Effect of the State of a Surface Layer on the Properties of Pd–P Catalysts in the Hydrogenation of Alkylanthraquinones

Article abstract of DOI:10.1134/S0023158419060028

Abstract: X-ray photoelectron spectroscopy (XPS) was used to study the state of the surface layer of a Pd–P catalyst, which exhibits high selectivity in the hydrogenation of 2-ethyl-9,10-anthraquinone. The preparation of the catalyst from Pd(acac)₂ and white phosphorus in hydrogen in a toluene–1-octanol solution led to the formation of Pd–P nanoparticles whose surface was enriched in electron-deficient palladium, apparently, as a constituent of an amorphized 2D layer based on a solid solution of Pd and P. Octyl esters of phosphoric acids, adsorbed on the surface, acted as stabilizers. A relationship between the properties of the Pd–P catalyst in the hydrogenation of 2-ethyl-9,10-anthraquinone and the state of the surface layer was considered.

Full text of DOI:10.1134/S0023158419060028

ISSN 0023-1584, Kinetics and Catalysis, 2019, Vol. 60, No. 6, pp. 808–817. © Pleiades Publishing, Ltd., 2019.
Russian Text © The Author(s), 2019, published in Kinetika i Kataliz, 2019, Vol. 60, No. 6, pp. 788–796.
Effect of the State of a Surface Layer on the Properties
of Pd–P Catalysts in the Hydrogenation of Alkylanthraquinones
L. B. Belykha, *, T. P. Sterenchuka, N. I. Skripova, V. V. Akimovb, V. L. Tausonb, A. S. Romanchenkoc,
K. L. Gvozdovskayaa, S. B. Sanzhievaa, and F. K. Shmidta
aIrkutsk State University, Irkutsk, 664003 Russia
bVinogradov Institute of Geochemistry, Siberian Branch, Russian Academy of Sciences, Irkutsk, 664033 Russia
cInstitute of Chemistry and Chemical Technology, Krasnoyarsk Scientific Center, Siberian Branch,
Russian Academy of Sciences, Krasnoyarsk, 660036 Russia
* e-mail: belykh@chem.isu.ru
Received February 15, 2019; revised April 22, 2019; accepted June 7, 2019
Abstract—X-ray photoelectron spectroscopy (XPS) was used to study the state of the surface layer of a Pd–P
catalyst, which exhibits high selectivity in the hydrogenation of 2-ethyl-9,10-anthraquinone. The preparation
of the catalyst from Pd(acac)2 and white phosphorus in hydrogen in a toluene–1-octanol solution led to the
formation of Pd–P nanoparticles whose surface was enriched in electron-deficient palladium, apparently, as
a constituent of an amorphized 2D layer based on a solid solution of Pd and P. Octyl esters of phosphoric
acids, adsorbed on the surface, acted as stabilizers. A relationship between the properties of the Pd–P catalyst
in the hydrogenation of 2-ethyl-9,10-anthraquinone and the state of the surface layer was considered.
Keywords: hydrogenation, 2-ethyl-9,10-anthraquinone, palladium–phosphorus catalysts, selectivity, XPS
DOI: 10.1134/S0023158419060028
INTRODUCTION
Chemoselective hydrogenation of alkylanthraqui-
nones to alkylanthrahydroquinones and the subse- hydrogen peroxide [1]
quent oxidation of alkylanthraquinones with oxygen is
the main industrial method for the production of
O
OH
OH
O
R
R
R
H2
Pd
O2
H2O2
+
O
O
Alkilantraquinones
Alkilantrahydroquinones
In this case, the hydrogenation of alkylanthraqui- genation of alkylanthraquinone is not carried out until
nones is invariably accompanied by side reactions,
which decrease the yield of the target product. These
side reactions primarily include the hydrogenation of
aromatic rings and the hydrogenolysis of C–OH
bonds in the resulting alkylanthrahydroquinone [2].
Various approaches are used to increase the yield of
hydrogen peroxide. In order to reduce the contribu-
its complete conversion, and the conversion of
alkylanthraquinone usually does not exceed 60% [2].
In this regard, it is reasonable to develop new highly
selective catalysts that reduce only the carbonyl groups
of alkylanthraquinones without accelerating the
hydrogenolysis of C–OH bonds in alkylanthrahydro-
tion of side processes in industry, the catalytic hydro- quinones and without affecting the aromatic rings.
Abbreviations: CSR, coherent scattering region; HR TEM, high-resolution transmission electron microscopy; XRD, X-ray diffrac-
tion; FWHM, full width at half maximum of spectral lines; EDX, energy-dispersive X-ray spectroscopy.
