Content uploaded by Sergey Sarvadiy
Author content
All content in this area was uploaded by Sergey Sarvadiy on May 27, 2019
Content may be subject to copyright.
453
ISSN 1995-0780, Nanotechnologies in Russia, 2018, Vol. 13, Nos. 9–10, pp. 453–463. © Pleiades Publishing, Ltd., 2018.
Original Russian Text © A.K. Gatin, M.V. Grishin, N.V. Dokhlikova, S.A. Ozerin, S.Yu. Sarvadii, B.R. Shub, 2018, published in Rossiiskie Nanotekhnologii, 20 18, Vol. 13, Nos. 9–10.
Adsorption Properties of the Film Formed by Gold and Copper
Nanoparticles on Graphite
A. K. Gatina, M. V. Grishina,*, N. V. Dokhlikovaa, S. A. Ozerina, S. Yu. Sarvadiia, and B. R. Shuba
a Semenov Institute of Chemical Physics, Russian Academy of Sciences, Moscow, 119991 Russia
*e-mail: mvgrishin68@yandex.ru
Received October 11, 2018; accepted October 17, 2018
Abstract—Some physicochemical properties of homogeneous and heterogeneous films formed by Au and Cu
nanoparticles on graphite are studied by scanning tunneling microscopy and spectroscopy. It is found that the
nanoparticles have a shape close to spherical with a diameter of 3‒6 nm, the gold particles do not contain
impurities, and the copper particles can be coated with oxide. The adsorption properties of nanostructured
coatings with respect to hydrogen, carbon oxide, and oxygen are determined. Copper oxide is reduced by car-
bon oxide and hydrogen, but the latter is also adsorbed onto oxide-free copper particles and gold. Exposure
to oxygen results in the reformation of the oxide on copper. The possibility of rearranging the electronic struc-
ture of copper nanoparticles during hydrogen adsorption is confirmed by the results of quantum chemical
simulation.
DOI: 10.1134/S1995078018050063
INTRODUCTION
The progress of modern industrial chemistry is
largely due to the use of new catalysts based on
nanoparticles. The use of nanoparticles, including
multicomponent systems, is the main direction when
creating new catalysts, e.g., bimetallic catalysts con-
sisting of gold and copper nanoparticles are used in the
oxidation processes of various hydrocarbons: benzyl
alcohol, propylene, methanol, and others [1–8]. In
addition, gold–copper coatings are considered effec-
tive catalysts for the low-temperature oxidation of CO
[9, 10]. In many cases, bimetallic nanostructures show
significantly better optical, electronic, and catalytic
properties when compared with monometallic [11].
This is due to the fact that they can possess not only
the properties of each of the components separately,
but also receive new ones due to the interaction
between them; i.e., they show the so-called synergistic
effect.
Usually, studies of catalysts, even consisting of
nanoparticles, are carried out by standard methods,
which practically exclude the possibility of detecting
differences between individual nanoparticles. To
answer the question of the contribution of each of the
components of the system under study to its proper-
ties, new approaches are needed to provide local
“chemical sensitivity,” as well as high spatial and tem-
poral resolution. The first conditions correspond to
probe methods for studying the surface and, in partic-
ular, the method of scanning tunneling microscopy
and spectroscopy (STM/STS) [12–16]. The latter
condition is satisfied by applying the quantum chemi-
cal simulation of real nanosized objects. A comparison
of data obtained in the course of STM/STS experi-
ments and as a result of calculations makes it possible
to obtain the most detailed information about the
structure and properties of nanostructured systems.
The aim of this work was to determine physical param-
eters and adsorption properties of single nanoparticles
that are included in the composition of gold–copper
film on the surface of highly oriented pyrolytic graph-
ite (HOPG).
EXPERIMENTAL
The experiments were carried out under ultra-
high vacuum conditions (residual gas pressure P ≤ 2 ×
10–10 Torr). This made it possible to eliminate the
uncontrolled change in the chemical composition of
the samples due to residual gases and provided the
reliability of the information.
For the formation of the coating, the impregnation
method was used. An aqueous solution of HAuCl4
and/or Cu(NO3)2 with a metal concentration of 2.4 ×
10–5 g/mL was applied on HOPG, dried, and calcined
at T = 750–850 K in vacuum.
To analyze the structural features and electronic
structure of single nanoparticles, scanning tunneling
microscopy and spectroscopy were used. The studies
were carried out using tips prepared from PtIr-wire
according to standard procedures. The tips were
cleaned in vacuum by applying an electric field of high
NANOSTRUCTURES,
INCLUDING NANOTUBES
454
NANOTECHNOLOGIES IN RUSSIA Vol. 13 Nos. 9–10 2018
GATIN et al.
intensity. Only those tips that showed reproducible
S-shaped curves characteristic of nanocontacts
formed by metals when measuring current–voltage
characteristics [12–16] were used in the work.
The results of the chemical interaction of gas phase
molecules with single nanoparticles were determined
by scanning tunneling spectroscopy. To this end, mea-
surements were made of the current–voltage charac-
teristics of STM nanocontacts containing nanoparti-
cles deposited on a substrate (hereinafter, the CVC of
nanoparticles); the current–voltage characteristics of a
tunnel contact that does not contain nanoparticles are
referred to below as the CVC of HOPG. As a result of
the chemical action, the CVC of nanoparticles could
undergo qualitative changes: the appearance of sec-
tions of zero current (band gap), a series of local max-
ima, etc. [17, 18]. These features ref lect the formation
of the oxide layer on the surface of the nanoparticles
and the adsorption of single molecules and particles.
