Phenol and dihydroxybenzene hydrogenation catalysts based on polyamide dendrimers and rhodium species

Article abstract of DOI:10.1134/S0965544114060115

Heterogeneous catalysts based on rhodium nanoparticles and first- and second-generation poly(amidoamine) (PAMAM) dendrimers crosslinked with hexamethylene diisocyanate have been synthesized. It has been found that catalyst samples with a particle size of 0.88 to 1.96 nm, depending on the PAMAM dendrimer generation, are effective in the hydrogenation of phenol, hydroquinone, resorcinol, and pyrocatechol (85°C, 30 atm O₂, 2 h). In phenol hydrogenation, the selectivity for cyclohexanone is 100%. Cyclohexanone has not undergone further transformation under the reaction conditions. The main products of dihydroxybenzene hydrogenation have been trans-cyclohexanediols, with their proportion decreasing in the order: resorcinol > hydroquinone > pyrocatechol. The selectivity has been found to be 100% for 1,3-cyclohexanediol, 97-99% for 1,4-cyclohexanediol, and 33-91% for 1,2-cyclohexanediol. The catalysts based on the second-generation dendrimer have shown a high activity in dihydroxybenzene hydrogenation (TOF of 6600 to 35000 h^-1).

Full text of DOI:10.1134/S0965544114060115

ISSN 0965ꢀ5441, Petroleum Chemistry, 2014, Vol. 54, No. 6, pp. 412–419. © Pleiades Publishing, Ltd., 2014.
Original Russian Text © E.M. Zakharyan, Gouqiung Ma, A.L. Maksimov, E.A. Karakhanov, Z.D. Voronina, 2014, published in Neftekhimiya, 2014, Vol. 54, No. 6, pp. 422–429.
Phenol and Dihydroxybenzene Hydrogenation Catalysts
Based on Polyamide Dendrimers and Rhodium Species
E. M. Zakharyana, Gouqiung Maa, A. L. Maksimova, E. A. Karakhanova, and Z. D. Voroninab
a Faculty of Chemistry, Moscow State University, Moscow, Russia
b Topchiev Institute of Petrochemical Synthesis, Russian Academy of Sciences, Moscow, Russia
eꢀmail: alenka.am@mail.ru
Received June 4, 2014
Abstract—Heterogeneous catalysts based on rhodium nanoparticles and firstꢀ and secondꢀgeneration
poly(amidoamine) (PAMAM) dendrimers crosslinked with hexamethylene diisocyanate have been syntheꢀ
sized. It has been found that catalyst samples with a particle size of 0.88 to 1.96 nm, depending on the
PAMAM dendrimer generation, are effective in the hydrogenation of phenol, hydroquinone, resorcinol, and
pyrocatechol (85°C, 30 atm O2, 2 h). In phenol hydrogenation, the selectivity for cyclohexanone is 100%.
Cyclohexanone has not undergone further transformation under the reaction conditions. The main products
of dihydroxybenzene hydrogenation have been transꢀcyclohexanediols, with their proportion decreasing in
the order: resorcinol > hydroquinone > pyrocatechol. The selectivity has been found to be 100% for 1,3ꢀ
cyclohexanediol, 97–99% for 1,4ꢀcyclohexanediol, and 33–91% for 1,2ꢀcyclohexanediol. The catalysts
based on the secondꢀgeneration dendrimer have shown a high activity in dihydroxybenzene hydrogenation
(TOF of 6600 to 35000 h–1).
Keywords: hydrogenation, phenol, cyclohexanone, rhodium nanoparticles, dendrimers, dihydroxꢀ
ybenzenes
DOI: 10.1134/S0965544114060115
One of the principal processes of petroleum chemꢀ latter with a salt of metal, such as Pt2+, Pd2+, Rh3+, and
istry is hydrogenation of aromatic compounds. Benꢀ the subsequent reduction with excess sodium borohyꢀ
zene hydrogenation gives cyclohexane [1], and phenol dride in water or alcohol (Scheme 1).
hydrogenation gives cyclohexanol and cyclohexꢀ
anone, which are intermediates for preparation of
caprolactam and adipic acid [2]. Phenol hydrogenaꢀ
tion on Pd [3], Pt [4], Ru [5], and Rh catalysts [6, 7]
was described. The main product was the mixture of
cyclohexanone and cyclohexanol. Hydrogenation is
accompanied by side hydrogenolysis processes to form
benzene and cyclohexane [8, 9]. The formation of
cyclohexenꢀ2ꢀone is also possible in some cases [10].
