The complexes of copper with grafted ionic liquids in the environmentally important processes

Article abstract of DOI:10.1016/j.apcata.2013.10.041

The complexes of copper (II) chloride with quaternary ammonium salts grafted on various mineral supports (the modifications of SiO₂ - KSK, silochrome, Perlkat) are highly active in two environmentally important processes: the reaction of carbon tetrachloride with alkanes and the oxidative coupling of thiols. The effectiveness of the catalysts depends on the transition metal, the ionic liquid, and the support nature. In CCl₄ transformations, the highest activity is shown by the complexes with alkylamines derivatives grafted on silica with wide pores. In case of thiol oxidation, complexes with heterocycles derivatives grafted on Perlkat with narrow pores are most active. The effectiveness of the CCl₄ transformations catalysts increases upon the addition of alcohol to the reaction mixture.

Full text of DOI:10.1016/j.apcata.2013.10.041

Applied Catalysis A: General 470 (2014) 81–88
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
The complexes of copper with grafted ionic liquids in the
environmentally important processes
Irina Tarkhanova, Vladimir Zelikman, Mikhail Gantman∗
M.V. Lomonosov Moscow State University, Leninskiye Gory, 1 b. 3, 119991 Moscow, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 4 June 2013
Received in revised form 15 October 2013
Accepted 21 October 2013
Available online 30 October 2013
The complexes of copper (II) chloride with quaternary ammonium salts grafted on various mineral sup-
ports (the modifications of SiO2 – KSK, silochrome, Perlkat) are highly active in two environmentally
important processes: the reaction of carbon tetrachloride with alkanes and the oxidative coupling of
thiols. The effectiveness of the catalysts depends on the transition metal, the ionic liquid, and the support
nature. In CCl4 transformations, the highest activity is shown by the complexes with alkylamines deriva-
tives grafted on silica with wide pores. In case of thiol oxidation, complexes with heterocycles derivatives
grafted on Perlkat with narrow pores are most active. The effectiveness of the CCl4 transformations
catalysts increases upon the addition of alcohol to the reaction mixture.
Keywords:
Copper complexes
Thiol oxidation
Ionic liquids
© 2013 Elsevier B.V. All rights reserved.
Pyridinium
Imidazolium
Alkane chlorination
Immobilized metal complexes
1. Introduction
These processes are the examples of the model radical reactions
of various nature: alkanes chlorination occurs as a radical-chain
In recent years, ionic liquids (IL) have been widely used as
media for performing chemical processes and as components of
catalytic compositions. A promising trend is the development of
heterogeneous metal complex catalysts based on IL grafted on min-
eral supports [1–3]. The immobilized catalysts were found to be
active in various processes: e.g., hydroformylation [4–6], hydro-
genation [7,8], hydrodechlorination [9], methanol carbonylation
[10], Friedel–Crafts reaction [11,12], oxidation of phenols [13], alco-
hols [14] and thiophene [15], Heck reaction [16]. At the same time
there are very few radical reactions that can be catalyzed by such
complexes. Kharasch addition of CCl4 to alkenes is the example of
such process [17,18]. We present two other radical processes that
can be catalyzed by immobilized copper complexes with ionic liq-
uids. The first one is the liquid-phase oxidation of thiols with air
oxygen, leading to the formation of disulfides:
process, while thiol oxidation is a non chain process. The coor-
dinating and redox properties of sulfur- and halogen-containing
compounds differ significantly. Nevertheless, both reactions con-
tain redox stages characteristic for metal complex initiation. The
above mentioned processes are also of practical significance. The
first one is used in petrochemistry [19,20]. The second one is
a promising method for disposal of banned carbon tetrachloride
(CTC), as it allows the initial organic chlorine to be fully utilized to
form valuable commercial products [21,22].
We have previously shown the metal complexes with ILs grafted
on silica gel by covalent bonding between hydroxyl groups of the
surface and functional groups of the ionic liquid to be more active
in the chlorination process than the catalytic systems, obtained by
physical adsorption of the IL complex of the mineral support [22].
Compositions based on copper (II) chloride and quaternary
ammonium bases either in solutions or adsorbed on a surface
were proposed as catalysts for this reaction [21,22]. The main
shortcoming of the method is fast catalyst deactivation, which is
caused by necessity to perform the process at high temperature.
We have stated that this shortcoming can be avoided if transi-
tion metal complexes with thermally stable quaternary ammonium
bases (QAB) derived from imidazole, pyridine or short-chain alky-
lamines immobilized on the silica gel surface are used together with
aliphatic alcohols (ethanol or propanol). Using this approach, the
yield of chloroalkanes was increased and the catalyst stability was
4RSH + O2 = 2RSSR + 2H2O
and the liquid phase interaction between alkanes and carbon tetra-
chloride:
RH + CCl4 → RCl + CHCl3
∗
Corresponding author. Tel.: +7 49 5939 3498; fax: +7 49 5932 8846.
E-mail address: Mikhail.gantman@gmail.com (M. Gantman).