808

EFFECT OF THE STATE OF A SURFACE LAYER ON THE PROPERTIES
809
phorus in benzene (1.5 × 10–5 mol in terms of the
atomic form of phosphorus) was added dropwise in a
stream of hydrogen, and the contents were stirred at
room temperature for 5 min. Then, 1-octanol (10 mL)
was added; the temperature was raised to 90°C, and
the catalyst was formed with vigorous stirring of the
reaction mixture in hydrogen for 30–45 min until the
quantitative conversion of Pd(acac)2. The conversion
of Pd(acac)2 was monitored by UV spectroscopy using
Supported palladium catalysts are predominantly
used as catalysts for the hydrogenation of alkylanthra-
quinones, and their selectivity is changed by adding
promoters [3] or by modifying the support surface [4,
5]. Colloidal solutions of nanoparticles are the most
convenient models for studying catalytic systems and
optimizing their properties. Previously, Sterenchuk et
al. [6] used the colloidal solutions of Pd as an example
to found that the catalyst selectivity in the hydrogena-
tion of 2-ethyl-9,10-anthraquinone cannot be
increased only by changing the particle size of palla-
dium. Along with the hydrogenation of C=O groups in
2-ethyl-9,10-anthraquinone, large star-like palladium
crystallites (~d = 127 nm) accelerated predominantly
the hydrogenolysis of C–OH bonds, while small pal-
ladium nanoclusters (d = 1.5–2.5 nm) catalyzed the
hydrogenation of aromatic rings in 2-ethyl-9,10-anth-
rahydroquinone. Modification of the palladium cata-
lyst with phosphorus made it possible to increase the
yield of hydrogen peroxide from 69 to 96–98% [6].
an absorption band at 330 nm (ε330 = 10630 L cm–1
mol–1). The resulting black-brown colloidal solution
was cooled to 30°C. A drop of the Pd–P catalyst solu-
tion was applied onto carbonized copper gauze (200
mesh) and dried at room temperature in a box with an
inert atmosphere.
Method 2. A Pd–P catalyst for studying by X-ray
photoelectron spectroscopy (XPS) and X-ray diffrac-
tion (XRD) analysis was prepared as follows: To a
solution of Pd(acac)2 (0.4566 g, 1.55 × 10–3 mol) in
68 mL of toluene placed in a thermostatically con-
trolled long-neck flask, 3.1 mL of a solution of phos-
phorus in benzene (4.5 × 10–4 mol in terms of the
atomic form of phosphorus) was added dropwise in a
stream of hydrogen, and the contents were stirred at
room temperature for 5 min. Then, 1-octanol (44.5
mL) was added; the temperature was raised to 90°C,
and the catalyst was formed with vigorous stirring of
the reaction mixture in hydrogen for 30–45 min until
the quantitative conversion of Pd(acac)2. At the end of
the reaction, the catalyst system formed was cooled to
room temperature and transferred into a finger-type
vessel in an inert atmosphere. Solvents (2/3 of the vol-
ume) were distilled off in a vacuum, and diethyl ether
was added until the formation of a black precipitate.
The precipitate was sequentially washed with benzene
and diethyl ether in an atmosphere of argon and dried
in a vacuum (50°C/1 Torr). The yield was 0.1090 g.
In this work, we studied the phase composition and
the state of a surface layer of the Pd–P catalyst in order
to determine the causes of the modifying effect of
phosphorus on the properties of palladium catalysts
for hydrogenation.
EXPERIMENTAL
Materials
Solvents (benzene, toluene, and 1-octanol) were
purified in accordance with standard procedures [7].
For deeper drying, benzene and toluene were addi-
tionally distilled from LiAlH4 using a fractionating
column. 1-Octanol was twice distilled (conventional
distillation and distillation from sodium octylate) in a
vacuum (2 Torr) at 59°C. All of the solvents were
stored in an atmosphere of argon in sealed ampoules
over 4A molecular sieves.
Bis(acetylacetonato)palladium(II) Pd(acac)2 was
prepared according to a published procedure [8] with
the subsequent recrystallization from acetone.
Research Methods
White phosphorus was mechanically cleaned of
surface oxidation products and washed in anhydrous
benzene immediately before use. A solution of white
phosphorus in benzene was prepared and stored in an
inert atmosphere in a Schlenk flask, which was pre-evac-
uated and filled with argon. 31P NMR (δ, ppm): –520 s.
The UV spectra of the catalyst solutions were mea-
sured on an SF-2000 spectrophotometer (Russia) in
quartz cells with an optical path length of 0.01 cm. The
conversion of Pd(acac)2 was monitored based on an
absorption band at 330 nm (ε330 = 10630 L cm–1 mol–1).
Additionally, the interaction products of Pd(acac)2
with elemental phosphorus were analyzed on a
GCMS-QP2010 Ultra Shimadzu chromatography–
mass spectrometry instrument (Japan) equipped with
a 30-m GsBP-5MS capillary column with a poly(5%
diphenyl, 95% dimethylpolysilphenylene siloxane)
phase. Electron impact ionization was used, and the
ionization energy was 70 eV. The experimental mass
spectra were compared with published data (Wiley,
Pd–P Catalyst Preparation Methods
Method 1. A Pd–P (in situ) catalyst for analysis by
high-resolution transmission electron microscopy
(HR TEM) was prepared by the same method as that
used for studying its properties in the hydrogenation of
2-ethyl-9,10-anthraquinone. To
a
solution of
Pd(acac)2 (0.0152 g, 5 × 10–5 mol) in 7 mL of toluene
placed in a thermostatically controlled thermostated
‘duck’ type glass vessel, 1 mL of a solution of phos- NIST, and NIST05 reference libraries).