Only a few examples of for selective hydrogenation of
phenol to cyclohexanone are known [11–14].
This method was used to prepare effective Rhꢀconꢀ
taining catalyst for selective hydrogenation on the
basis of firstꢀ and secondꢀgeneration PAMAM denꢀ
drimers.
EXPERIMENTAL
As starting materials, the firstꢀ and secondꢀgeneraꢀ
tion
PAMAM(NH2)8) and G2 (PAMAM(NH2)16) synꢀ
thesized as described in [20] were used. 1,6ꢀHexameꢀ
poly(amidoamine)
dendrimers
G1
(
In this study, we investigated the hydrogenation of thylene diisocyanate (HMDI, Aldrich) was used as the
phenol and dihydroxybenzenes over rhodium catalysts crosslinking agent.
based on poly(amidoamine) (PAMAM) dendrimers as
an agent facilitating the stabilization of the resulting
species [15, 16].
Analysis
Fourierꢀtransform IR spectra were recorded on a
Thermo Scientific Nicolet IR2000 instrument in the
multiple frustrated total internal reflection mode using
a Multiꢀreflection HATR accessory containing a ZnSe
45° crystal for different wavelength ranges with a 4 nm
resolution.
We proposed a method for preparation of heterogeꢀ
neous catalysts based on metal nanoparticles incorpoꢀ
rated into crosslinked dendrimer matrices. Heterogeꢀ
neous supports were prepared on the basis of poly(proꢀ
pylenimine) dendrimers by their bonding to one
another with crosslinking agents, such as diisocyanates
[17–19]. The materials based on both metal nanoparꢀ
The samples obtained were studied by Xꢀray photoꢀ
ticles and dendrimers were prepared by treating the electron spectroscopy (XPS) using a Kratos Axis Ultra
412

PHENOL AND DIHYDROXYBENZENE HYDROGENATION CATALYSTS
413
H2N
H2N NH2
N
NH2
N
NH2
NH2
H2N N
H2N
NH2
NH2
NH2
NH2
N N
N N
H2N
H2N
N NH2
N
N
N
N N
N
N
N N
N NH2
NH2
H2N N
H2N
N
NH2
H2N
NH2
NH
2 H2N
G1 (PAMAM(NH2)8)
G2 (PAMAM(NH2)16)
H
N
NCO
N
OCN
N
NH2
NH2
≡
HMDI
O
Scheme 1. Preparation of nanoparticles encapsulated in a crosslinked dendrimer matrix.
DLD instrument equipped with an OPXꢀ150 retardꢀ C–N–H ), 1479, 1437, 1336 (CH2 ), 1257, 1124, 1086,
δ
δ
ingꢀfield electron energy analyzer. Photoelectrons 1047, 1005 (C–Nst in NHꢀC(ꢀO)).
were excited with Xꢀray radiation from an aluminum
anode (Al = 1486.6 eV) at a tube potential of 12 kV
XPS (eV): 284.2 (24%), 285.2 (50%), 286.6 (11%),
287.9 (16%), 289.2 (2%) (C 1s, 67.1%), 399.4 (55%),
400.6 (32%), 401.9 (13%) (N 1s, 21.5%), 530.9 (71%),
531.9 (23%), 533.1 (6%) (O 1s, 11.4%).
K
α
and an emission current of 20 mA. Photoelectron
peaks were calibrated with reference to the C1s carbon
line with the binding energy of 284.8 eV. Rhodium was
quantitatively determined with atomic adsorption
spectroscopy on an AAnalyst 400 instrument. A transꢀ
mission electron microscopy study was conducted
using a LEO912 AB OMEGA transmission electron
microscope.
Synthesis of G2ꢀHMDI. The process was conꢀ
ducted as described above. The reactants were
PAMAM(NH2)16 (1.064 g, 0.327 mmol) and hexameꢀ
thylene diisocyanate (0.5 mL, 3.28 mmol) in absolute
THF (50 mL) . The product was isolated to yield
1.064 g of a yellowish solid powder.
IR (cm–1): 3338 (N–Hst in NHꢀC(=O)), 2974 (C–
Hst), 2837 (C–Hst, CH2–Nst), 1637 (C=Ost in NHꢀ
Catalyst Preparation
C(=O)), 1587, 1568, 1537, 1518, Nꢀ1518 (N–H ,
δ
Synthesis of G1ꢀHMDI. A 1.019ꢀg portion of
PAMAM(NH2)8 (0.761 mmol) and waterꢀfree methaꢀ
nol (20 mL) were placed in a 100ꢀmL oneꢀneck flask
fitted with a magnetic stirrer and a reflux condenser.