0926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.10.041

82
I. Tarkhanova et al. / Applied Catalysis A: General 470 (2014) 81–88
Fig. 1. The scheme of the immobilization of the ionic liquids complexes on the supports.
enhanced, and thus the catalysts became suitable for repeated use
without loss of activity.
stage the modifier of the surface (3-chloropropyltrimethoxysilane)
interacted with amine to form quaternary ammonium salt, and
then the surface was modified with this salt. The modification with
the triethylpropylammonium chloride is described in [22].
To obtain the grafted complexes, ethanolic solution of metal
chloride (CuCl2, MnCl2, CoCl2) reacted with the modified SiO2 in
a flask under intense magnetical stirring for 1 h and slight heating
(50–70 ◦C). After that the solid phase was filtered off and dried on
air.
2. Experimental
2.1. The catalysts synthesis
The following supports were used to obtain the cata-
lysts: silochrome C-120 (surface area ≈120 m2/g, pore diame-
ter = 40–45 nm), silica gel KSK-2 (surface area ≈300 m2/g, pore
diameter = 12–15 nm), Perlkat (surface area ≈300–450 m2/g, pore
diameter = 9–10 nm).
2.2. Physico-chemical study of the catalysts
The Cu(II), Co(II) and Mn(II) chloride complexes with immobi-
lized N-ethyl-Nꢀ-propylimidazolium chloride (EtPrImCl–SiO2) on
KSK were prepared by a reported procedure [9]. The difference was
that the quaternization was carried out in a vacuum rather than in
dry argon.
The grafted catalysts (on KSK, Perlkat and silochrome) with
propylpyridinium chloride was prepared by immobilizing the
cation (see Fig. 1), using the technique, described in [23] with a
slight modification.
Typical synthesis of the catalyst was performed as follows: At
the first stage 3-chloropropyltrimethoxysilane was added to dehy-
drated SiO2 (silochrome, KSK or Perlkat). Toluene was used as a
solvent. Weight ratio of the support and modifier was 2:1. The
process was performed in a flask equipped with a Dean and Stark
distillation head and a reflux condenser. The reaction mixture was
boiled until the water stopped accumulating; then the solid phase
was filtered off, washed with dry toluene using 10 ml of solvent per
gram of support, and dried on air. On the next step the quaterniza-
tion was done as follows: 2.8 ml of pyridine reacted with 2.0 g of the
modified in a sealed glass ampoule at 160 ◦C under intense stirring
for 6 h. After this the excess of pyridine was washed off with 25 ml
of ethanol under intense stirring, and then the solid was filtered
off and dried under 50–70 ◦C. The difference from the technique,
described in [23] is as follows. In the literature method on the first
The grafting of the complex was monitored by IR spectroscopy
for KBr pellets using an Infralum FT-801 FT IR spectrophotome-
ter. Treatment of silica gel with the modifying agent induced an
irreversible decrease in the intensity of the bands for the termi-
nal surface OH groups (3745 cm−1) and appearance of the bands
at 2800–3000 cm−1, which did not disappear even after long-term
keeping of the samples in a thermostat at 180–200 ◦C.
The diffuse reflectance electronic spectra were measured on an
HR4000CG-UV-NIK Ocean Optics instrument with an integrating
sphere in the range of 300–1200 nm.
The thermogravimetric measurements were carried out on a
Derivatograph-C instrument in air and under argon at a heating
rate of 10 ◦C/min for ∼15 mg samples.
The amount of metal on the surface was determined by com-
plexonometry. The amount of modifier on the support surface was
stated by C-, N-, H-elemental analysis on a Vario MAX apparatus.
13C NMR spectra were recorded on a BRUKER AVANCE-II NMR
400 with using of solid-state 4 mm H/X MAS probe. A RAMP cross-
polarization pulse sequence with SW-TPPM high-power proton
decoupling was used. The contact time–2 ms; recycle delay–2 s. The
chemical shift scale was referred relatively to the CH2 group of solid
adamantane. A sample for NMR–MAS investigation contained the
catalytic sample and 10-fold excess of dodecane-1-thiol. After com-
pletereductionof copper(whichwasregisteredas thediscoloration

I. Tarkhanova et al. / Applied Catalysis A: General 470 (2014) 81–88
83
of the catalyst) the sample was placed into the ceramic beaker and
the spectrum was registered. After the investigation the thiol was
still present in the system.
2.3. Catalytic experiments
The oxidation of thiol was performed in a glass flask at room
temperature under vigorous magnetic stirring. Isooctane (25 ml),
a catalyst (0.1 g), and dodecane-1-thiol (5 × 10−2 M) were placed
in the flask; air oxygen dissolved in the reaction mixture was
used as the oxidizer. The amount of thiol was measured by
means of potentiometric titration with diamine silver nitrate in
the presence of ion-selective argentite electrode [24]. The cat-
alytic activity (turnover frequency (TOF)) was calculated as the
ratio of the initial rate of thiol concentration decrease to the
concentration of copper. The total stability (turnover number
(TON)) was calculated as the ratio of oxidized thiol and copper
amounts.