The control of the elemental composition of the
sample surface was carried out using Auger spectros-
copy. The control of the gase-phase composition in
the ultrahigh vacuum chamber at all stages of work,
including the synthesis of bimetallic coatings, the inlet
of gaseous reactants, and the isolation of the products
of their interaction, was carried out using a quadru-
pole mass spectrometer.
The numerical experiment was carried out in the
framework of the density functional theory (DFT).
For the calculations, the software packages OpenMX
3.8 (OMX) [19] and QuantumEspresso 5.1.1 (QE) [20]
were used. In both, the generalized gradient approxi-
mation and the PBE functional were chosen. Relativ-
istic effects of electron motion were taken into account
when generating ultra-soft pseudopotentials. The
accuracy of the chosen basic sets approximately corre-
sponded to the double Sletter basis, both for the OMX
package with numerical atomic-centered basic orbitals
and for the QE package with a flat-wave basis. The
specified goals did not include the achievement of
experimental values; therefore, in order to avoid ambi-
guities, the overlapping of bases was not taken into
account in the calculations.
RESULTS AND DISCUSSION
Gold nanoparticles on graphite. The film formed by
Au nanoparticles on HOPG was studied in STM (see
Fig. 1). It was found that, on the surface of HOPG, as
a rule, clusters of nanoparticles having a shape close to
spherical (diameter of 3‒5 nm, Fig. 1a) were formed
near the boundaries of the graphene terraces. About
10% of the substrate surface is coated with nanoparti-
cles. The shape of the CVC of nanoparticles makes it
possible to conclude that their electronic structure
corresponds to the electronic structure of the metal
and the conductivity of an STM nanocontact with a
nanoparticle is much higher than the conductivity of
an STM nanocontact without a nanoparticle. This
means that gold contains almost no foreign inclusions,
which is confirmed by the measured Auger spectra of
the sample under study. The results of these studies
completely coincide with the data obtained by us ear-
lier [18].
The study of the adsorption properties of gold
nanoparticles with respect to hydrogen, carbon oxide,
and oxygen was carried out by us earlier [17]. As a
result, it was found that (1) the adsorption of hydrogen
is dissociative and results in a rearrangement of the
electronic structure of nanoparticles from metallic to
Fig. 1. (Color online) Nanos tructured gold f ilm on HOPG :
(a) image of the sample surface area of 82 × 82 nm and
(b) CVC of HOPG (curve 1) and nanoparticles (curve 2).
210
Voltage, V
−1−2
1.5
(a)
(b) Tunnel current, nA
1.0
0.5
0
1
2
−0.5
−1.0
−1.5
NANOTECHNOLOGIES IN RUSSIA Vol. 13 Nos. 9–10 2018
ADSORPTION PROPERTIES OF THE FILM 455
semiconducting; (2) the interaction of CO molecules
with atomic hydrogen adsorbed on gold results in the
formation of HCO particles (formyl radicals); and
(3) oxygen oxidizes formyl radicals, as a result of
which molecules of carbon dioxide and water arise.
The adsorption of carbon oxide and oxygen on gold
nanoparticles is possible only after the preliminary
adsorption of hydrogen.
Cu nanoparticles on graphite. An investigation of a
film consisting of Cu nanoparticles deposited on
HOPG in a scanning tunneling microscope, along
with some general properties, revealed a number of
significant differences of this coating from a film based
on gold nanoparticles on the same substrate (Fig. 2).
It was found that, in most cases, the shape of cop-
per nanoparticles is close to spherical and the diameter
is 4‒6 nm. The vast majority of nanoparticles are
grouped near the boundaries of the graphene sheets
constituting the substrate, and they form irregularly
shaped clusters consisting of hundreds of nanoparti-
cles (Fig. 2a). The nanoparticles also cover about 10%
of the substrate surface. Measurements in the STM
showed that, unlike gold nanoparticles, the CVC of
which are almost identical, the CVC of copper
nanoparticles can be divided into three types (Fig. 2b).
The insignificant part of nanoparticles corresponds to
the CVC of curve 2 in Fig. 2b. On these curves, there
is no region of zero current (band gap), and the curves
themselves lie above the CVC of graphite. Further, the
index Cu-2 will denote this type of nanoparticles.
CVCs similar to curve 3 in Fig. 2b correspond to more
than half of nanoparticles. On this curve, a section of
zero current with a width of about 1 V is detected (the
band gap with a width of 1 eV). This means that copper
is coated with a layer of semiconductor, apparently,
oxide. These nanoparticles will be below denoted as
Cu-3. Finally, on the CVC of 25–30% of nanoparti-
cles, there is a section of zero current (a forbidden
zone) with a width of 1.6–2 V; i.e., they are also coated
with a semiconductor layer—an oxide, the character-
istics of which, however, differ from those of an oxide
on the surface of Cu-3 nanoparticles. The latter type
of nanoparticles will be designated as Cu-4. Thus, on
the surface of HOPG there are copper nanoparticles
that differ in their electronic structure: Cu-2, without
a band gap; Cu-3, with a band gap of about 1 eV wide;
and Cu-4, with a band gap of about 1.6‒2 eV. In
Fig. 2b, edges of the band gaps on the CVC are indi-
cated by arrows corresponding to the color of the
curves.