KINETICS AND CATALYSIS Vol. 60 No. 6 2019

810
BELYKH et al.
The XRD analysis of the Pd–P catalyst sample was
Activity, min–1
Yield of H2O2, %
performed on a Shimadzu XRD-7000 S diffractome-
20
18
16
14
12
10
8
6
4
2
100
90
ter (CuKα radiation; Ni filter; λ = 1.5418 Å).
The TEM images were obtained on a Tecnai G2
electron microscope (FEI, the United States) with an
accelerating voltage of 200 kV. The images were
recorded using a CCD camera (Soft Imaging System,
Germany). The microscope was equipped with an
energy-dispersive X-ray characteristic spectrometer
(EDX, Phoenix) with a semiconductor Si(Li) detec-
tor. The parameters of particle images were measured
using the iTEM 5.0 and DigitalMicrographs 1.94.1613
software. The fast Fourier transformation (FFT) and
inverse fast Fourier transformation (IFFT) methods
were used for analyzing periodic structures and filter-
ing the images. An area containing at least 100 parti-
cles was treated to find the average size.
1
80
70
60
50
40
30
20
10
0
2
0
0.2
0.4
0.6
0.8
1.0
P : Pd
Fig. 1. Effect of the phosphorus content on the properties
of the Pd catalyst in the hydrogenation of 2-ethyl-9,10-
anthraquinone: (1) the yield of H O and (2) the catalyst
2
2
The XPS spectra were measured on a PHOIBOS
150 MCD 9 photoelectron spectrometer (SPECS,
Germany) using monochromated AlKα radiation from
an X-ray tube (1486.74 eV). The survey spectrum was
recorded with a step of 1 eV at an analyzer transmission
energy of 20 eV, and high-resolution spectra (narrow
scans), with a step of 0.1 eV and a transmission energy
of 10 eV. The C1s line (285.0 eV) of carbon was used as
a reference. Nonuniform charging was eliminated by
irradiating the sample with slow electrons. The sam-
ples were etched with Ar+ ions for 2 min using a PU-
IQE 12/38 ion source (SPECS) at an accelerating volt-
age of 2.5 kV and an ion current of 20 μA, which
ensured an etching rate of ~1 nm/min. The experi-
mental data were processed using the CasaXPS pro-
gram. The spin–orbital splitting of the Pd3d5/2–3/2 and
P2p3/2–1/2 doublet lines was approximated by two
Lorentz–Gaussian curves with doublet separations of
activity. [Pd] = 2.5 mmol/L; [2-ethyl-9,10-anthraqui-
none]/[Pd] = 42.3; solvent, 10 mL of 1-octanol + 10 mL
of toluene; 50°C;
= 2 atm.
P
H2
genation of 2-ethyl-9,10-anthraquinone [6]. At P : Pd
ratios of 0.3–1.0, phosphorus decreased the activity of
the palladium catalyst in the hydrogenation of 2-ethyl-
9,10-anthraquinone by almost an order of magnitude;
in this case, the yield of hydrogen peroxide increased
from 69% (Pd black) to 96–98% (Fig. 1). According to
the TEM data, the average particle size of the Pd–P
catalyst (P : Pd = 0.3) was almost 3.5 times smaller
than that of the Pd black (Fig. 2); therefore, such a
sharp drop in the activity of the Pd–P catalyst com-
pared to that of the Pd black cannot be associated with
a difference in dispersity. To experimentally substanti-
ate the reasons for the modifying action of phospho-
5.26 and 0.84 eV, respectively, at area ratios of 3/2 for rus, the state of a surface layer of the Pd–P catalyst
Pd3d5/2–3/2 and 2/1 for P2p3/2–1/2. The full width at half formed in toluene–1-octanol solution was studied by
XPS.
maximum (FWHM) of ≤1.5 eV of the spectral lines
was taken according to published data for analogous
substances and similar conditions used for measuring
the spectra. We assumed that, if the FWHM exceeded
1.5 eV (1.2 eV for phosphorus), these lines should be
approximated by several curves corresponding to dif-
ferent chemical species of the element. Because the
oxygen 1s spectral line strongly overlaps with the pal-
ladium 3p3/2 line, the number of chemical forms of
palladium and the position of their lines were con-
nected with the second Pd3p1/2 doublet line in the sur-
vey spectrum.
The surface composition of the palladium catalyst
was represented by four elements: Pd, P, C, and O.