The dendrimer dissolved almost immediately. Then
hexamethylene diisocyanate (0.5 mL, 3.169 mmol) in
absolute THF (20 mL) was added with stirring. The
CH2 , C–N–H ), 1469, 1367 (CH2 ), 1057, 1024
(C
δ
δ
δ
⎯
Nst in NHꢀC(ꢀO)).
XPS (eV): 284.6 (20%), 285.6 (58%), 286.8 (5%),
288.4 (17%), (C 1s, 65.1%), 399.4 (88%), 400.5
(12%), (N 1s, 23.0%), 531.1 (80%), 532.1 (18%),
533.6 (2%) (O 1s, 11.9%).
Synthesis of G1ꢀPAMAMꢀ1. PAMAMꢀG1ꢀ
HMDI (1/1) (1200 mg) and ethanol (20 mL) were
placed in a 50ꢀmL flask fitted with a magnetic stirrer
reaction was conducted for 12 h at 70
reaction mixture was evaporated on a rotary evaporaꢀ
tor at 50 . The product was isolated to yield 1.184 g
of a white solid powder.
°C, then the
°C
and a reflux condenser. A solution of 30 mg RhCl3
⋅
4H2O (0.09 mmol) in 10 mL of distilled water was
IR (cm–1): 3330 (N–Hstin NHꢀC(=O)), 2933 added to the resulting suspension. The reaction was
(C–Hst), 2850 (C–Hst, CH2–Nst), 1626, 1585 (C=Ost conducted for 24 h at room temperature. Then,
in NHꢀC(=O)), 1554, 1512, Nꢀ1512 (N–H , CH2 , sodium borohydride (150 mg, 3.96 mmol) was added
δ
δ
PETROLEUM CHEMISTRY Vol. 54
No. 6
2014

414
ZAKHARYAN et al.
Table 1. Amounts of reactants in catalyst syntheses
Catalyst
G1ꢀPAMAMꢀ1
Starting material
Solvent
Catalyst yield, mg
G1ꢀHMDI (1200 mg)
G1ꢀHMDI (300 mg)
G1ꢀHMDI (505 mg)
G2ꢀHMDI (1709 mg)
G2ꢀHMDI (500 mg)
EtOH
648
230
G1ꢀPAMAMꢀ2
G1ꢀPAMAMꢀ3
G2ꢀPAMAMꢀ4
G2ꢀPAMAMꢀ5
MeOH + CHCl3
Toluene
125
EtOH
1049
222
EtOH + MeOH + CHCl3
with stirring. The reaction mixture darkened, and ically sealed, filled with hydrogen to 5, 15, or 30 atm,
effervescence was observed. The reaction was run for and the thermostat was switched on. The reaction was
12 h at 60°C. After completion of the reaction, the susꢀ run with intense agitation at 85°C for 15 min and 2 h;
pension was evaporated to dryness on a rotary evapoꢀ then, the reactor was cooled to below room temperaꢀ
rator at an elevated temperature. The resulting precipꢀ ture. The products were analyzed by GLC using a
itate was washed two times with water and two times ChromPack CP9001 gas chromatograph equipped
with methanol and dried in air at elevated temperaꢀ with a flameꢀionization detector and a 30 m
×
0.2 mm
ture. The yield of the product, a gray powder, was column with the bonded SEꢀ30 phase. Chromatoꢀ
646 mg.
The G1ꢀPAMAMꢀ2, G1ꢀPAMAMꢀ3, G2ꢀ
grams were recorded and processed with a computer
using the Maestro 1.4 software.
PAMAMꢀ4, and G2ꢀPAMAMꢀ5 catalysts were preꢀ
pared in the same way, the amounts of reactants and
the catalyst yields are given in Table 1.
Thus, the catalysts based on PAMAM dendrimers
of the first and second generations were prepared using
three different solvents: ethanol, methanol + chloroꢀ
form, and toluene.
RESULTS AND DISCUSSION
Polymer network matrices were prepared from the
1stꢀ and 2ndꢀgeneration poly(amidoamine) dendrimꢀ
ers PAMAM(NH2)8 and PAMAM(NH2)16 (Fig. 1).
The crosslinker was 1,6ꢀhexamethylene diisocyanate
OCN(CH2)6NCO (HMDI), a compound that forms
conformationally mobile bonds between the dendrimꢀ
ers. The obtained IR and XPS data show that the proꢀ
cess follows the conventional mechanism of the reacꢀ
tion of an isocyanate with an amine to form the –NHꢀ
C(=O)ꢀNHꢀ linkage.