The initial distribution of the CVC of copper
nanoparticles over various states of their electronic
structure is shown in Fig. 3a. Let us consider the prob-
able chemical composition of the Cu-2, Cu-3, and
Cu-4 nanoparticles, correlating the results of local
measurements in the STM with the literature data.
The electronic structure of CuO, Cu2O, and Cu4O3
copper oxides was considered in [21], as well as in a
number of other works [22‒26]. According to the data
presented in them, all these oxides are semiconduc-
tors, and Cu2O has the band gap of the greatest
width—2.17–2.62 eV [22], while the band gaps of CuO
and Cu4O3 are in the range of 1.35–1.7 eV [22‒26].
One can conclude that Cu-4 particles are copper-
coated with a layer of Cu2O oxide (possibly nonstoi-
chiometric), and Cu-3 particles are copper-coated
with CuO or Cu4O3 oxide, the electronic characteris-
tics of which are close. Apparently, Cu-2 nanoparti-
Fig. 2. (Color online) Nanostructured copper film on
HOPG: (a) image of a sample surface area of 399 × 399 nm
and (b) CVC of HOPG (curve 1) and nanoparticles
(curves 2–4). The arrows indicate the boundaries of the
band gaps on the corresponding color curves.
1.51.00.50−0.5−1.0−1.5 Voltage, V
1.5
2.0
(a)
(b)
Tunnel current, nA
1.0
0.5
0
1
2
3
4
−0.5
−1.0
−1.5
−2.0
456
NANOTECHNOLOGIES IN RUSSIA Vol. 13 Nos. 9–10 2018
GATIN et al.
cles can be correlated with particles of both Cu-3 and
Cu-4 adsorbed on HOPG defects, as well as copper
nanoparticles lacking an oxide layer on the surface.
The electronic structure of oxidized copper nanopar-
ticles can be strongly perturbed due to the interaction
with defects, the electronic structure of which also dif-
fers greatly from the electronic structure of the defect-
free surface of graphite [27]. These considerations are
indirectly confirmed by the fact that the proportion of
Cu-2 nanoparticles is insignificant and accounts for
less than 5% of the total number of nanoparticles.
The sample exposures successively in CO, H2, and
O2 significantly change the electronic structure of Cu
nanoparticles deposited on HOPG. Thus, after expo-
sure in CO, the proportion of particles coated with
oxide (states Cu-3 and Cu-4) is drastically reduced
and, along with Cu-2 nanoparticles, particles in the
new state Cu-5 appear (Fig. 3b). CVCs without the
band gap correspond to these particles, but the CVC
curve of the Cu-5 nanoparticles in the entire measure-
ment range of the spectroscopic dependences lies
below the CVC curve of the HOPG (an example of the
CVC of such particles is curve 3 in Fig. 4). In [28], CO
adsorption and CO2 synthesis on a CuO/CeO2 sample
were observed at room temperature, although the
authors explained the formation of CO2 by the interac-
tion of CO with the substrate—CeO2. In another paper
[29], it was noted that oxidized copper was reduced
using CO to a metallic state already at relatively low
temperatures of 473 K. Thus, our results can be inter-
preted as the complete (Cu-2 particles) or partial
(Cu-5 particles) reduction of copper oxides. The
incomplete reduction of copper oxides in a stream of
gaseous CO also finds its explanation in the frame-
work of studies [30]: for the adsorption of CO, Cu0
atoms are preferable to Cu+ atoms. According to the
results of our research, the number of copper atoms of
the first type is extremely small (Fig. 3a), which
resulted in the adsorption of the number of CO mole-
cules insufficient for complete metal reduction.
The sample exposure in H2 again significantly
changed the electronic structure of most nanoparti-
cles: the number of Cu-3 and Cu-4 particles, as well as
Cu-2, increased, but the number of Cu-5 particles
sharply decreased (Fig. 3c). It was shown in [31] that
the reduction of copper oxides with hydrogen is a
complex multistage process and includes an induction
period, as well as the incorporation of H into an oxide,
which as a result turns directly into a metal without the
formation of intermediate oxides (Cu4O3 or Cu2O).
The activation energy for the CuO → Cu transition is
about 14.5 kcal/mol, while the value of the activation
energy for the Cu2O → Cu transition is approximately
equal to 27.4 kcal/mol. In addition, the results of
quantum chemical calculations presented below
showed that, during the adsorption of hydrogen on
impurity-free copper, their electronic structure is
rearranged from the metallic to the semiconductor
Fig. 3. (Color online) Evolution of electronic structure of Cu nanoparticles in the presence of various gases: (a) initial state,
(b) exposure in CO, (c) exposure in H2, and (d) exposure in O2.