Table 1 summarizes concentration ratios between the
chemical forms of the elements. The initial spectrum
of the Pd3d line consists of three 3d5/2–3/2 spin–orbital
doublets characterizing various chemical states of pal-
ladium (Fig. 3a). The binding energy of electrons of
the high-intensity Pd3d5/2 component with a maxi-
mum at 335.5 eV corresponds to palladium similar to
palladium in the reduced state Pd(0), conventionally
designated as Pdδ+ [9]. Shifts in the binding energies of
the other two components (335.8 and 337.4 eV) rela-
tive to the binding energy of the Pd3d5/2 level of the
bulk metal (335.2 eV [9]) indicate a positive charge on
palladium. The Pd3d5/2 binding energy of 337.4 eV is
RESULTS AND DISCUSSION
The introduction of phosphorus before the stage of
Pd(acac)2 reduction with hydrogen affects the activity
and selectivity of the palladium catalyst in the hydro-
characteristic of Pd2+ in Pd(acac)2 [10, 11]. The con-
KINETICS AND CATALYSIS
Vol. 60
No. 6
2019

EFFECT OF THE STATE OF A SURFACE LAYER ON THE PROPERTIES
(а) (b)
811
10 1/nm
0.2 μm
200 nm
5 1/nm
Fig. 2. TEM images of (a) Pd black and (b) the Pd–P catalyst. The insets (a, b) show the electron diffraction spectra.
centration of the Pd(acac)2 precursor among three dif- (Eb(Pd3d5/2) = 335.5 eV, Eb(O1s) = 528.9 eV [12]),
ferent forms of palladium did not exceed 5.2%.
small palladium clusters (Eb(Pd3d5/2) = 336.1 eV for
Pd10 [13]), palladium phosphide, or a solid solution
(Eb(Pd3d5/2) = 335.8–336.2 eV) [14, 15]. The absence
of the O1s component with a maximum at 528.9 eV
from the XPS spectrum (Table 1, Fig. 4c) excludes the
The main chemical state of palladium (72 at %) in
the Pd–P catalyst is represented by a component with
a Pd3d5/2 binding energy of 335.8 eV (Table 1). The
electron-deficient form of palladium can result from
the formation of a 2D surface palladium oxide formation of 2D surface palladium oxide. This con-
Table 1. Binding energies and concentrations of the elements in various states in the surface layers of the Pd–P catalyst iso-
lated from the Pd(acac)2–0.3P–Н2 system after etching with argon
The most probable chemical
Spectral line Binding energy, eV
FWHM, eV
Element concentration, at %
species*
Pdδ+ (0.23)**
335.5
0.7
0.9
1.5
0.6
1.2
3.1
Pd3d5/2
PdxP (0.72)
13.6
9.2
335.8
9.8
0.7
3.2
1.2
Pd(acac)2 (0.05)
PdxP (0.35)
337.4
130
Pdδ+–Р (0.13)
131.1
P2p3/2
2−
133.5
1.2
1.7
3.1
–
R–O–
(0.18)
PO3
PO34−
134.6
532.4
1.2
(0.34)
Pd3p3/2 (Pdδ+, PdxP)
2.80
O–С + O–P + O=С (1.0)
O1s
С1s
16.5
60.7
533.1
534.5
285.0
1.70
3.3
16.5
–
(R–O– 2−, R–OН, Pd(acac)2)
PO3
Pd3p3/2 (Pd(acac)2)
С–С + СН2 + СН3 (0.83)
1.4
50.6
(R–O– 2−, R–OН, Pd(acac)2)
PO3
С–О–P (0.12)
286.8
289.1
1.2
1.1
7.2
2.9
(R–O– 2−, R–OН)
PO3
С=О + СО2 (0.05) (Pd(acac)2)
* The atomic fraction of this chemical species is indicated in parentheses.
δ+
** The form Pd is similar to palladium in the reduced state Pd(0).
KINETICS AND CATALYSIS Vol. 60 No. 6 2019

812
BELYKH et al.
(а)
(b)
Intensity
Intensity
PdxP
Pdδ+
PdxP
Pdδ+
Pd(acac)2
Pd(acac)2
348 344 340 336 332
Binding energy, eV
348 344 340 336 332
Binding energy, eV
(c)
(d)
Intensity
PdxP
Intensity
PdxP
PO33–
PO34–
PO33–
PO34–
Pdδ+–P
Pdδ+–P
138 136 134 132 130 128
Binding energy, eV
138 136 134 132 130 128
Binding energy, eV
Fig. 3. (a, b) Pd3d and (c, d) P2p XPS spectra of the Pd–P catalyst (a, c) before and (b, d) after etching with argon. The broken
line represents the experimental spectrum. The line enveloping the experimental spectrum corresponds to the sum of the decom-
position components of the XPS line, and marked lines correspond to the components of XPS line decomposition into the chem-
0
ical forms of (a, b) palladium (Pd , Pd P, and Pd(acac) ) or (c, d) phosphorus.
x
2
clusion was also confirmed by the quantitative analysis layer based on a solid solution of Pd and P. It is diffi-
of the surface composition after etching with argon
(Table 2).
cult to make a conclusion on the nature of surface
compounds because the P2p3/2 binding energies in
various metal phosphides and in solid solutions are
similar. Based on the relative atomic concentrations of
phosphorus and palladium (Table 1), the empirical
formula of the main surface compound PdxP, charac-
terized by binding energies of 335.8 eV (Pd3d5/2) and
130 eV (P2p3/2), can be represented as Pd3.06P. The
empirical formula Pd3.06P of the surface compound
corresponds to the composition of a palladium phos-
phide, Pd3P or Pd3P0.8 [20], but the X-ray phase anal-
ysis data given below are inconsistent with the forma-
tion of the bulk phase of crystalline palladium phos-
phide.