The catalytic experiments with CCl4 were performed in sealed
evacuated glass tubes as described in [22]. To prepare a stan-
dard sample, a catalyst, decane and CCl4 were placed in a glass
tube. To remove dissolved air, the cycle “freezing with liq-
uid nitrogen-pumping out-defreezing” was repeated three times,
after that the ampoule was soldered. This ampoule was placed
in a thermostat, equipped with a mechanism for intense stir-
ring of the mixture. The activity of the supported catalysts
was characterized by the overall yield of chlorinated products
(monochlorodecanes and dichlorodecanes), over a 5 h period
at 175 ◦C at the constant initial reactant ratio: 0.05 g of the
heterogenous catalyst, 0.4 ml of CCl4, 0.2 ml of decane and
alcohol (methanol-29 ␮l, ethanol-42 ␮l, isopropyl alcohol-55 ␮l,
propan-1-ol-54 ␮l, 2-methylpropan-2-ol-68 ␮l, butan-1-ol-66 ␮l,
pentan-1-ol-78 ␮l).
In all the studied processes the intense of stirring had no affect
on the rate of the reaction, it means, that diffusion processes had
no influence on the kinetics.
The products were identified by GC/MS on a THERMO TRACE
DSQ II instrument. Quantitative analysis was carried out by GLC-
FID.
3. Results
3.1. The composition and structure of catalysts
The obtained catalysts were air-stable powders containing
5–10 wt.% of organic material and 1.5–2.1 wt.% of the metal. Copper
containing catalysts turned out to be the most active, so their prop-
erties were studied in detail. Table 1 presents the data on elemental
analysis of all the copper containing catalytic systems.
The diffuse reflectance electronic spectra of copper complexes
are shown in Fig. 2. Each spectrum exhibits two absorption bands
typical for Cu2+ chloride complexes: the charge transfer band at
420–480 nm and the band at 750–820 nm corresponding to d–d
transitions of Cu2+ in a distorted octahedron [25]. Such com-
plexes with axial chloride bridges are formed through interaction
2−
of proximate CuCl4 ions. Analogous spectra have been reported
earlier [22]. The charge transfer band at 430 nm corresponds to the
absorption of mononuclear complexes CuCl42−, the band at 480 nm
2−
responds Cu2Cl6 complexes.
Fig. 3 shows the thermogravimetric analysis data for the copper
complexes. It can be seen that in air, the mass loss occurs in the
250–350 ◦C range; thus, the obtained catalysts are stable against
thermolysis in oxidizing medium and are suitable for the use in
the reaction under consideration proceeding at temperatures above
200 ◦C.

84
I. Tarkhanova et al. / Applied Catalysis A: General 470 (2014) 81–88
1
3.2. The products of the studied processes and selectivity
(a)
1.0
0.8
0.6
0.4
0.2
0.0
5
2
All the studied complexes were active in the reactions under
consideration. In the CCl4 transformations
CCl4 + C10H22 → CHCl3 + C10H21Cl
3
in the temperature range of 160–190 ◦C, the major reaction
products included mainly isomeric monochlorodecanes, also
dichlorodecanes were formed (their content in the product mix-
ture did not exceed 15% of the total decane conversion) and minor
amount of trichlorodecanes (less than 0.5%) was found among the
products. In the present manuscript the ratio of decane conver-
sion (not the ratio of monochlorodecanes formation) was used to
characterize the catalytic activity of the studied complexes. Fig. 4
presents the data for a number of the chloride metal complexes
with EtPrImCl–KSK.Table 2 represents the data on the activity of
copper complexes with different ILs on various supports. Fig. 5
shows the results of testing of the CuCl2–EtPrImCl–KSK catalyst in
several successive cycles. The catalyst activity is seen to markedly
drop from one cycle to another.
4
400
600
800
wave length, nm
(b)
0.6
The decane conversion and catalyst stability can be increased
1.5-2-fold by adding aliphatic alcohols to the reaction system. In
this case the copper complexes were the most active too (Fig. 4). The
productivity of the CuCl2–EtPrImCl–KSK and CuCl2–PrEt3NCl–KSK
catalysts in the presence of alcohol remains constant during sev-
eral successive cycles within the determination error (Fig. 6). The
addition of alcohol made no changes in the composition of the
decane transformations products. Thiol oxidative coupling led to
the formation of the corresponding disulfide with no by products.
2
0.4
0.2
0.0
1
3
4RSH + O2 → 2RSSR + 2H2O
400
500
600
700
800
900
Table 3 contains data on the activity and stability of the studied
copper complexes on various supports in the thiol oxidation.
wave length, nm
4. Discussion
(c)
0.7
1
4.1. The influence of metal nature on the activity of the complex
in CCl4 transformations
0.6
0.5
0.4
0.3
0.2
0.1
0.0
2
The activity of the obtained catalysts increases in the range
Co(II) < Mn(II) < Cu(II) (see Fig. 4); the highest yield of chlorodecanes
(34%) was obtained for CuCl2–EtPrImCl–KSK.