Metal nanoparticles were encapsulated in denꢀ
drimers according to the procedure proposed in [18,
19, 21, 22]. Solvents of different polarities were used
for metal deposition.
Catalyst Testing
A were placed in a thermostatted steel autoclave
equipped with an insert vial and a magnetic stirrer was
charged with the catalyst and the substrate to have
4 mg of catalyst per 1 mL of substrate. The solvent was
added whenever necessary. The autoclave was hermetꢀ
The obtained materials were characterized by IR,
XPS, TEM, and AAS methods. The metal content for
each of the materials was determined by the AAS
method (Table 2).
The IR spectra of the crosslinked dendrimers and
the catalysts on their basis display vibration bands of
N–H bonds (3330 cm–1), methylene groups (2940,
2855 cm–1), and urea and amide carbonyl groups
(1684, 1645, 1622 cm–1), with this signal being due to
the linear moieties.
Position Fraction, %
26.0
284.6
285.6
286.9
288.3
19
61
4
20.8
15.8
10.4
52
16
According to the XPS data, the spectrum of C1s
electrons consists of four components (Fig. 1) suggestꢀ
ing the presence of the CH2CH2 (
CH2CH2NH ( b = 285.5 eV [23]), CH2NHC(O)NH
Eb = 286.8 eV [23]), and CH2CH2C(O)NH (Eb
288.1 eV [23]) moieties. The binding energies in all the
samples are practically equal to one another, and their
integral intensities differ insignificantly (Table 2).
Eb = 284.9 eV [23]),
0
292 291 290 289 288 287 286 285 284 283 282 281 280
E
(
=
Binding energy, eV
Fig. 1. Spectrum of C1s electrons for G1ꢀPAMAMꢀ3.
PETROLEUM CHEMISTRY Vol. 54
No. 6
2014

PHENOL AND DIHYDROXYBENZENE HYDROGENATION CATALYSTS
415
Table 2. Physicochemical characteristics of dendrimerꢀbased supports
XPES
AAS
TEM
, nm
Catalyst
DM
atomic conꢀ
binding energy (eV) and intensity (%)
Rh, %
d
centration, %
C 1s, 66.3
284.8 (30%), 285.8 (50%), 287.3 (4%), 288.5 (16%)
Rh 3d5/2, 0.08 309.7
0.88 0.003 (98%),
1.96 0.006 (2%)
G1ꢀPAMAMꢀ1
0.43
1.12
1.77
0.33
0.52
0.99
0.83
0.65
0.69
0.76
N 1s, 19.3
O 1s, 12.3
C 1s, 70.7
399.5 (89%), 400.5 (11%)
531.2 (74%), 532.4 (19), 533.4 (7%)
284.8 (37%), 285.7 (48%), 287.1 (3%), 288.4 (12%)
Rh 3d5/2, 0.09 309.5
0.90 0.02 (65%),
1.43 0.04 (35%)
G1ꢀPAMAMꢀ2
G1ꢀPAMAMꢀ3
G2ꢀPAMAMꢀ4
G2ꢀPAMAMꢀ5
N 1s, 17.6
O 1s, 11.7
C 1s, 66.1
399.5 (84%), 400.3 (16%)
531.2 (66%), 532.1 (28%), 533.4 (6%)
284.6 (19%), 285.6 (61%), 286.9 (4%), 288.3 (16%)
0.92 0.04 (39%),
1.65 0.04 (61%)
Rh 3d5/2, 0.21 309.5
N 1s, 21.1
O 1s, 12.6
C 1s, 58.1
N 1s, 14.0
O 1s, 10.9
N 1s, 14.0
C 1s, 69.0
399.5 (92%), 400.6 (8%)
531.0 (80%), 532.1 (17%), 533.4 (3%)
284.5 (27%), 285.6 (47%), 287.4 (6%), 288.5 (20%)
399.5 (92%), 401.4 (8%)
0.94 0.02 (62%),
1.90 0.04 (38%)
531.4 (64%), 533.0 (27%), 534.1 (9%)
399.5 (92%), 401.4 (8%)
284.5 (46%), 285.6 (39%), 288.1 (15%)
Rh 3d5/2, 0.03 309.6
1.04 0.024 (70%),
1.52 0.027 (30%)
N 1s, 13.6
O 1s, 8.9
399.5 (73%), 400.5 (23%), 402.0 (4%)
531.2 (60%), 532.4 (31%), 533.8 (9%)
The signal at 531.2 eV in the spectrum of O1s elecꢀ is considerably higher than that for the metal and oxiꢀ
trons confirms the presence of the amide carbonyl dized rhodium (Eb (Rh) = 307.2, Eb (Rh2O3) =
bond (
firms the presence of the isocyanate group (Eb
532.0 eV [23]), and the signal of 533.4 eV shows the corresponds to partially oxidized rhodium atoms Rh
E
b = 531.5 eV [23]), the signal of 532.1 eV conꢀ 309.2 eV [23]) and is slightly below that for trivalent
=
rhodium (Eb (Rh3+) = 310.0 eV [23]). Obviously, it
δ
+
Rh–O bond (
Eb = 533.0 eV [23]) (Fig. 2).