At the same time, the phosphorus 2p XPS spec-
trum of the Pd–P catalyst consisted of four Р2р3/2-1/2
spin–orbital doublets with the following positions of
the maxima of the high-intensity component Р2р3/2:
130, 131.1, 133.5, and 134.6 (Fig. 3c). The P2p3/2 bind-
ing energies of 133.5 and 134.6 eV indicate the pres-
ence of the oxidized forms of tetracoordinated pen-
2−
tavalent phosphorus (R–O–
3−) on the sur-
PO3 , PO4
face [16], whereas the positions of maximums at130
and 131.1 eV are characteristic of elemental phospho-
rus (Eb(P2p3/2) = 129.9–130.4 eV) [17, 18], phospho-
rus in alloys (Eb(P2p3/2) = 130.1 eV) and transition
metal phosphides (Eb(P2p3/2) = 128.5–130 eV) [17,
The diffraction curve of the Pd–P catalyst exhib-
19], or a nanoscale (several monolayers) disordered ited two broad reflections at 2θ angles of 39.59° and
(insufficiently coordinated amorphous) 2D surface 71.21° (Fig. 5). The interplanar distances of 2.2765
KINETICS AND CATALYSIS
Vol. 60
No. 6
2019

EFFECT OF THE STATE OF A SURFACE LAYER ON THE PROPERTIES
813
(а)
(b)
Intensity
Intensity
C–C
C–C
C–O–P
C–O–P
C=O
C=O
292 290 288 286 284 282
Binding energy, eV
292 290 288 286 284 282
Binding energy, eV
(c)
(d)
Intensity
Intensity
Pd3p3/2
(Pdδ+, PdxP)
Pd3p3/2
(Pdδ+, PdxP)
Pd3p3/2
Pd3p3/2
(Pd(acac)2)
(Pd(acac)2)
O–C, O–P
O=C
O–C, O–P
O=C
540
536
532
528
540
536
532
528
Binding energy, eV
Binding energy, eV
Fig. 4. (a, b) C1s and (c, d) O1s XPS spectra of the Pd–P catalyst (a, c) before and (b, d) after etching with argon. The broken
line represents the experimental spectrum. The line enveloping the experimental spectrum corresponds to the sum of the decom-
position components of the XPS line, and marked lines correspond to the components of XPS line decomposition into the chem-
ical forms of (a, b) carbon or (c, d) oxygen.
and 1.3241 Å are closest to the interplanar distances of nm. The difference by a factor of 1.8 between CSR
sizes (1.6 and 0.9 nm) calculated from the half-widths
of both reflections (2θ = 39.59° and 71.21°) clearly
indicates the imperfection of the crystal lattice of the
Pd–P catalyst. The electron diffraction data (Fig. 2b,
inset) and the HR TEM images (Fig. 6b, inset) also
indicate a low structural ordering of the sample. The
size of ordering regions in the particles of the Pd–P
catalyst does not exceed 0.6–1.0 nm (Fig. 6b, inset).
The distortion of the palladium crystal lattice is
associated with the entry of phosphorus (a metalloid)
into the composition of the particles. According to the
EDX analysis data, the ratio between P and Pd for dif-
the planes (111) (d/n = 2.2458 Å) and (220) (d/n =
1.37537 Å) of palladium crystallites (# 00-046-1043).
The observed broadening of diffraction lines in the
X-ray diffraction pattern can be due to a decrease in
the particle size to less than 100 nm and microstresses
caused by chemical inhomogeneity and defective
structure. The size of the coherent scattering region
(CSR) calculated from the half-width of the main
reflection (at reflection angles 2θ = 39.59°) was no
greater than 1.6 nm. However, according to the TEM
data, the number average particle diameter of the Pd–
P catalyst was 34.4 13.7 nm (Fig. 2b). Such a signif-
icant difference in the CSR size (as a rule, it is identi- ferent sections of the Pd–P catalyst was 0.23 : 1
fied with the average crystallite size) and the TEM (Fig. 6). Therefore, the empirical formula of the bulk
phase was Pd4.3P. The interplanar distances in the dif-
fraction curve, which are most characteristic of palla-
dium, and the low structural ordering resulting from
data indicates that the broadening of diffraction lines
in the X-ray diffraction pattern is related to the defect
structure of the sample and low structural ordering
rather than a decrease in particle size to less than 100 the incorporation of phosphorus into the palladium
KINETICS AND CATALYSIS Vol. 60 No. 6 2019

814
BELYKH et al.