Cu-2Cu-3Cu-4
1.0
0.8
(a) Initial state
0.6
0.4
0.2
0Cu-2Cu-5Cu-4 + Cu-3
1.0
0.8
(b) Exposure in СО
0.6
0.4
0.2
0
Cu-2Cu-3Cu-4 + Cu-3
1.0
0.8 (c) Exposure in H2
0.6
0.4
0.2
0Cu-2Cu-5Cu-4 + Cu-3
1.0
0.8 (d) Exposure in O2
0.6
0.4
0.2
0
NANOTECHNOLOGIES IN RUSSIA Vol. 13 Nos. 9–10 2018
ADSORPTION PROPERTIES OF THE FILM 457
type. Thus, exposure in hydrogen of a sample contain-
ing copper nanoparticles, both oxide-coated and
oxide-free, results (1) in the reduction of copper oxide
(and an increase in the number of nanoparticles with
CVC like curve 2 in Fig. 2) and (2) in the adsorption of
hydrogen on oxide-free copper nanoparticles (the
process described in [32]), as well as the reduction of a
small number of nanoparticles coated with Cu2O
oxide. It is unreduced nanoparticles coated with a
layer of Cu2O, as well as particles coated with adsorbed
hydrogen, that contribute to the number of particles
with a CVC like curve 4 in Fig. 2b. At the same time,
spectroscopic measurements in the STM have shown
that there is no interaction between the adsorbed CO
molecules and hydrogen. This is confirmed by the
provisions of [30, 33]. They indicate that the adsorp-
tion and oxidation of CO due to oxygen from the
oxide, followed by the desorption of CO2, is preferable
to the interaction with hydrogen.
Finally, the exposure in О2 predictably resulted in
the oxidation of the overwhelming number of
nanoparticles, and the proportion of particles in the
states Cu-5 and Cu-2 decreased in comparison with
the previous experiment (Fig. 3d).
Thus, the results of the studies of coating based on
copper nanoparticles result in the following conclu-
sions:
1. Initially, copper nanoparticles coated with vari-
ous oxides, namely CuO, Cu4O3, and Cu2O, are on the
surface of graphite.
2. Exposure into CO results in the reduction of
copper oxides.
3. The results of exposure in H2 are similar to the
action of CO, but in addition there is a rearrangement
of the electronic structure of oxide-free copper
nanoparticles under the action of adsorbed hydrogen.
4. Exposure into O
2 results in the reoxidation of
copper nanoparticles.
Bimetallic gold–copper film on graphite. Studies in
the STM showed that clusters of spherical nanoparti-
cles (2–6 nm in diameter) located predominantly at
the edges of the terraces are located on the surface of
HOPG (Fig. 4a). Up to 15% of the HOPG surface is
coated with nanoparticles. Spectroscopic measure-
ments in the STM revealed three types of nanoparti-
cles that differ significantly in their electronic struc-
ture (Fig. 4b). Curves 2, 3, and 4 in Fig. 4b correspond
to these particles. A quantitative analysis of various
types of curves, similar to that performed for a film of
Cu nanoparticles, shows (Fig. 5a) that, in the bimetal-
lic Au–Cu coating, the content of “metallic”
nanoparticles, the CVCs of which have the shape of
curve 2 in Fig. 4b (below, NP-2 particles), as well as
nanoparticles, to which CVCs like curve 3 in Fig. 4b
correspond (hereinafter referred to as NP-3), signifi-
cantly increased. Finally, a noticeable number of par-
ticles are characterized by CVCs like curve 4 in Fig. 4b
(below, NP-4 particles). The number of NP-2 and
NP-4 nanoparticles is approximately the same and is
equal to 30%, and the number of NP-3 particles is
40% of the total number. Such a distribution is obvi-
ously due to the presence in the system of gold
nanoparticles, which have an electronic metal struc-
ture. In addition, it seems that the interaction of gold
and copper nanoparticles prevents the complete oxi-
dation of the latter. This assumption is confirmed by
the conclusions of [34], where the existence of nonox-
ide gold–copper nanostructured coatings on the sur-
face of reduced graphene oxide was also shown by
X-ray diffraction. In addition, the results of a number
Fig. 4. (Color online) Accumulations of Au and Cu
nanoparticles on HOPG: (a) topographic image of a sam-
ple surface area of 434 × 434 nm and (b) examples of CVC
of HOPG (curve 1) and nanoparticles (curves 2–4). The
arrows indicate the boundary of the band gap on curve 4.
1.51.00.50−0.5−1.0−1.5 Voltage, V
1.5
2.0
(a)
(b)
Tunnel current, nA
1.0
0.5
0
1
2
3
4
−0.5
−1.0
458
NANOTECHNOLOGIES IN RUSSIA Vol. 13 Nos. 9–10 2018
GATIN et al.
of studies [35, 36] indicate that, under conditions
when one of the components of the system is oxidized,
mixed particles (alloys) are not formed; i.e., NP-3
particles are unlikely to be correlated with nanoparti-
cles from the AuCu alloy. A comparison of the infor-
mation with the results of experiments with monome-
tallic films consisting of gold and copper nanoparticles
made it possible to conclude that NP-2 nanoparticles
are gold, NP-3 nanoparticles are copper (the surface
of which may contain only a small amount of oxygen
atoms (Cu-5)), and NP-4 nanoparticles are copper
coated with a layer of oxide (Cu-3 and Cu-4). The
results of X-ray photoelectron spectroscopy studies
[9] showed that, in gold–copper systems, like those
studied in this work, gold is in the Au0 state and copper
is in Cu0 and Cu2+ states, which corresponds to our
data. Let us emphasize that, in the monometallic film
formed by copper nanoparticles on HOPG, the frac-
tion of oxide-free particles is small.