Such dependence is typical for the halogenated hydrocarbons
transformations in the presence of transition metal complexes.
Thus, we observed this phenomenon in the Kharasch reaction cat-
alyzed by immobilized metal complexes with amino alcohols [26].
The highest activity of copper complexes as compared with the
other metals can be explained not only by the fact that copper
much easier changes its oxidation state [27]. Another reason is the
well known ability of univalent copper halides and their complexes
(CuYLn) to interact with alkyl halides (RX) via the following way:
RX + CuYLn ꢀ R• + CuYLnX [28,29]. The reversibility and high rates
of the radicals interaction with halogen-containing copper (2+)
compounds present a base of atom transfer radical polymerization
(ATRP) [30]. In the process under investigation carbon tetrachloride
plays the role of RX and trichloromethyl and decyl (formed by the
reaction of decane with CCl3) radicals are the R respectively. The
ability of copper compounds to interact with CCl4 is well-known
[31]. We have also studied these processes and have shown, that
key step in the reaction is copper reduction to univalent state by
donor ligand followed by the oxidation of Cu+ to Cu2+ with CCl4
[32]. These peculiarities of copper complexes are obviously of great
3
400
600
800
wave length, nm
Fig. 2. (a) The electronic spectra of 1-CuCl2–EtPrImCl–KSK (fresh), 2-
CuCl2–EtPrImCl–KSK (after the oxidation of thiol), 3–CuCl2-PrPyCl–KSK (fresh), 4-
CuCl2–PrPyCl–KSK (after the oxidation of thiol), 5-CuCl2– PrEt3NCl–KSK (fresh). (b)
The electronic spectra of 1-CuCl2–EtPrImCl–silochrome (fresh), 2-CuCl2–PrPyCl–
silochrome (fresh), 3-CuCl2–PrEt3NCl–silochrome (fresh). (c) The electronic spectra
of 1-CuCl2–EtPrImCl–Perlkat (fresh), 2-CuCl2–PrPyCl–Perlkat (fresh), 3-CuCl2–
PrEt3NCl–Perlkat (fresh).

I. Tarkhanova et al. / Applied Catalysis A: General 470 (2014) 81–88
85
(a)
(b)
100
1
100
80
2
2
1
80
3
3
4
4
T,oC
T,oC
200
400
600
200
400
600
Fig. 3. TGA-DTG curves for EtPrImCl–CuCl2–KSK (a) and PrPyCl–CuCl2–KSK (b) (1, 3–on air; 2, 4–under argon). Analogous curves for the CuCl2– PrEt3NCl–KSK catalyst see in
[20].
Table 2
Decane conversion (%) on various supports.
Quaternary ammonium base in the composition of complex (L)a
Support
KSK-2
L:Cu
Silochrome
Perlkat
L:Cu
Decane conversion
L:Cu
Decane conversion
Decane conversion
EtPrImCl
PrPyCl
PrEt3NCl
2.5
2.1
1.9
52
29
65
1.8
3.2
1.2
42
47
54
1.8
2.1
1.4
41
33
52
о
Initial conditions: m(cat.) = 0.05 g, n-PrOH–1.2 M; CCl4–6 M, C10H22–1.5 M, 5 h, 175 C.
a
L is the grafted quaternary ammonium base.
Table 3
Activity of catalysts, TOF (mole/(mole Cu*h) and TON (mole/(mole Cu) in dodecane thiol oxidation.
Support
KSK-2
TOF
Silochrome
TOF
Perlkat
Metal complex
TON
TON
TOF
TON
CuCl2–EtPrImCl
CuCl2–PrPyCl
CuCl2–PrEt3NCl
26
12
0.3
130
76
0.5
36
39
12
200
80
15
41
54
6
220a
330a
12
Isooctane (25 ml), a catalyst (0.1 g), dodecane-1-thiol (5 × 10−2 M), 293 K.
a
Here we give the number of cycles, that were performed by the catalysts without any loss of activity, as we were unable to bring those catalysts to the total loss of activity.
importance for the reaction under consideration, since its key stage
is similar to the mentioned one [21,22]:
Fig. 7 shows the results of experiments carried out in the pres-
ence of various aliphatic alcohols with the CuCl2–PrEt3NCl–KSK as
a catalyst. Methanol, propan-2-ol, and pentan-1-ol barely affect the
yield. Ethanol and butan-1-ol increase the efficiency of the catal-
ysis, and the highest product yield is obtained in the presence of
propan-1-ol. The addition of tert-butyl alcohol induces resinifica-
tion of the catalyst by the products of alcohol oxidation, and, as a
consequence, a sharp decrease in the target product yields.