Table 2. Binding energies and concentrations of the elements in various states in the surface layers of the Pd–P catalyst iso-
lated from the Pd(acac)2–0.3P–Н2 system after etching with argon
Spectral
line
The most probable chemical
species*
Binding energy, eV
FWHM, eV
Element concentration, at %
Pdδ+ (0.16)**
PdxP (0.77)
335.6
335.9
337.7
130.0
131.1
0.6
1.0
1.5
0.6
1.2
1.9
Pd3d5/2
12.0
9.5
9.3
0.8
2.8
1.1
Pd(acac)2 (0.07)
PdxP (0.29)
Pdδ+–Р (0.11)
P2p3/2
2−
133.3
134.5
532.2
1.2
1.2
2.5
1.4
4.2
–
R–O–
(0.15)
PO3
3− (0.45)
PO4
Pd3p3/2 (Pdδ+, PdxP)
O–С + P–O + O=С (1.0)
O1s
С1s
19.6
58.9
533.3
534.5
285.0
1.7
3.3
1.4
19.6
–
(R–O– 2−, R–OН, Pd(acac)2)
PO3
Pd3p3/2 (Pd(acac)2)
С–С + СН2 + СН3 (0.85)
49.9
(R–O– 2−, Pd(acac)2)
PO3
С–О–P (0.12)
286.9
289.0
1.2
1.5
7.0
2.1
(R–O– 2−, R–OН)
PO3
С=О + СО2 (0.03) (Pd(acac)2)
* The atomic fraction of this chemical species is indicated in parentheses.
δ+
** The form Pd is similar to palladium in the reduced state Pd(0).
crystal lattice suggest the formation of a structurally mation of oxidized phosphorus species. This conclu-
sion is based on the data given below.
disordered solid solution between palladium and
phosphorus. The possibility of the formation of substi-
tutional solid solutions was experimentally demon-
strated by Moreau et al. [21] using nickel and phos-
phorus as an example despite differences in the atomic
radii and electronegativities.
First, the surface compositions before and after
argon etching differed only slightly (Tables 1 and 2). In
this case, the fraction of phosphorus in the oxidized
state after etching with argon did not decrease. Sec-
ond, the C1s spectrum of the Pd–P catalyst contained
three chemical forms of carbon (Fig. 4a) correspond-
ing to the C–O–P (Eb(C1s) = 286.6 eV) and C=O
(Eb(C1s) = 289.1 eV) fragments and C–C/C–H ali-
phatic radicals (Eb(C1s) = 285.0 eV) [23] (Table 1).
2−
The quantitative analysis of the surface composi-
tion of the Pd–P catalyst showed that a total P : Pd
ratio of 0.68 (Table 1) was higher by a factor of almost
2 than the initial ratio (P : Pd = 0.3). That is, the sur-
face of the Pd–P catalyst was enriched in phosphorus
compared to the bulk phase. In this case, more than a
The presence of the С–O–
fragment indicated
PO3
the presence of phosphoric acid alkyl esters on the sur-
half of phosphorus (52%) occurred in an oxidized face of the Pd–P catalyst.
form (Table 1). The presence of the oxidized form of
The formation of octyl phosphite in the reaction
phosphorus was most often associated with the oxida-
tion of surface transition metal phosphides with oxy-
gen [22]. However, for the Pd–P catalyst obtained by
the low-temperature method from Pd(acac)2 and P4 in
medium was confirmed by chromatography–mass
spectrometry. Using UV spectroscopy, we found that
~15% Pd(acac)2 in toluene–1-octanol solution in an
inert atmosphere reacts with phosphorus at a P : Pd
hydrogen, other processes also contributed to the for- ratio of 0.3:
KINETICS AND CATALYSIS
Vol. 60
No. 6
2019

EFFECT OF THE STATE OF A SURFACE LAYER ON THE PROPERTIES
Intensity
815
25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100105 110 115
2θ, deg
Fig. 5. X-ray diffraction pattern of the Pd–P catalyst. Vertical lines show the positions of the reflections of crystalline palladium.
Pulse, arb. units
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
Pd Lα
5 nm
2 nm
(а)
(b)
Pd Lβ
CuKα
C K
P L
P K
Si K
Pd Lβ2
O K
0
1.3
2.6
3.9
5.2
6.5
7.8
9.1
Energy, keV
Fig. 6. EDX spectrum of the Pd–P catalyst. The insets (a, b) show HR TEM images with different scales.
The analysis of a surface layer of Pd–P particles
(P : Pd = 0.3) formed in the toluene–1-octanol solu-
tion allowed us to formulate the following reasons for
the modifying effect of phosphorus on the properties
of palladium catalysts in the chemoselective hydroge-
nation of 2-ethyl-9,10-anthraquinone:
3Pd(acac)2 + 2P + 6H2O
= 3Pd(0) + 6Hacac + 2H3PO3.