Adsorption properties of a nanostructured gold–
copper film were determined with respect to CO, H2,
and O2. For this, samples were exposed in the above
gases at T = 300 K and P = 1 × 10–6 Torr for 30 min.
As a result of adsorption, a change in the shape of the
CVC of nanoparticles occurred, which reflects a
change in their electronic structure due to a change in
the elemental composition. At that, no change in the
morphology of the studied film took place.
Exposure of a sample in CO results in a change in
the ratio between nanoparticles with different elec-
tronic structures, i.e., CVC curves of different shapes.
The number of NP-3 nanoparticles increases sharply
and the number of NP-4 nanoparticles decreases
(Fig. 5b). Thus, a restructuring of the electronic struc-
ture of the surface of the nanoparticles from the semi-
conductor type to the metallic type is observed. This
means that the concentration of oxygen atoms on the
surface of nanoparticles in a given film decreases due
to the interaction with CO molecules. Similar results
were obtained in [37] for catalytic systems, including
nanoparticles with the composition AuCu/CuO. This
work emphasizes that the presence of copper oxide
(CuO) is necessary for the oxidation of CO, while the
AuCu system is inactive in this reaction. The fact that
the number of NP-3 nanoparticles, and not NP-2,
increased, means that the exposure into CO used by us
is insufficient for the complete reduction of oxidized
copper.
The effect of hydrogen on the electronic structure
of nanoparticles is ambiguous (Fig. 5c). As one can see
from this figure, there was a sharp increase in the
number of NP-4 nanoparticles (with the surface layer
of a semiconductor). To explain these results, it is nec-
essary to remember that hydrogen not only reduces the
oxidized copper nanoparticles, but, adsorbed on gold
and oxide-free copper nanoparticles, it also causes the
transformation of their electronic structure from
metallic to semiconductor. Thus, the group of NP-4
Fig. 5. (Color online) Effect of adsorption on the electronic
structure of a nanostr uctured Au–Cu film: (a) initial state,
(b) exposure in CO, (c) exposure in H2, and (d) exposure
in O2.
0.4 (a)
(b)
(c)
(d)
Particle number, rel. units
Initial
0.3
0.2
0.1
0NP-4 NP-3 NP-2
0.4
0.5
0.6 CO only
0.3
0.2
0.1
0NP-4 NP-3 NP-2
0.4 H2
0.3
0.2
0.1
0NP-4 NP-3 NP-2
0.4
0.5
O2
0.3
0.2
0.1
0NP-4 NP-3 NP-2
Type of nanoparticles
NANOTECHNOLOGIES IN RUSSIA Vol. 13 Nos. 9–10 2018
ADSORPTION PROPERTIES OF THE FILM 459
particles includes not only nanoparticles coated with
oxide (Cu-4 and Cu-3), but also copper and gold
nanoparticles, the surface of which contains adsorbed
hydrogen.
The result of exposure into oxygen of the sample,
which was previously maintained in carbon oxide, and
then in hydrogen, is paradoxical. On the one hand, as
expected, exposure in O2 causes the reoxidation of
some copper nanoparticles (Fig. 5d), which results in
an increase in NP-4 particles. However, on the other
hand, the number of NP-2 particles also increased.
Obviously, this is due to two processes taking place
simultaneously on Cu and Au nanoparticles coated
with hydrogen adatoms and free from hydrogen.
A release of Au and Cu nanoparticles (which consti-
tute the group of particles NP-2) from the adsorbed
hydrogen and the oxidation of copper nanoparticles
take place. Let us also note that, when the surface of
Au and Cu nanoparticles is released from hydrogen,
the latter can be desorbed mainly in the composition
of H2O molecules. This means that, in the presence of
copper (or copper oxide), as in the presence of oxi-
dized nickel nanoparticles [17], a two-stage process of
water synthesis is realized, rather than a three-stage
process characteristic of gold nanoparticles on HOPG
[18]. It should also be noted that an increase in the
exposure time of this sample in oxygen resulted in the
complete oxidation of copper nanoparticles.
The sequential exposure of the sample in H2 and
CO (and vice versa) practically did not result in the
interaction of these gases with each other on the sur-
face of nanoparticles: no traces of adsorption of mole-
cules with C–H and C–O bonds were found by STM
methods.
Quantum chemical study of hydrogen adsorption on
Au and Cu nanoparticles. Clusters of Aun and Cun (n =
13‒147), both free (Au and Cu) and associated with a
graphene cluster (Au), simulating a substrate, were the
models of gold and copper nanoparticles. The atomic
structure of clusters was calculated using the classical
Morse potential at the first stage of computations and
quasi-Newton methods within the DFT frameworks
at the second stage. Since the search for a global min-
imum was not the aim of the work, clusters with the
configuration of atoms corresponding to the local
minimum of the total energy satisfied the tasks posed.
The stable isomers of copper clusters, even with filled
icosahedral shells, n = 13, 55, and 147, were not
objects with a precisely defined type of symmetry, like
the models of gold nanoparticles in our previous work
[38]. Let us note that the distribution of density of
electron states for Cu13 and Au13 clusters has separate
local maxima, which gradually merge with the
increase in size of the Cu55 and Au55 cluster, forming
similarities of zone in the solid.