(1 <
δ
< 3) coordinated to nitrogen or oxygen atoms
[24]. The formation of Rh3+ suggests incomplete
In the spectrum of N1s electrons (Fig. 3) consisting
of two components; the signal at 399.5 eV indicates the
presence of the C–N bond of the CH2СH2NH2 and
CH2СH2C(O)NH fragments, and that at 400.6 eV in
the СH2NH2/Rh fragment. The energy shifts for oxyꢀ
gen and nitrogen relative to the conventional values
(531.5 eV and 398.6 eV, respectively [23]) in the specꢀ
tra of the samples are caused by partial electron transꢀ
fer onto neutral metal particles.
reduction of Rh + cations, which form complexes with
δ
the internal amino groups of PAMAM dendrimers
[25, 26].
It has been found that the amount of the metal
deposited on the polymer depends on both the denꢀ
drimer generation and the solvent used for metal depꢀ
osition. The amount of the metal adsorbed on the supꢀ
port and, hence, the total metal content in the samples
For all the samples, except G2ꢀPAMAMꢀ4, for increases with an increase in the fraction of the nonꢀ
which the rhodium line overlaps the C1s line, the Rh polar solvent,. Under identical conditions, the metal
3d5/2 binding energy for is 309.6 eV (Fig. 4). This value content decreases with the dendrimer generation.
PETROLEUM CHEMISTRY Vol. 54
No. 6
2014

416
ZAKHARYAN et al.
Position Fraction, %
The proportion of large particles also increases with
the dendrimer generation.
29.0
24.2
19.4
14.6
9.8
531.0
532.1
533.4
80
17
3
The catalysts obtained showed a high selectivity in
the phenol hydrogenation to cyclohexanone. Rhodꢀ
ium catalysts usually mediate hydrogenation either to
an alcohol and ketone mixture [6, 7, 27] or selectively
to the alcohol [28]. There is the only report on phenol
hydrogenation to ketone over rhodium nanoparticles
in the presence of potassium tertꢀbutoxide and nitroꢀ
genꢀcontaining ligands, with the activity of the cataꢀ
lytic system being low (TOF = 30 h–1) [13].
The PAMAMꢀbased catalysts obtained in this
study were found to be selective for cyclohexanone at
a substantially higher activity above that attained for
rhodium nanoparticles mediating hydrogenation to
the ketone (Rh/C) [13] and comparable with the
activity of heterogeneous rhodium catalysts that mediꢀ
ate nonselective hydrogenation [29]. This can be
explained by ready desorption of the intermediate enol
owing to the formation of additional hydrogen bonds
with the functional groups of the support. A compariꢀ
son of the catalysts based on firstꢀ and secondꢀgeneraꢀ
tion dendrimers dhows that the conversion is proporꢀ
tional to the rhodium content of the sample. The maxꢀ
imum conversion among the firstꢀgeneration samples
was achieved on the G1ꢀPAMAMꢀ1 (1.77% Rh) cataꢀ
lyst, and that between the second generation denꢀ
drimer catalysts was on G2ꢀPAMAMꢀ5 (Table 3).
Note that cyclohexanone is not hydrogenated under
the reaction conditions.
5.0
537 536 535 534 533 532 531 530 529 528 527 526525
Binding energy, eV
Fig. 2. Spectrum of O1s electrons for G1ꢀPAMAMꢀ3.
It has been found from the TEM data that in the
case of the composites based on rhodium nanopartiꢀ
cles, the particle size and the distribution mode both
are affected by the dendrimer generation (Figs. 5–9).