[CuCl] + CCl4 ꢀ [CuCl2] + CCl3
4.2. The influence of alcohols on the reaction
According to gas chromatography/mass spectrometry data,
ethanol is converted to give vinyl chloride CH2 CHCl, acetaldehyde
CH3CH O, and a minor amount of chloroethane CH3CH2Cl.
For the most active additive, propan-1-ol, we measured the
dependence of the yield of the decane chlorination product on
the amount of propan-1-ol. The obtained relation passes through
60
40
35
30
25
20
15
10
5
without alcohol
50
with alcohol
40
30
20
10
0
0
1
2
3
Co
Mn
Cu
Cycle number
Fig. 4. The activity of catalysts CoCl2–EtImPrCl–KSK; MnCl2–EtImPrCl–KSK;
CuCl2–EtImPrCl–KSK in CCl4 transformations Initial conditions: [C10H22] = 1,5 M, [n-
PrOH] = 1,2 M (if added), m(cat) = 0.05 g 5 h., 175 ◦C.
Fig. 5. The activity of CuCl2–EtImPrCl–KSK catalyst in the successive cycles in the
absence of alcohol. Initial conditions: m(cat) = 0,05 g, [CCl4] = 6 M, [C10H22] = 1,5 M,
5 h, 175 ◦C.

86
I. Tarkhanova et al. / Applied Catalysis A: General 470 (2014) 81–88
70
60
50
40
30
20
10
0
CCl4
+RH
CHCl
.
Cu2+
Cu+
2+
red
+ CCl
Cu Cl
.
3
3 +R
.
+R
-RCl
-RCl
+CCl4
Fig. 9. The scheme of the mechanism of CCl4 interaction with alkane.
metal reduction may occur through the elimination of the hydrogen
atom from the ␣-C atom [33].
[CuCl2] + C2H5OH → [CuCl] + CH3CHOH + HCL,
1
2
3
4
Cycle number
however, the alcohol itself acts as a substrate in the substitution,
thus competing with the alkane (RH):
Fig. 6. The activity of CuCl2–Et3NPrCl–KSK (light column), and CuCl2–EtImPrCl-
KSK (dark column) catalyst in the successive cycles in the presence of ethanol
(m(cat) = 0,05 g; [EtOH] = 1,2 M; [C10H22] = 1,5 M, 5 h, 175 ◦C.
CH3CH2OH + CCl3 → CH3CHOH + CHCl3
70
60
50
40
30
20
10
0
RH + CCl3 → R• + CHCl3
The ratio of the rates of these processes is determined by the
nature and the concentration of the alcohol: the maximum effect is
achieved by using C2–C3 alcohols in an optimal ratio to an alkane.
An increase in the alcohol concentration may change this ratio of
the rates in favor of the competing process. The use of alcohol with
longer or branched chain alkyl group also favors this. It can be seen
from the comparison of the energy of ␣
alcohols (methanol-411, ethanol-400, propanol-400, butanol-397,
pentanol-397 kJ/mole) [34]. The last value is quite equal to the C
bond energy in the alkane methylene group. So, alcohols with a
chain, containing more than four carbon atoms, are capable not
only of subtracting of hydrogen atom from ␣ C, but from other car-
bon atoms also. As a result, butanol taken in the same amount as
propanol increases the yield of the target product to a lesser extent
and pentanol almost does not affect the process. The secondary
and tertiary alcohols inhibit the formation of chloroalkanes as
compared with primary alcohols: this is due to their easy transfor-
mations resulting in exhaustion in the very beginning of the process
and resinification of the catalyst initiator. In case of secondary alco-
hol these transformations are connected with its oxidation. In case
of tertiary alcohol we mean, that it can easily interact with carbon
tetrachloride to form 2-methyl-2-chloropropane, which under the
reaction conditions transforms to 2-methylprop-2-ene. The differ-
ence in the ethanol and methanol efficiency can be accounted for by
different solubility of these alcohols in CCl4 [35]. Thus, 1-propanol
provides the highest yield of the target product, being the promotor
of choice for this reaction.
C H bonds in various
H
–
MeOH EtOH i-PrOH n-PrOH t-BuOH n-BuOH n-AmOH
Fig. 7. The dependence of the reaction yield on the nature of alcohol. Catalyst
CuCl2–Et3NPrCl–KSK, m(cat) = 0.05 g, [alcohol] = 1.2 M, [C10H22] = 1,5 M, 5 h, 175 ◦C.
a maximum (Fig. 8). That is, when the alcohol amount in the sys-
tem exceeds some critical value, the yield of the target product
decreases. We have observed the analogous effect also on obtained
by means of adsorption catalysts [22].
The non-monotonic dependences shown in Fig. 7 and Fig. 8
can be interpreted using the published scheme of radical reaction
between alkanes and CCl4 in the presence of metal-complex Red/Ox
initiators [21,22] (Fig. 9).
