et al.
Molecular Catalysis 470 (2019) 145–151
ruthenium interface. First, DFT calculations demonstrated that forma-
tion of hydrogen bonds between the C]O groups adsorbed on the Ru
surface and adjacent co-adsorbed water molecules lowers the activation
energy of the first hydrogenation step, leading in consequence to en-
hanced hydrogenation rate [35]. Second, the dissociation of water in-
creases the surface concentration of H atoms, which contributes to the
hydrogenation of carbonyl group [36]. In addition, Li. et al. [37]
showed that oxidized ruthenium species, whose presence at the surface
of the catalyst used in this work has been revealed by XPS, facilitate the
process of water splitting. Moreover, the existing long-range electron
polarization of water film [38] on the catalyst surface most likely in-
fluences the orientation of the substrate leading to the preferential re-
duction of C]O group. The addition of water improves the selectivity
to the desired product, 1-phenylethanol, as well (Table 2). It may be
speculated that it affects the adsorption mode of ACT on the catalyst
surface in the way which facilitates the hydrogenation of C]O group,
while limiting to some extent the access to aromatic ring.
methoxyacetophenone, respectively. The result agrees with the ob-
servation of Zaccheria et al. [4], who observed acceleration of hydro-
genation upon insertion of electron donating substituents. On the other
hand an electron withdrawing group, like –Cl– (entry 6), slows down
the process and after the same period of time only 29% conversion is
observed. It should also be mentioned that for 4′-chloroacetophenone
no products indicating hydrodechlorination process could be detected
[2]. The observed trends may be assigned to changes in the polarization
of C]O group, which increases upon action of an electron donating
substituent, or decreases in the presence of an electron withdrawing
group.
When comparing the results for methylacetophenones possessing
the −CH substituent (electron donating group) in ortho, meta or para
3
position (entries 2, 3, and 4), one can notice that the highest conversion
(81%), pointing to a promoting effect, is achieved only in the case of 4′-
methylacetophenone, for which the methyl group is located in the para
position. For both the 2′-methylacetophenone (ortho) and the 3′-me-
thylacetophenone (meta), the obtained conversions are similar to that
of the unsubstituted ACT, i.e. 66% and 62%, respectively, despite the
presence of the same electron donating group. When looking for an
explanation of this phenomenon, one should consider that a methyl
group localized closer to the carbonyl bond constitutes an obvious steric
hindrance which adversely affects the reaction rate. Apparently, the
resulting inhibiting effect obliterates the promoting influence of elec-
tron donation, leaving the activity essentially at the same level.
No steric hindrance is noticed in the case of ethyl phenyl ketone
(entry 7) with no functional group on phenyl ring and an extra CH2
group in the aliphatic chain. In this case the obtained conversion (68%)
is similar to that of parent acetophenone (61%).
3.3. Substituent effect
The influence of the functional groups on the catalytic activity of the
studied catalyst was investigated for seven ACT derivatives (Table 3).
When comparing acetophenone derivatives with substituents at
para position the impact of substituent type, i.e. electron donating vs.
electron withdrawing character, can easily be observed. The electron
donating groups (−CH
tend to increase the reaction rate and after 6 h of catalytic test 81% and
9% conversions were observed for 4′-methylacetophenone and 4′-
3 3
and −OCH ) (entries 4 and 5, respectively)
9
Table 3
Hydrogenation of acetophenone and its derivatives.
4. Conclusions
Ketone
Conversion
%]
TOF
Selectivity to desired product
[%]
[h 1]
−
The product–solvent interactions are of key importance for under-
[
standing the performance of Ru/FCN catalyst in hydrogenation of
acetophenone. It is proposed that the solvent-induced removal of the
reaction product (1-phenylethanol) from the Ru catalyst surface, to
make room for the adsorption of substrate molecules, is determining the
rate of the catalytic reaction. Solvating efficiency towards 1-pheny-
lethanol increases with the solvent’s ability to form dipole-dipole in-
teractions and hydrogen bonds, therefore the solvents of choice should
display possibly high polarity combined with protic character. For this
reason catalytic activity is observed only in MeOH and IPA, which are
both polar and protic, while the use of polar but aprotic (THF) and
apolar aprotic solvents (toluene, isooctane and cyclohexane) renders
the system inactive. The effect of 10% admixture of water (highly polar
component capable of strong H-bonding) is interpreted in similar terms.
6
1
35
38
77
1
.
.
6
6
87
2
6
8
2
1
36
75
87
76
3
.