All of the samples are characterized by bimodal disꢀ
tribution, which can be associated with the presence in
the dendritic matrix structure of two types of sites that
bond metal nanoparticles. It may be assumed that the
particles of up to 1.3 nm in size are accommodated
between the branches and the particles of 1.5–2 nm in
size locate are in space between crosslinked dendrimꢀ
ers. As seen from the data in Table 2, particles of a
smaller size up to 1.04 nm prevail in the samples other
than G1ꢀPAMAMꢀ3 and make 98% of total particles
in the G1ꢀPAMAMꢀ1 sample. As the rhodium perꢀ
centage in the sample based on the firstꢀgeneration
The catalyst activity (TOF, turnover frequency) was
calculated as the number of moles of the substrate
converted per mole of rhodium per unit time. The valꢀ
ues of TOFs were calculated as the number of moles of
the substrate converted per mole of rhodium on the
surface per unit time [30, 31]. The most active catalyst
dendrimer increases, the proportion of larger particles based on the crosslinked firstꢀgeneration dendrimer
was G1ꢀPAMAMꢀ1 (TOF = 1900 h–1 and TOFs
=
also increases (to 61% in the case of G1ꢀPAMAMꢀ3).
9.0
8.5
8.0
7.5
7.0
6.5
30.0
Position Fraction, %
399.5
400.6
92
8
24.8
19.6
14.4
9.2
4.0
321
318
315
312
309
306
405 404 403 402 401 400 399 398 397 396 395
Binding energy, eV
Binding energy, eV
Fig. 3. Spectrum of N1s electrons for G1ꢀPAMAMꢀ3.
Fig. 4. Spectrum of Rh 3d electrons for G1ꢀPAMAMꢀ3.
PETROLEUM CHEMISTRY Vol. 54 No. 6 2014

PHENOL AND DIHYDROXYBENZENE HYDROGENATION CATALYSTS
417
Number of particles
1200
Number of particles
350
300
250
200
150
100
50
1000
800
600
400
200
0
0.5 1.0 1.5 2.0 2.5
d, nm
0
0.5 1.0 1.5 2.0 2.5
, nm
d
100 nm
50 nm
Fig. 5. TEM data and particle size distribution of G1ꢀ
PAMAMꢀ1.
Fig. 6. TEM data and particle size distribution of G1ꢀ
PAMAMꢀ2.
Number of particles
1200
1000
800
600
400
200
0
0.5 1.0 1.5 2.0 2.5
, nm
d
Number of particles
300
250
200
150
100
50
0
0.5
1.0
1.5
2.0
2.5
, nm
50 nm
50 nm
d
Fig. 7. TEM data and particle size distribution of G1ꢀ
PAMAMꢀ3.
Fig. 8. TEM data and particle size distribution of G2ꢀ
PAMAMꢀ4.
1900 h–1, and G2ꢀPAMAMꢀ5 (TOF = 2300 h–1 and size of about 1 nm. As the particle size increases, the
TOFs = 3000 h–1) was the most active of the specific activity per active center decreases. Note that
crosslinked secondꢀgeneration dendrimer catalysts. the difference in TOFs between the most active cataꢀ
The maximum activity was shown by the highꢀdisperꢀ lysts of first (G1ꢀPAMAMꢀ1) and second (G2ꢀ
sity catalysts in which the majority of particle had a PAMAMꢀ5) generations can be explained by lower
PETROLEUM CHEMISTRY Vol. 54
No. 6
2014

418
ZAKHARYAN et al.
Number of particles
500
pyrocatechol, whereas the cisꢀproducts prevailed in
[5] with their relative amount decreasing in the order:
hydroquinone > resorcinol > pyrocatechol. For examꢀ
ple, the proportion of the transꢀisomer in hydrogenaꢀ
tion over the rhodium catalysts based on PAMAM
dendrimers was 100% for 1,3ꢀcyclohexanediol, 97–
100% for 1,4ꢀcyclohexanediol, or 33–91% for 1,2ꢀ
cyclohexanediol.
400
300
200
100
0
0.5 1.0 1.5 2.0 2.5
, nm
In hydroquinone hydrogenation, the conversion to
1,4ꢀcyclohexanediol also depended on the rhodium
content of the catalyst (Table 4), and the maximum
conversion was 53% for the G1ꢀPAMAMꢀ3. This can
be due to both the amount of the metal (both total and
surface metal) and the particle size.
d
Note that the specific activity grows for the cataꢀ
lysts based on the secondꢀgeneration dendrimers and
drops for the catalysts based on the firstꢀgeneration
dendrimers with an increase in the particle size.