The position of the hydrogen atom. Table 1 presents
the energies and bond lengths of the Au13 cluster with
hydrogen in various positions relative to the nearest
gold atoms calculated by us. According to the calcula-
tions, the values of energy and bond lengths of Au–H
in different positions of the hydrogen atom above the
surface of the gold cluster differed slightly and were on
average 3.67 eV and 1.77 Å, respectively. However, the
position “bridge” (i.e., associated with two atoms of
gold) was the most advantageous. Previously, using
the example of the Au13Cu54 system, we have shown
that interaction with a graphene cluster (Cu54) results
in a distortion of the atomic structure and charging of
the gold cluster (Au13). The most energetically favor-
able places for hydrogen adsorption, located at the
gold-graphene interface, have been revealed [38].
Carrying out similar calculations for a free Cu13
copper cluster showed that the most stable position of
a hydrogen atom on the cluster surface is “hollow”
(i.e., associated with three gold atoms) (Table 2). The
average energy and bond length are 2.97 eV and 1.62 Å,
respectively. One can assume that the difference in the
most advantageous positions of the hydrogen atom on
the cluster surface is due to the different degrees of
hybridization of the valence s–d orbitals of copper and
gold. In copper, the mixing of s–d states is less; there-
fore, the repulsive interaction between the filled d-zone
and the s orbital of the hydrogen atom will be less pro-
nounced, which will make it possible for the hydrogen
atom to occupy the most symmetric position relative
to the copper atoms on the cluster surface. This ques-
tion, however, is not included in the tasks posed in this
paper. For more thorough research, additional calcu-
lations are required.
Effect of hydrogen adsorption on the electronic struc-
ture of clusters. In order to determine how strongly the
interaction with hydrogen affects the electronic struc-
ture of gold and copper clusters, the distribution of the
projected (local) density of states on atoms without
and with a hydrogen atom was calculated. Since the
interaction with a single copper atom was estimated,
the position of the atop hydrogen atom was chosen.
The local electron density of d-states on gold and cop-
Table 1. Values of energy and bond length of Au13 cluster
with a hydrogen atom in various positions relative to the
nearest gold atoms
Atop Bridge Hollow Average
Energy, eV –3.62 –3.75 –3.64 3.67
Bond length, Å 1.6 1.8 1.9 1.77
Table 2. Values of energy and bond length of Cu13 cluster
with a hydrogen atom in various positions relative to the
nearest gold atoms
Atop Bridge Hollow Average
Energy, eV –2.59 –3.13 –3.18 2.97
Bond length, Å 1.5 1.65 1.7 1.62
460
NANOTECHNOLOGIES IN RUSSIA Vol. 13 Nos. 9–10 2018
GATIN et al.
per atoms interacting with hydrogen decreases when a
bond with hydrogen is formed (Figs. 6a, 6b). The
appearance of antibinding states of the d-zone of cop-
per and gold clusters and the s orbital of the hydrogen
atom, σ*, are also observed. Changes in the distribu-
tion of the local density of states for neighboring atoms
are much smaller (Figs. 6c, 6d). Thus, acts of succes-
sive adsorption for clusters of gold and copper can be
considered independent in the sense of the smallness
of the spatial perturbation region defined by the act of
adsorption, i.e., E(k)–E(k+1) (E(k), E(k + 1)),
where E(k) is the adsorption energy of the kth hydro-
gen atom.
Sequential adsorption of hydrogen atoms. To esti-
mate the effect of the interaction with hydrogen atoms
on the energy structure depending on the number of
atoms in a metal cluster, the calculation of the density
of states for gold clusters Au
13, Au55, and Au147 and
copper Cu13, Cu55, and Cu147 without hydrogen atoms
and with a surface completely filled with hydrogen
atoms was carried out (Fig. 7). As the number of
hydrogen atoms increases, the “center of gravity” of
the density of states shifts from the Fermi level towards
negative energy values both in the case of gold clusters
and copper clusters, which means, according to the
resonant chemisorption model, a decrease in the
chemical activity of the cluster. The measure of chem-
ical activity in this case is the value of the binding
energy.
Table 3 presents the calculated values of the bind-
ing energy of hydrogen atoms in gold Au13, Au20, Au55,
Au100, and Au147 and copper Cu13, Cu20, Cu55, Cu100,
and Cu147 clusters with different surface filling: one
hydrogen atom and full filling. As one can see, a com-
plete correlation with changes in the energy spectra of
the clusters does not take place, which indicates a
complication of the resonance model of chemisorp-
tion. This means that in the objects under study the
value of the binding energy is determined not only by
the coordination saturation of the copper and gold
atoms, but also by the distribution of the local density
of states near the Fermi level. Since the clusters chosen
by us have low nuclearity, their chemical properties
can be determined to a significant degree not only by
the local density of d-states lying below the Fermi
Fig. 6. (Color online) Projected density of states on (a) gold atom bound to the hydrogen atom, (b) gold atom adjacent to it,
(c) copper atom bound to the hydrogen atom, and (d) copper atom adjacent to it in the system without hydrogen (PDOS1) and
with hydrogen (PDOS2).