2
Thus, addition of H O enhances solvating ability of MeOH and IPA,
which results in an improved activity, while its effect on poorly sol-
vating THF is insufficient to promote desorption of the product, so that
no meaningful activity is observed. The profound impact of water ad-
dition to apolar solvents (toluene, isooctane and cyclohexane), which
4
5
.
.
9
2
9
9
150
20
72
85
2
form biphasic systems with H O, is related to the hydrophilic/hydro-
phobic character of Ru/FCN catalyst. Its hydrophilic part (functional
groups, to which Ru is bonded) tends towards the aqueous phase and
away from the apolar solvent, therefore the surface of Ru may easily
come in contact with water. The latter acts as an excellent scavenger of
adsorbed reaction product and ensures increase of the catalytic activity.
The effect of substituents depends on their electron-donating/
withdrawing character, and/or on their position on the aromatic ring.
Electron donating groups tend to increase the reaction rate, while the
electron withdrawing group slows down the process. Steric effect for
hydrogenation of methylacetophenone isomers, manifested by the
6
7
.
.
6
8
40
64
Catalytic conditions: temperature: 40 °C, H2 pressure: 1 bar, reaction time: 6 h,
solvent system: IO/H2O (1:1). Conversions determined by GC analysis of crude
reaction mixtures. TOF calculated from initial hydrogenation rate (below 10%
conversion) as moles of converted substrate per mole of Ru per hour.
highest conversion obtained for the CH
tion, is observed.
3
group attached at ortho posi-
150

D. Duraczyńska, et al.
Molecular Catalysis 470 (2019) 145–151
Acknowledgement
catalysts for the stereoselective hydrogenation of paracetamol to 4-trans-acet-
amidocyclohexanol: effect of support, metal precursor, and solvent, J. Catal. 229
(
2005) 439–445, https://doi.org/10.1016/j.jcat.2004.11.018.
This work was supported by the Polish National Science Center
NCN), grant OPUS 2012/07/B/ST5/00770.
[
[
18] N.M. Bertero, A.F. Trasarti, C.R. Apesteguia, A.J. Marchi, Solvent effect in the li-
(
quid-phase hydrogenation of acetophenone over Ni/SiO : a comprehensive study of
the phenomenon, Appl. Catal. A Gen. 394 (2011) 228–238, https://doi.org/10.
1
2
016/j.apcata.2011.01.003.
References
19] A.F. Trasarti, N.M. Bertero, C.R. Apesteguia, A.J. Marchi, Liquid-phase hydro-
genation of acetophenone over silica-supported Ni, Co and Cu catalysts: influence of
metal and solvent, Appl. Catal. A Gen. 475 (2014) 282–291, https://doi.org/10.
[
[
[
1] R.V. Malyala, C.V. Rode, M. Arai, S.G. Hedge, R.V. Chaudhari, Activity, selectivity
and stability of Ni and bimetallic Ni–Pt supported on zeolite Y catalysts for hy-
drogenation of acetophenone and its substituted derivatives, Appl. Catal. A Gen.
1
016/j.apcata.2014.01.038.
[20] H. Cheng, J. Hao, H. Wang, C. Xi, X. Meng, S. Cai, F. Zhao, (R,R)-DPEN-modified
—an efficient heterogeneous catalyst for enantioselective hydrogena-
1
93 (2000) 71–86, https://doi.org/10.1016/S0926-860X(99)00442-1.
2 3
Ru/γ-Al O
2] P. Tundo, S. Zinovyev, A. Perosa, Multiphase catalytic hydrogenation of p-
Chloroacetophenone and acetophenone. A kinetic study of the reaction selectivity
toward the reduction of different functional groups, J. Catal. 196 (2000) 330–338,
https://doi.org/10.1006/jcat.2000.3054.
3] A. Vargas, T. Bürgi, M. von Arx, R. Hess, A. Baiker, Relation between electronic
structure of α-Substituted ketones and their reactivity in racemic and en-
antioselective platinum-catalyzed hydrogenation, J. Catal. 209 (2002) 489–500,
https://doi.org/10.1006/jcat.2002.3653.
tion of acetophenone, J. Mol. Catal. A: Chem. 278 (2007) 6–11, https://doi.org/10.
1016/j.molcata.2007.08.015.
[
21] A. Bukowska, W. Bukowski, J. Noworol, Three-component terpolymers of glycidyl
methacrylate with good swelling characteristics, J. Appl. Polym. Sci. 106 (2007)
3
800–3807, https://doi.org/10.1002/app.26009.
[22] A.A. Soliman, M.A. Amin, A.A. El-Sherif, C. Sahin, C. Varlikli, Synthesis, char-
acterization and molecular modeling of new ruthenium(II) complexes with nitrogen
and nitrogen/oxygen donor ligands, Arab. J. Chem. 10 (2017) 389–397, https://
doi.org/10.1016/j.arabjc.2015.04.001.
23] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, J. Chastain (Ed.), Handbook of
X-Ray Photoelectron Spectroscopy: A Reference Book of Standard Spectra for
Identification and Interpretation of XPS Data, 2nd ed., Perkin-Elmer Corporation
Physical Electronics Division, United States of America, 1992, pp. 114–115.
[24] Electron Spectroscopy Database, (2018) Accessed January http://www.lasurface.
com.
[25] NIST X-ray Photoelectron Spectroscopy Database, (2018) Accessed January http://
srdata.nist.gov/xps/.
[
[
[
[
[
4] F. Zaccheria, N. Ravasio, R. Psaro, A. Fusi, Heterogeneous selective catalytic hy-
[
drogenation of aryl ketones to alcohols without additives, Tetrahedron Lett. 46
(2005) 3695–3697, https://doi.org/10.1016/j.tetlet.2005.03.159.
5] O.J. Sonderegger, G.M.-H. Wo, T. Bürgi, A.J. Baiker, Enantioselective hydrogena-
tion of aromatic ketones over cinchona-modified rhodium: a new opportunity? J.
Catal. 230 (2005) 499–506, https://doi.org/10.1016/j.jcat.2005.01.008.
6] R. Hess, T. Mallat, A. Baiker, Enantioselective hydrogenation of aromatic ketones:
structural effects, J. Catal. 218 (2003) 453–456, https://doi.org/10.1016/S0021-
9
517(02)00179-3.
7] V. Vetere, A.B. Merlo, M.L. Casella, Chemoselective hydrogenation of aromatic
[26] S. Song, Y. Wang, N. Yan, A remarkable solvent effect on reductive amination of
ketones, Mol. Catal. 454 (2018) 87–93, https://doi.org/10.1016/j.mcat.2018.05.
ketones with Pt-based heterogeneous catalysts, Substituent effects, Appl. Catal. A.
0
17.
4
91 (2015) 70–77, https://doi.org/10.1016/j.apcata.2014.11.010.
8] Y. Chaofen, Y. Jun, S. Xiaodong, Z. Yanqin, W. Qi, C. Hua, (1S,2S)-
Diphenylethylenediamine modified Ir/SiO catalyst for asymmetric hydrogenation
[27] H. Wang, C. Liu, D. Zhang, Decisive effects of solvent and substituent on the re-
activity of Ru-catalyzed hydrogenation of ethyl benzoate to benzyl alcohol and
ethanol: a DFT study, Mol. Catal. 440 (2017) 120–132, https://doi.org/10.1016/j.
mcat.2017.06.026.
2
of acetophenone and its derivatives, Chin. J. Catal. 33 (2012) 1154–1160, https://
doi.org/10.1016/S1872-2067(11)60403-8.
9] P. Tundo, A. Perosa, S. Zinovyev, Modifier effects on Pt/C, Pd/C, and Raney-Ni
[
28] Dielectric constants based onhttp://macro.lsu.edu/HowTo/solvents/Dielectric
[
catalysts in multiphase catalytic hydrogenation systems, J. Mol. Catal. A Chem.
%20Constant%20.htm.
H
5-6.pdf (1 bar, 25 °C).
[
29]
2
solubility data based on https://srdata.nist.gov/solubility/IUPAC/SDS-5-6/SDS-
2
04–205 (2003) 747–754, https://doi.org/10.1016/S1381-1169(03)00360-1.
10] D. Duraczyńska, A. Drelinkiewicz, E.M. Serwicka, D. Rutkowska-Żbik, E. Bielańska,
R.P. Socha, Preparation and characterization of RuCl – diamine group functiona-
[
[
[
[30] L.J. Diorazio, D.R.J. Hose, N.K. Adlington, Toward a More Holistic Framework for
Solvent Selection, Org. Process Res. Dev. 20 (2016) 760–773, https://doi.org/10.
1021/acs.oprd.6b00015.
31] L. Abu-Farah, F. Al-Qaessia, A. Schönbuche, Cyclohexane/water dispersion beha-
viour in a stirred batch vessel experimentally and with CFD simulation, Procedia
Comput. Sci. 1 (2012) 655–664, https://doi.org/10.1016/j.procs.2010.04.070.
[32] S. Wald, F.R. Wurm, K. Landfester, D. Crespy, Stabilization of inverse miniemul-
sions by silyl-protected homopolymers, Polymers 8 (2016) 303–312, https://doi.
org/10.3390/polym8080303.