Under the same conditions, resorcinol was hydroꢀ
genated to 1,3ꢀtransꢀcyclohexanediol with 100% yield
and 100% selectivity (Scheme 2, Table 4). It likely that
for these catalysts, there is an optimal substrate size
corresponding to the distance between branches of the
dendrite matrix. In fact, almost 100% conversion is
attained even in the reaction run for 15 min. The G2ꢀ
PAMAMꢀ4 catalyst showed a high activity of TOF =
35500 h–1or TOFs = 51400 h–1. As in the case of hydꢀ
roquinone hydrogenation, the samples based on the
secondꢀgeneration dendrimer were more active than
those based on the firstꢀgeneration dendrimer.
100 nm
Fig. 9. TEM data and particle size distribution of G2ꢀ
PAMAMꢀ5.
diffusion constraints in the catalyst based on the latter
dendrimer.
The catalysts were also examined in the hydrogenaꢀ
tion of dihydric phenols: hydroquinone, resorcinol,
and pyrocatechol. Unlike the ruthenium catalysts
Ru/C in [5] on which a high conversion was attained
owing to the reaction time of 24 h and a pressure of
5 atm, the Rh catalysts based on PAMAM dendrimers
showed a high activity in the hydrogenation of subꢀ
strates for 2 h at a pressure of 30 atm.
Trace amounts of phenol, benzene, and cyclohexꢀ
ane were found as products, indicating the occurrence
of possible C–O bond hydrogenolysis reactions [8, 9].
In comparison with the previous substrates, pyroꢀ
catechol is characterized by a very low rate of hydrogeꢀ
nation to the mixture of isomeric products: transꢀ and
cisꢀ1,2ꢀcyclohexanediols. This can be explained by the
high affinity of rhodium for the oxygen of the substrate
OH groups.
Note that the main hydrogenation products in our
case were the transꢀisomers, with their proportion
decreasing in the order: resorcinol > hydroquinone >
The highest conversion (13%) was attained on the
G1ꢀPAMAMꢀ3 sample containing 1.77% rhodium.
Table 3. Phenol hydrogenation in the presence of Rh cataꢀ
lysts based on dendrimers
The principal hydrogenation product was the trans
ꢀ
isomer. As the conversion increased, the trans to cisꢀ
isomer ratio decreased from 10 : 1 to 3 : 1. The catalyst
activity drops in the order: G1ꢀPAMAMꢀ3 > G1ꢀ
PAMAMꢀ2 > G2ꢀPAMAMꢀ5 > G1ꢀPAMAMꢀ1 >
G1ꢀPAMAMꢀ4. The catalyst with the smallest metal
content showed the lowest activity.
Catalyst
Conversion, % TOF, h–1 TOFs, h–1
G1ꢀPAMAMꢀ1
G1ꢀPAMAMꢀ2
G1ꢀPAMAMꢀ3
G2ꢀPAMAMꢀ4
G2ꢀPAMAMꢀ5
25
42
70
17
36
1907
1230
1297
1690
2271
1926
1482
1996
2449
2989
In summary, the catalysts based on both PAMAM
dendrimers and rhodium nanoparticles have syntheꢀ
sized and examined in the hydrogenation of phenol
and dihydroxybenzenes. Phenol hydrogenation over
the catalysts based on PAMAM dendrimers yields
cyclohexanone with 100% selectivity. It is noteworthy
that the main products in the hydrogenation of paraꢀ,
metaꢀ, and orthoꢀdihydroxybenzenes were respective
transꢀcyclohexanediols. The maximum activity was
Reaction conditions: substrate (2.7 mmol), catalyst (5 mg),
distilled water (250 L), 2 h, 85 C, 30 atm H .
µ
°
2
PETROLEUM CHEMISTRY Vol. 54
No. 6
2014

PHENOL AND DIHYDROXYBENZENE HYDROGENATION CATALYSTS
419
Table 4. Hydroquinone and resorcinol hydrogenation in the presence of Rh catalysts based on dendrimers
Hydroquinone
TOF, h–1
Resorcinol
TOF, h–1
Catalyst
conversion, %
TOFs, h–1
conversion, %
TOFs, h–1
G1ꢀPAMAMꢀ1
G1ꢀPAMAMꢀ2
G1ꢀPAMAMꢀ3
G2ꢀPAMAMꢀ4
G2ꢀPAMAMꢀ5
31
51
53
27
33
8439
5328
3503
9573
7421
8520
6419
100
99
27209
10363
6610
27484
12486
10169
51383
29487
5389
100
100
100
13874
9770
35455
22410
Reaction conditions: substrate (2.3 mmol), catalyst (1 mg), distilled water (250
µ
L), 2 h, 85°C, 30 atm H .
2
14. H. Liu, T. Jiang, B. Han, et al., Sience 326, 1250
observed in the hydrogenation of resorcinol (TOF =
35000 h–1).
(2009).