10−1−2−3−4−5
8
7(a)
σ*
PDOS1
PDOS2
6
5
4
3
2
1
Density of states, eV−1
0
10−1−2−3-4 Energy (E − Ef), eV
−5
12
11
10
7
8
9
(b)
σ*
PDOS1
PDOS2
6
5
4
3
2
1
Density of states, eV−1
0
10−1-2−3−4−5
8
9
7
(c)
PDOS1
PDOS2
6
5
4
3
2
1
0
10−1−2−3−4 Energy (E − Ef), eV
−5
12
11
10
7
8
9
(d)
PDOS1
PDOS2
6
5
4
3
2
1
0
NANOTECHNOLOGIES IN RUSSIA Vol. 13 Nos. 9–10 2018
ADSORPTION PROPERTIES OF THE FILM 461
level, but also by the sp-states, which are smaller in
value but near the Fermi level.
Thus, the acts of successive adsorption in the sys-
tems considered by us can be considered independent.
When the number of adsorbed hydrogen atoms
increases, the “center of gravity” of the density of
states of copper and gold clusters shifts toward lower
energy values. Changes in the energy spectra during
hydrogen adsorption correspond to changes in the
binding energy of hydrogen atoms with gold and cop-
per clusters within the framework of the resonance
model of the d-zone, which indicates a similar mech-
anism for the hydrogenation of copper and gold
nanoparticles.
Fig. 7. (Color online) Projected densities of states of Aun, Cun, AunHm, and CunHm clusters, where n = 13, 55, and 147 and m =
12, 39, and 86: (a) for Au13 and Au13H12 clusters, (b) for Au55 and Au55H39 clusters, (c) for Au147 and Au147H86 clusters, (d) for
Cu13 and Cu13 H12 clusters, (e) for Cu55 and Cu55H39 clusters, and (f) for Cu147 and Cu147H86 clusters.
0−5
120
(a)
(b)
(c)
(d)
(e)
(f)
Density of states, eV−1
100 Au13
Au13H12
80
60
40
20
0−5
200 Au55
Au55H39
150
100
50
0
0
10−1−2−3−4
Energy, eV
−5−6−7
500 Au147
Au147H86
400
300
200
100
0−5
140
Density of states, eV−1
100
120 Cu13
Cu13H12
80
60
40
20
0−5
200
300
350
Cu55
Cu55H39
150
250
100
50
0
0
00
10−1−2−3−4
Energy, eV
−5−6−7
500
600
700
800
Cu147
Cu147H86
400
300
200
100
462
NANOTECHNOLOGIES IN RUSSIA Vol. 13 Nos. 9–10 2018
GATIN et al.
CONCLUSIONS
As a result of the research carried out it was found
that (1) Au and Cu nanoparticles have a shape close to
spherical with a diameter of about 3‒6 nm and in most
cases are part of homogeneous and heterogeneous
clusters; (2) the surface of Cu nanoparticles in homo-
geneous clusters is coated with oxides Cu3O4, Cu2O,
or CuO; (3) the exposure of these nanoparticles into
CO results in a partial reduction of the oxide and
exposure into H2 simultaneously reduces the remain-
ing oxidized nanoparticles and rearranges the elec-
tronic structure from the metallic type to the semicon-
ductor in oxide-free copper nanoparticles; (4) the
interaction of Cu and Au nanoparticles, which are part
of heterogeneous clusters, favors the predominant for-
mation of CuO layer on copper; (5) with respect to CO
adsorption, a heterogeneous film is similar to copper
nanoparticles, and with respect to hydrogen adsorp-
tion it is similar to gold nanoparticle film. A quantum
chemical simulation of hydrogen adsorption on gold
and copper nanoparticles showed that the electronic
structure of gold and copper clusters when interacting
with hydrogen changes in a similar way.
ACKNOWLEDGEMENTS
This work was supported by the Russian Science
Foundation, project no. 18-73-00195. The resources
of the Interdepartmental Supercomputer Center of the
Russian Academy of Sciences were used in the calcu-
lations.
REFERENCES
1. D. Kim, J. Resasco, Y. Yu, A. M. Asiri, and P. D. Yang,
“Synergistic geometric and electronic effects for elec-
trochemical reduction of carbon dioxide using gold–
copper bimetallic nanoparticles,” Nat. Commun. 5,
4948–4956 (2014).
2. S. Neatu, J. A. Macia-Agullo, P. Concepcion, and
H. Garcia, “Gold–copper nanoalloys supported on
TiO2 as photocatalysts for CO2 reduction by water,”
J. Am. Chem. Soc. 136, 15969–15976 (2014).
3. R. He, Y. C. Wang, X. Y. Wang, Z. Wang, G. Liu,
W. Zhou, L. Wen, Q. Li, X. Wang, X. Chen, J. Zeng,
and J. G. Hou, “Facile synthesis of pentacle gold–cop-
per alloy nanocrystals and their plasmonic and catalytic
properties,” Nat. Commun. 5, 4327–4337 (2014).
4. C. L. Bracey, P. R. Ellis, and G. J. Hutchings, “Appli-
cation of copper-gold alloys in catalysis: current status
and future perspectives,” Chem. Soc. Rev. 38, 2231–
2243 (2009).
5. T. Pasini, M. Piccinini, M. Blosi, R. Bonelli,
S. Albonetti, N. Dimitratos, J. A. Lopez-Sanchez,
M. Sankar, Q. He, C. J. Kiely, G. J. Hutchings, and
F. Cavani, “Selective oxidation of 5-hydroxymethyl-2-
furfural using supported gold-copper nanoparticles,”
Green Chem. 13, 2091–20 99 (2011).