A third of the introduced phosphorus was spent in
the redox process. The phosphoric acids formed in the
redox process enter into an esterification reaction with
1-octanol used as a co-solvent. Octyl phosphites
and/or octyl phosphates adsorbed on Pd–P catalyst
particles act as stabilizers. A higher temperature
(90°C) required for the reduction of Pd(acac)2, which
contains O,O-chelate-bound acac ligands, with
hydrogen in toluene–1-octanol solution, in contrast
to dimethylformamide [24, 25], leads to the formation
of particles of structurally disordered solid Pd–P solu-
tions, the surface of which is enriched in electron-
deficient palladium.
As noted above, a sharp drop in the activity and a
significant increase in the selectivity of the Pd–P cat-
alyst compared to Pd black cannot be explained by a
change in the particle size (Figs. 1 and 2). First, the
average particle size of the Pd–P catalyst was almost
3.5 times smaller than that of the Pd black (Fig. 2). If
the geometrical factor played the main role, the activ-
ity in terms of total palladium should increase as a
result of an increase in the dispersity of the Pd–P cat-
KINETICS AND CATALYSIS Vol. 60 No. 6 2019

816
BELYKH et al.
FUNDING
alyst. Second, both large (d = 127 nm) and small (d =
1.5–2.5 nm) palladium nanoclusters were character-
ized by a lower yield of hydrogen peroxide than that on
Pd–P particles (Fig. 1), although they also accelerated
various side processes [6].
The study was supported by the Russian Science Foun-
dation (project no. 17-73-10158).
REFERENCES
In our opinion, the most important factor deter-
mining changes in the properties of the palladium cat-
alyst upon phosphorus modification is the enrichment
of its surface in electron-deficient palladium presum-
ably in the form of a Pd–P solid solution and the pres-
ence of a stabilizer on the surface of the Pd–P catalyst
particles. As a rule, Pd catalysts with a high electron
density on palladium are more active in the hydroge-
nation of anthraquinone [3, 26]. Antibatic and sym-
batic effects of a decrease in the electron density of
palladium on the catalytic activity and selectivity,
respectively, were observed previously in the chemo-
selective hydrogenation of α,β-unsaturated aldehydes
to unsaturated alcohols [27]. On the one hand, the
appearance of a partial positive charge on palladium in
Pd–P particles weakens the activation of molecular
hydrogen and the H–H bond cleavage required for
catalytic hydrogenation [28, 29]. On the other hand,
the solubility of hydrogen, that is, the concentration of
nonselective hydrogen, in solutions of palladium with
phosphorus decreases in comparison with palladium
crystallites [30]. In this case, the solutions of palla-
dium with phosphorus contain more strongly bound
hydrogen [31], which is not favorable for the hydroge-
nation of an aromatic ring [32]. A high yield of H2O2
in the presence of amorphous Ni–Cr–B alloys, in
contrast to crystalline nickel catalysts, was also noted
by Liu et al. [32].
1. Ciriminna, R., Albanese, L., Meneguzzo, F., and
Pagliaro, M., ChemSusChem., 2016, vol. 9, p. 1.
2. Campos-Martin, J.M., Blanco-Brieva, G., and Fierro, J.L.,
Angew. Chem., 2006, vol. 45, no. 42, p. 6962.
3. Yuan, E., Wu, C., Hou, X., Dou, M., Liu, G., Li, G.,
and Wang, L., J. Catal., 2017, vol. 347, p. 79.
4. Yuan, E., Wu, C., Liu, G., Li, G., and Wang, L., J. Ind.
Eng. Chem., 2018, vol. 66, p. 158.
5. Kosydar, R., Drelinkiewicz, A., Lalik, E., and Gurgul, J.,
Appl. Catal., A, 2011, vol. 402, no. 1–2, p. 121.
6. Sterenchuk, T.P., Belykh, L.B., Skripov, N.I., San-
zhieva, S.B., Gvozdovskaya, K.L., and Shmidt, F.K.,
Kinet. Catal., 2018, vol. 59, no. 5, p. 585.
7. Gordon, A.J. and Ford, R.A., The Chemist’s Compan-
ion: A Handbook of Practical Data, Techniques, and Ref-
erences, Wiley, 1972.
8. US Patent 3474464, 1969.
9. Miquel, P., Yamin, Y., Lombaert, K., Dujardin, C.,
Trentesaux, M., Gengembre, L., and Granger, P., Surf.
Interface Anal., 2010, vol. 42, nos. 6–7, p. 545.
10. Mazalov, L.N., Trubina, S.V., Kryuchkova, N.A.,
Tarasenko, O.A., Trubin, S.V., and Zharkova, G.I., J.
Struct. Chem., 2007, vol. 48, no. 2, p. 253.
11. Selishchev, D.S., Kolobov, N.S., Bukhtiyarov, A.V.,
Gerasimov, E.Y., Gubanov, A.I., Kozlov, D.V., Appl.
Catal., B, 2018, vol. 235, p. 214.
12. Gabasch, H., Unterberger, W., Hayek, K., Klotzer, B.,
Kleimenov, E., Teschner, D., Zafeiratos, S., Havecker, M.,
Knop-Gericke, A., Schlog, R., Han, J., Ribeiro, F.H.,
Aszalos-Kiss, B., Curtin, T., and Zemlyanov, D., Surf.