[33] M.N.D.S. Corderio, Interfacial tension behaviour of Water/Hydrocarbon liquid–li-
quid interfaces: a molecular dynamics simulation, Mol. Simulat. 29 (2003)
817–827, https://doi.org/10.1080/0892702031000121905.
[34] C. Michel, P. Gallezot, Why Is Ruthenium an Efficient Catalyst for the Aqueous-
Phase Hydrogenation of Biosourced Carbonyl Compounds? ACS Catal. 5 (2015)
4130–4132, https://doi.org/10.1021/acscatal.5b00707 and references therein.
35] B.S. Akpa, C. D’Agostino, L.F. Gladden, K. Hindle, H. Manyar, J. McGregor, R. Li,
M. Neurock, N. Sinha, E.H. Stitt, D. Weber, J.A. Zeitler, D.W. Rooney, Solvent ef-
fects in the hydrogenation of 2-butanone, J. Catal. 289 (2012) 30–41, https://doi.
org/10.1016/j.jcat.2012.01.011.
[36] J. Tan, J. Cui, T. Deng, X. Cui, G. Ding, Y. Zhu, Y. Li, Water‐promoted hydro-
genation of levulinic acid to γ‐Valerolactone on supported ruthenium catalyst,
ChemCatChem 7 (2015) 508–512, https://doi.org/10.1002/cctc.201402834.
3
lized polymer, React. Funct. Polym. 7 (2010) 382–391, https://doi.org/10.1016/j.
reactfunctpolym.2010.03.003.
[
11] D. Duraczyńska, A. Drelinkiewicz, E. Bielańska, E.M. Serwicka, L. Litynska-
Dobrzyńska, Hydrogenation of acetophenone in the presence of Ru catalysts sup-
ported on amine groups functionalized polymer, Catal. Lett. 141 (2011) 83–94,
https://doi.org/10.1007/s10562-010-0448-3.
12] A. Drelinkiewicz, A. Knapik, A. Waksmundzka-Góra, A. Bukowska, W. Bukowski,
J. Noworól, Functional gel-type resin based palladium catalysts: the role of polymer
properties in the hydrogenation of the C=C bond of maleic and fumaric acids, the
isomers of dicarboxylic acids, React. Funct. Polym. 68 (2008) 1059–1071, https://
doi.org/10.1016/j.reactfunctpolym.2008.02.008.
13] A. Drelinkiewicz, W. Stanuch, A. Knapik, A. Ghanem, R. Kosydar, A. Bukowska,
W. Bukowski, Amine groups functionalized gel-type resin supported Pd catalysts:
physicochemical and catalytic properties in hydrogenation of alkynes, J. Mol. Catal.
A: Chem. 300 (2009) 8–18, https://doi.org/10.1016/j.molcata.2008.10.035.
14] M.G. Al -Shaal, W.R.H. Wright, R. Palkovits, Exploring the ruthenium catalysed
synthesis of γ-valerolactone in alcohols and utilisation of mild solvent-free reaction
conditions, Green Chem. 14 (2012) 1260–1263, https://doi.org/10.1039/
C2GC16631C.
[
[
[
[
15] A.S. Piskun, J. Ftouni, Z. Tang, B.M. Weckhuysen, P.C.A. Bruijnincx, H.J. Heeres,
Hydrogenation of levulinic acid to γ-valerolactone over anatase-supported Ru cat-
alysts: effect of catalyst synthesis protocols on activity, Appl. Catal. A Gen. 549
[
2
37] Y. Li, Q. Zhang, N. Zhang, L. Zhu, J. Zheng, Ru-RuO /C as an efficient catalyst for
(2018) 197–206, https://doi.org/10.1016/j.apcata.2017.09.032.
the sodium borohydride hydrolysis to hydrogen, Int. J. Hydrogen Energy 38 (2013)
13360–13367, https://doi.org/10.1016/j.ijhydene.2013.07.071.
[38] S. Meng, E.G. Wang, S. Gao, A molecular picture of hydrophilic and hydrophobic
[
16] J. Hajek, N. Kumar, P. Mäki-Arvela, T. Salmi, D.Yu. Murzin, Selective hydrogena-
tion of cinnamaldehyde over Ru/Y zeolite, J. Mol. Catal. A: Chem. 217 (2004)
1
45–154, https://doi.org/10.1016/j.molcata.2004.03.015.
interactions from ab initio density functional theory calculations, J. Chem. Phys.
1
19 (2003) 7617–7620, https://doi.org/10.1063/1.1617974.
[
17] B. Bachiller-Baeza, A. Guerrero-Ruiz, I. Rodriguez-Ramos, Ruthenium-supported
151