In the absence of other reducing agents, the organic fragment of
the quaternary base serves as the Red. In the presence of alcohol, the
In our previous work [22], we have studied in detail the pro-
cesses of copper complexes evolution and deactivation. We have
shown by means of EPR spectroscopy, that the donor molecule
coordinates copper ion together with chloride anion, after that
the reduction of copper takes place, followed by its oxidation with
carbon tetrachloride. Also the products of ethanol transformation
were studied there, to prove the mechanism we present in this
manuscript. It is important to note here, that after reaction passes,
the initial ESR spectrum of catalyst appears again. This proves, that
in the presence of donor the structure of copper complex after the
reaction passes restores.
4.3. The influence of the support nature on the catalytic activity
Apart from the specificity of the quaternary base taking part
in the catalytic complex formation and the type of donor addi-
tive, the catalyst activity is also affected by the nature of mineral
support(Table 2).
Fig. 8. The dependence of decane conversion on propanol concentration. Catalyst
CuCl2–Et3NPrCl–KSK (m(cat) = 0.05 g, [CCl4] = 6 M, [C10H22] = 1,5 M, 5 h, 175 ◦C).

I. Tarkhanova et al. / Applied Catalysis A: General 470 (2014) 81–88
87
which corresponds to the absorption of aggregated particles. On
the basis of such spectra peculiarities one may predict the differ-
ence of the synthesized catalysts activity in the thiol oxidation
or decane chlorination. The first stage of the chlorination is the
copper reduction with the ligand. Isolation of copper ion and
surrounding it with a large number of donor ligands favor this
process [40]. On the contrary in thiol oxidation the reduction of
copper takes place easily, and the reoxidation of copper to divalent
state becomes the key stage. These processes take place easily for
polynuclear complexes [38,39].
RSH
[Cu+nCln(RS)n]
[Cu2+nCl2n
]
O2
H2O2
RSSR
The catalyst deactivaton is one of the key problems of the thiol
oxidation. The most active pyridinium and imidazolium-containing
systems, grafted on the Perlkat surface demonstrated high stability.
In fact, Table 3 contains the data on activity and stability of all com-
plexes, except the most active (CuCl2–EtPrImCl and CuCl2–PrPyCl
grafted on Perlkat). It was impossible to bring those catalysts to the
total loss of their activity. On that reason in the Table 3 instead
of real TON for those catalytic systems we give the number of
cycles that were performed without any loss of activity. This fact
made studying of the evolution of these complexes a rather com-
plicated task. The evolution of these systems was studied by 13C
NMR-MAS spectroscopy. The spectra of initial compounds cannot
be interpreted as they contain diamagnetic ion Cu2+. At that the
copper-containing intermediate can be studied by this method. In
the systems containing both imidazolium and pyridinium deriva-
tive at the excess of thiol the reduction of copper takes place. In
liquid phase dodecane-1-thiol (14.7; 23.4; 29.2; 30.1; 30.5; 30.7;
32.7; 34.6) and corresponding disulfide were found (14.1; 22.8;
28.7; 29.6; 31.9; 40.1). In both liquid and solid phase peaks of ILs
were observed (129.6; 145.3; 144.6; 64.3; 21.2; 14.9 in case of
propylpyridinium chloride) (123.3, 124.6; 136.4; 50.0; 34.8, 23.3,
22.1, 10.3, 9.7 in case of ethylptopylimidazolium chloride). Also the
spectrum of solid phase contained peaks, with their shifts close to
the shifts of thiol. We suggested, that they can be referred to cop-
per (I) thiolate (15.5; 26.6; 30.5, 34.8; 41.8). We have synthesized
it as described in [24] and convinced that this suggestion is right.
Thus we can state, that during the thiol oxidation ILs and chloride
anions are partially replaced in the coordination sphere of copper
with thiol. Taking into account, that after the reaction complexes on
Perlkat have the structure identical to the initial one (no changes in
the UV–vis spectrum of catalysts were observed after the reaction
takes place), we can state, that as the disulfide forms, it leaves the
coordination sphere of copper. No changes in the ILs structure dur-
ing that process were observed. This fact actually agrees the data on
high stability of the system, but unfortunately gives no information
on the ways of catalysts deactivation.
Fig. 10. The scheme of the mechanism of thiol oxidation coupling.
The best results were obtained on macroporous supports. For
the imidazole and triethylamine derivatives the highest decane
conversion was obtained on KSK. Pyridinium derivatives are active
on silochrome. This is most likely due to abnormally high L:M
ratio in the system. Indeed, the increase of the amount of donor
molecules in the coordination sphere of metal promotes the
reduction of Cu2+ and leads to isolation of complexes. Isolated
mononuclear complexes were shown to possess higher activity
in the reaction [21]. The catalytic activity of CuCl2–PrEt3NCl–SiO2
demonstrates this fact well. On the surface of KSK this catalyst
exhibits the highest activity among the discussed complexes, at
that this is the only mononuclear complex on this support (Fig. 3a,
ꢀmax = 430 nm). On other supports mononuclear complexes were
not found. The lowest activity was displayed by the complexes
on the Perlkat surface. This can be caused by two reasons: The
first one is the pore diameter. The Perlkat surface has the smallest
pores among all three used supports. This causes low availability
of some complex particles, located in narrow pores, and weakly
solvating ligands (CCl4 and alkanes) are unable to interact with
Cu2+. Another possible reason, which causes the activity decrease,
is the presence of strong Lewis acid – aluminium oxide – on the Per-
lkat surface (about 5%). Carbon tetrachloride is known to interact
with Lewis acids to form cations and cationic complexes, possess-
ing superacidic properties instead of radicals [36]. Those cationic
compounds can destroy the coordination sphere of a metal ion,
impeding the reduction of Cu2+ to Cu+. Also these superacidic par-
ticles interact with alkanes inefficiently unlike radicals. Another
important fact is the elimination of HCl from chlorodecane (if it
forms), which takes place easily in the presence of superacids. This
process leads to the formation of alkene and its further polymeriza-
tion. [36]. Also acidic particles are able to poison the catalyst [37].