Sci., 2006, vol. 600, p. 2980.
Thus, changes in the properties (activity and selec-
tivity) of the palladium catalyst in the production of
hydrogen peroxide by the anthraquinone method are
associated with the enrichment of the surface of Pd–P
particles in electron-deficient palladium during the
formation of palladium catalysts in the presence of
elemental phosphorus in hydrogen in toluene–1-
octanol solution and the stabilization of particles with
the octyl esters of phosphoric acids.
13. Wu, T., Kaden, W.E., Kunkel, W.A., and Anderson, S.L.,
Surf. Sci., 2009, vol. 603, no. 17, p. 2764.
14. Successful Design of Catalysts: Future Requirements and
Development, Inui, T., Ed., Amsterdam: Elsevier, 1988,
p. 3.
15. Zhao, M., Chem. – Asian J., 2016, vol. 11, p. 461.
16. Bedia, J., Rosas, J.M., Rodrıguez-Mirasol, J., and
Cordero, T., Appl. Catal., B, 2010, vol. 94, nos. 1–2, p. 8.
17. Blanchard, P.E.R., Grosvenor, A.P., Cavell, R.G., and
Mar, A., Chem. Mater., 2008, vol. 20, no. 22, p. 7081.
ACKNOWLEDGMENTS
18. Rego, R., Ferraria, A.M., Botelho do Rego, A.M., and
Oliveira, M.C., Electrochim. Acta, 2013, vol. 87, p. 73.
The X-ray photoelectron spectra were measured on a
PHOIBOS 150 MCD 9 photoelectron spectrometer at the
Center for Collective Use of the Krasnoyarsk Scientific
Center, Siberian Branch, Russian Academy of Sciences.
19. Grosvenor, A.P., Cavell, R.G., and Mar, A., J. Solid
State Chem., 2008, vol. 181, no. 10, p. 2549.
20. Skripov, N.I., Belykh, L.B., Belonogova, L.N., Rokh-
in, A.V., Stepanova, T.P., and Shmidt, F.K., Russ. J.
Gen. Chem., 2012, vol. 82, no. 2, p. 206.
21. Moreau, L.M., Ha, D.H., Bealing, C.R., Zhang, H.,
Hennig, R.G., and Robinson, R.D., Nano Lett., 2012,
vol. 12, no. 9, p. 4530.
The electron microscopy images of the catalyst samples
were obtained on an electron microscope at the Baikal Cen-
ter for Nanotechnology, Irkutsk National Research Techni-
cal University (INRTU).
KINETICS AND CATALYSIS
Vol. 60
No. 6
2019

EFFECT OF THE STATE OF A SURFACE LAYER ON THE PROPERTIES
817
22. Sun, H., Xu, J., Fu, G., Mao, X., Zhang, L., Chen, Y., 28. Maccarrone, M.J., Lederhos, C.R., Torres, G., Betti, C.,
Zhou, Y., Lu, T., and Tang, Y., Electrochim. Acta, 2012,
vol. 59, p. 279.
Coloma-Pascual, F., Quiroga, M.E., and Yori, J.C.,
Appl. Catal., A, 2012, vols. 441–442, p. 90.
23. Puziy, A.M., Poddubnaya, O.I., and Ziatdinov, A.M., 29. Mironenko, R.M., Belskaya, O.B., Lavrenov, A.V., and
Appl. Surf. Sci., 2006, vol. 252, no. 23, p. 8036.
Likholobov, V.A., Kinet. Catal., 2018, vol. 59, no. 3,
p. 339.
24. Skripov, N.I., Belykh, L.B., Belonogova, L.N.,
Umanets, V.A., Ryzhkovich, E.N., and Shmidt, F.K.,
Kinet. Catal., 2010, vol. 51, no. 5, p. 714.
30. Stojewski, M., Kowalska, J., and Jurczakowski, R., J.
Phys. Chem. C, 2009, vol. 113, no. 9, p. 3707.
25. Belykh, L.B., Skripov, N.I., Sterenchuk, T.P., Kole-
snikov, S.S., and Schmidt, F.K., J. Nanosci. Nanotech-
nol., 2017, vol. 17, no. 1, p. 696.
31. Flanagan, B.T.B., Biehl, G.E., Clewley, J.D., Kund-
qvist, S., and Anderson, Y., J. Chem. Soc., Faraday
Trans. 1, 1980, vol. 76, p. 196.
26. Kosydar, R., Drelinkiewicz, A., Lalik, E., and Gurgul, J.,
32. Liu, B., Qiao, M., Wang, J., and Fan, K., Chem. Com-
mun., 2002, p. 1236.
Appl. Catal., A, 2011, vol. 402, nos. 1–2, p. 121.
27. Maki-Arvela, P., Hajek, J., Salmi, T., and Murzin,
Translated by Valentin Makhlyarchuk
D.Yu., Appl. Catal., A, 2005, vol. 292, p. 1.
KINETICS AND CATALYSIS Vol. 60 No. 6 2019