15. B. Devarakonda, R. A. Hill, and M. de Villiers, Int. J.
Pharm. 284, 133 (2004).
ACKNOWLEDMENTS
16. W. Soichiro and I. Michiko, J. Photochem. Photobiol.
A: Chem. 155, 57 (2003).
This work was supported in part by the Russian
Foundation for Basic Research, project nos. 12ꢀ03ꢀ
00325A and 14ꢀ03ꢀ3153.
17. E. A. Karakhanov, A. L. Maximov, V. A. Skorkin, et al.,
Pure Appl. Chem. 81, 2013 (2009).
18. E. A. Karakhanov, A. L. Maksimov, A. V. Zolotukhina,
and S. V. Kardashev, Pet. Chem. 50, 290 (2010).
REFERENCES
19. E. A. Karakhanov, A. L. Maksimov, A. V. Zolotukhina,
1. C. Vangelis, A. Bouriazos, S. Sotiriou, et al., J. Catal.
et al., Pet. Chem. 52, 289 (2012).
274, 21 (2010).
20. D. A. Tomalia, D. A. Kaplan, W. J. Kruper, et al., US
Patent No. 5 728 461 (1994).
2. J. Ritz, H. Fuchs, H. Kieczka, and W. C. Moran, in Ullꢀ
mann’s Encyclopedia of Industrial Chemistry, (Wiley–
VCH, New York, 2011), 7th Ed., p. 1.
21. Y. Borodko, C. M. Thompson, W. Huang, et al.,
J. Phys. Chem. C 115, 4757 (2011).
3. E. Diaz, A. F. Mohedano, L. Calvo, et al., Chem. Eng.
J. 131, 65 (2007).
22. R. M. Crooks, M. Zhao, L. Sun, et al., Acc. Chem.
Res. 34, 181 (2001).
4. Z. Guo, L. Hu, H.ꢀH. Yu, et al., R. Soc. Chem. Adv.
3477 (2012).
2,
23. NIST Xꢀray Photoelectron Spectroscopy Database
,
http://srdata.nist.gov/xps
5. T. Maegawa, A. Akashi, K. Yaguchi, et al., Chem. Eur.
J. 15, 6953 (2009).
24. Z. Nazarpoor, K. Khivantsev, E. Kyriakidou, et al.,
J. Colloid Interface Sci. 398, 22 (2013).
6. E. A. Karakhanov, A. G. Dedov, A. S. Loktev, and
V. S. Pshezhetskii, USSR Inventor’s Certificate
No. 1002292 (1983): Byull. Izobret., No. 9 (1983).
25. W. Huang, J. N. Kuhn, C.ꢀK. Tsung, et al., Nano Letꢀ
ters 8, 2027 (2008).
7. E. A. Karakhanov, V. S. Pshezhetskii, A. G. Dedov,
26. A. Siani, O. S. Alexeev, D. S. Deutsch, et al., J. Catal.
et al., Dokl. Akad. Nauk SSSR 275, 1098 (1984).
266, 331 (2009).
8. C. S. Chin, B. Lee, J. Moon, et al., Bull. Korean Chem.
27. S. Fujita, T. Yamada, Y. Akiyama, et al., J. Supercrit.
Soc. 16, 528 (1995).
Fluids 54, 190 (2010).
9. H. Ohde, M. Ohde, and C. M. Wai, Chem. Commun.
28. D.ꢀW. Wang, S.ꢀM. Lu, and Y.ꢀG. Zhou, Tetrahedron
8, 930 (2004).
Lett. 50, 1282 (2009).
10. K. Mori, K. Furubayashi, S. Okada, and H. Yamashita,
29. H. Wang, F. Zhao, S.ꢀI. Fujita, and M. Arai, Catal.
Chem. Commun. 48, 8886 (2012).
Commun.
9, 362 (2008).
11. T. Maegawa, A. Akashi, K. Yaguchi, et al., Chem. Eur.
30. A. Mastalir, Z. Kiraly, and F. Berger, Appl. Catal. A 269
,
J. 15, 6953 (2009).
161 (2004).
12. X. Yang, S. Liao, J. Zeng, and Z. Liang, Appl. Surf. Sci.
31. J. R. Anderson, Structure of Metallic Catalysts (Acaꢀ
257, 4472 (2011).
demic Press, New York, 1975).
13. M. L. Buil, M. A. Esteruelas, S. Niembro, et al., Orgaꢀ
nometallics 29, 4375 (2010).
Translated by S. Lebedev
PETROLEUM CHEMISTRY Vol. 54
No. 6
2014