This idea is confirmed by the fact that on Perlkat all the catalysts
display the lowest activity in CCl4 transformations.
To estimate the ways of catalysts deactivation, analogous sys-
tems on KSK surface were studied. We registered UV–vis spectra
of just obtained catalysts and of catalysts after loss of the activ-
ity. In case of imidazolium, the charge transfer band shifts to the
short wave-length region, while the d–d transfer band shifted to
the long-wave region. Such a change indicates that a significant
rearrangement of the molecule takes place and the fraction of
isolated molecules increases. As we have earlier stated, such com-
plexes are inactive in this reaction. The spectrum of complex with
pyridinium also changes. Both bands broaden greatly, so complexes
of various structure are formed on the surface and the activity
decreases.
4.4. The oxidation of thiols
The Table 3 data demonstrate that in the thiol oxidation the best
results are obtained on supports with smaller pores as opposed to
chlorination. This difference can be explained by the distinctions
in the reaction mechanisms. Thiols oxidation catalyzed by copper
complexes with the formation of hydrogen peroxide and disul-
fide is a many-electron process with the participation of bi- and
polynuclear structures [24,38]. As it has been previously shown, a
polynuclear mixed-ligand copper complex is the process interme-
diate [24,39] (Fig. 10).
Close location of metal ions on the surface favors the formation
of such structures. This conclusion is in good agreement with
the spectroscopic investigation result (Fig. 2b and c). Indeed,
the inactive complexes CuCl2–PrEt3NCl–silochrome/Perlkat
absorb light at 430 nm, which corresponds to isolated complexes
[25]. CuCl2–EtPrImCl–silochrome/Perlkat and CuCl2–PrPyCl–
silochrome/Perlkat absorb light in the long-wavelength region,
5. Conclusions
The results obtained in this work demonstrate that the copper
complexes with immobilized ionic liquids can be used in different
environmentally important processes. Two new radical reactions
that can be catalyzed by such complexes were found.

88
I. Tarkhanova et al. / Applied Catalysis A: General 470 (2014) 81–88
The CCl4 transformations are catalyzed by mononuclear copper
[15] X. Wang, H. Wan, M. Han, L. Gao, G. Guan, Ind. Eng. Chem. Res. 51 (2012)
3418–3424.
complexes with ILs on the basis of quaternary ammonium salts with
short chain alkylamines. Surface with wide pores should be used
to obtain the most active complexes.
Polynuclear copper complexes with ILs on the basis of alkylpyri-
dinium salts are the most active in thiols oxidation catalysis.
Supports with narrow pores are preferred for these catalysts.
[16] H. Hagiwara, Y. Sugawara, K. Isobe, T. Hoshi, T. Suzuki, Org. lett. 6 (2004)
2325–2328.
[17] T. Sasaki, M. Tada, C. Zhong, T. Kume, Y. Iwasawa, J. Mol. Catal. A: Chem. 279
(2008) 200–209.
[18] T. Sasaki, C. Zhong, M. Tada, Y. Iwasawa, Chem. Commun. 250 (2005) 6–2508.
[19] J.C. Bricker, R.R. Frame, Eur. Pat. #394751, 1989.
[20] A. Grünecker, H. Kilkendey, W. Stockmair, K. Schumann, P. Jakob, G. Berold. Ger.
Pat. # 3008284, 1998.
[21] E.N. Golubeva, D.N. Kharitonov, D.I. Kochubey, V.N. Ikorskii, J. Phys. Chem. A
113 (2009) 10219–10223.
Acknowledgements
[22] V.M. Zelikman, I.G. Tarkhanova, E.V. Khomyakova, Kinet. Catal. 53 (2012)
222–230.
[23] M.H. Valkenberg, C. de Castro, W.F. Holderich, Green Chem. 4 (2002) 88–93.
[24] I.G. Tarkhanova, M.G. Gantman, A.O. Chizhov, V.V. Smirnov, React. Kinet. Mech.
Catal. 101 (2010) 267–278.
[25] D.W. Smith, Coord. Chem. Rev. 21 (1976) 93–158.
[26] V.V. Smirnov, I.G. Tarkhanova, A.I. Kokorin, D.S. Tsvetkov, Kinet. Catal. 46 (2005)
861–866.
[27] R. Mukherjee, Comprehensive Coordination Chemistry II, vol. 6, Elsevier, 2003,
pp. 747–910.
The work was financially supported by the Russian Foundation
for the Basic Research and Darville Enterprises Limited. Mineral
support Perlkat was kindly provided by BASF–The Chemical Com-
pany.
References
[28] K. Matyjaszewski, T.E. Patten, J. Xia, J. Amer. Chem. Soc. 119 (1997)
674–680.
[29] K. Ohno, A.Q. Goto, T. Fukuda, J. Xia, K. Matyjaszewski, Macromolecules 31
(1998) 2699–2701.
[30] K. Matyjaszewski, Controlled/Living Radical Polymerization: Progress in ATRP,
NMP, and RAFT, ACS, Washington D.C., USA, 2000, pp. 486.
[31] J.M. Munoz-Molina, W.M.C. Sameera, E. Alvarez, F. Maseras, T.R. Belderrainand,
P.J. Perez, Inorg. Chem. 50 (2011) 2458–2467.
[32] V.V. Smirnov, M.M. Levitskii, I.G. Tarkhanova, S.M. Nevskaya, E.N. Golubeva,
Kinet. Catal. 42 (2001) 560–564.
[1] H. Olivier-Bourbigou, L. Magna, D. Morvan, Appl. Catal., A 373 (2010) 1–56.
[2] C. Van Doorslaer, J. Wahlen, P. Mertens, K. Binnemans, D. De Vos, Dalton Trans.
39 (2010) 8377–8388.
[3] H. Li, P.S. Bhadury, B. Song, S. Yang, R. Soc. Chem. Adv. 2 (2012) 12525–12551.
[4] C.P. Mehnert, R.A. Cook, N.C. Dispenziere, M. Aferwork, J. Amer. Chem. Soc. 124
(2002) 12932–12933.
[5] A. Riisager, R. Fehrmann, M. Haumann, P. Wasserscheid, Eur. J. Inorg. Chem. 4
(2006) 695–701.
[6] A. Riisager, P. Wasserscheid, V. Van Hal, R. Fehrmann, J. Catal. 219 (2003)
452–455.
[33] R.Kh. Freidlina, E.C. Chukovskaya, Synthese 5 (1974) 477–485.
[34] E.T. Denisov, I.B. Afanas’ev, Oxidation and Antioxidants in Organic Chemistry
and Biology, CRC Press, Taylor Francis group, Boca Raton, USA, 2005.
[35] A. Seidell, Solubilities of inorganic and organic compounds: a compilation of
quantitative solubility data from the periodical literature, vol. 2, third ed., Van
Nostrand, New York, 1940, pp. 194.
[36] I.S. Akhrem, A.V. Orlinkov, M. Vol’pin, J. Chem. Soc. Chem. Commun. 8 (1993)
671–672.
[37] I. Mochida, A. Uchino, H. Fujitsu, K. Takeshita, J. Catal. 43 (1976) 264–272.
[38] B.C. Gilbert, S. Silvester, P.H. Walton, J. Chem. Soc. Perkin Trans. 2 (6) (1999)
1115–1121.
[7] C.P. Mehnert, E.J. Mozeleski, R.A. Cook, Chem.Commun. 24 (2002) 3010–3011.
[8] P.M. Arvela, J.-P. Mikkola, P. Virtanen, H. Karhu, T. Salmi, D.Y. Mursin, in: E.M.
Gaineaux, et al. (Eds.), Scientific Bases for the Preparation of Heterogeneous
Catalysts, Elsevier B.V., 2006, pp. 87–94.
[9] X. Wu, Y.A. Letuchy, D.P. Eyman, J. Catal. 161 (1996) 164–177.
[10] A. Riisager, B. Jorgensen, P. Wasserscheid, R. Fehrmann, Chem. Commun. 9
(2006) 994–996.
[11] A. Riisager, R. Ferhmann, M. Haumann, P. Wasserscheid, Top. Catal. 40 (2006)
91–102.
[12] J. Joni, M. Haumann, P. Wasserscheid, Adv. Synth. Catal. 351 (2009) 423–431.
[13] S.V. Sirotin, I.F. Moskovskaya, Yu.G. Kolyagin, A.V. Yatsenko, B.V. Romanovsky,
Russ. J. Phys. Chem. A 85 (2011) 390–396.
[39] I.G. Tarkhanova, M.G. Gantman, T.N. Rostovschikova, Phosphorus Sulfur Silicon
Relat. Elem. 187 (2012) 88–100.
[40] J.K. Kochi, R.V. Subramanian, J. Amer. Chem. Soc. 87 (1965) 4856–4866.
[14] G. Noguera, J. Mostany, G. Agrifoglio, R. Dorta, Adv. Synth. Catal. 347 (2005)
231–234.