6. C. Della Pina, E. Falletta, and M. Rossi, “Highly selec-
tive oxidation of benzyl alcohol to benzaldehyde cata-
lyzed by bimetallic gold–copper catalyst,” J. Catal.
260, 384–386 (2008).
7. J. C. Bauer, G. M. Veith, L. F. Allard, Y. Oyola,
S. H. Overbury, and S. Dai, “Silica-supported Au-
CuOx hybrid nanocrystals as active and selective cata-
lysts for the formation of acetaldehyde from the oxida-
tion of ethanol,” ACS Catal. 2, 2537–2546 (2012).
8. J. Llorca, M. Dominguez, C. Ledesma, R. J. Chi-
mentão, F. Medina, J. Sueiras, I. Angurell, M. Seco,
and O. Rossell, “Propene epoxidation over TiO2-sup-
ported Au–Cu alloy catalysts prepared from thiol-
capped nanoparticles,” J. Catal. 258, 187–198 (2008).
9. L. Wang, Y. Zhong, H. Jin, D. Widmann, J. Weiss-
muller, and R. J. Behm, “Catalytic activity of nano-
structured Au: scale effects versus bimetallic/bifunc-
tional effects in low-temperature CO oxidation on
nanoporous Au,” Beilstein J. Nanotechnol. 4, 111– 128
(2013).
10. X. Liu, A. Wang, L. Li, T. Zhang, C.-Y. Mou, and
J.-F. Lee, “Synthesis of Au–Ag alloy nanoparticles
supported on silica gel via galvanic replacement reac-
tion,” J. Catal. 278, 288–296 (2011).
11. S. Dutta, C. Ray, S. Sarkar, M. Pradhan, Y. Negishi,
and T. Pal, “Silver nanoparticle decorated reduced
graphene oxide (RGO) nanosheet: a platform for SERS
based low-level detection of uranyl ion,” Electrochem.
Acta 180, 1075–1084 (2015).
12. General Principles and Applications to Clean and Absor-
bate-Covered Surfaces, Ed. by H.-J. Guntherodt and
R. Wiesendanger (Springer, Berlin, 1992).
13. G. Binnig, H. Rohrer, C. Berber, and E. Weibel, “Tun-
neling through a controllable vacuum gap,” Appl. Phys.
Lett. 40, 178–180 (1981).
14. E. Meyer, H. J. Hug, and R. Bennewitz, Scanning
Probe Microscopy (Springer, Berlin, 2004).
15. R. J. Hamers and Y. J. Wang, “Atomically-resolved
studies of the chemistry and bonding at silicon sur-
faces,” Chem. Rev. 96, 1261–1290 (1996).
16. R. J. Hamers, R. M. Tromp, and J. E. Demuth, “Sur-
face electronic structure of Si (111)-(7 × 7) resolved in
real space,” Phys. Rev. Lett. 56, 1972–1975 (1986).
17. A. K. Gatin, M. V. Grishin, S. Yu. Sarvadii, and
B. R. Shub, “Interaction of gaseous reagents on gold
and nickel nanoparticles,” Russ. J. Phys. Chem. B 12,
317–324 (2018).
18. M. V. Grishin, A. K. Gatin, N. V. Dokhlikova,
A. A. Kirsankin, A. I. Kulak, S. A. Nikolaev, and
B. R. Shub, “Adsorption and interaction of hydrogen
Table 3. Binding energy of hydrogen atoms with gold and
copper clusters
mAu13, eV Au20, eV Au55, eV Au100, eV Au147, eV
1 –3.13 –3.22 –3.50 –3.40 –3.25
full –3.14 –3.26 –3.14 –3.09 –3.03
mCu13, eV Cu20, eV Cu55, eV Cu100, eV Cu147, eV
1 –3.29 –3.35 –3.15 –3.32 –3.31
full –3.27 –3.23 –3.22 –2.96 –3.01
NANOTECHNOLOGIES IN RUSSIA Vol. 13 Nos. 9–10 2018
ADSORPTION PROPERTIES OF THE FILM 463
and oxygen on the surface of separate crystalline gold
nanoparticles,” Kinet. Catal. 56, 532–539 (2015).
19. T. Ozaki, “Variationally optimized atomic orbitals for
large-scale electronic structures,” Phys. Rev. B 67,
155108 (2003).
20. P. Giannozzi, S. Baroni, N. Bonini, M. Calandra,
R. Car, C. Cavazzoni, D. Ceresoli, G. L. Chiarotti,
M. Cococcioni, I. Dabo, A. Dal Corso, S. de Giron-
coli, S. Fabris, G. Fratesi, R. Gebauer, et al.,
“QUANTUM ESPRESSO: a modular and open-
source software project for quantum simulations of
materials,” J. Phys.: Condens. Matter 21, 395502
(2009).
21. M. Heinemann, B. Eifert, and C. Heiliger, “Band
structure and phase stability of the copper oxides Cu2O,
CuO, and Cu4O3,” Phy s . R ev. B 87, 115111 (2013).
22. B. K. Meyer, A. Polity, D. Reppin, M. Becker, P. Her-
ing, P. J. Klar, Th. Sander, C. Reindl, J. Benz, M. Eick-
hoff, C. Heiliger, M. Heinemann, J. Bläsing, A. Krost,
S. Shokovets, C. Müller, and C. Ronning, “Front
cover: binary copper oxide semiconductors: from mate-
rials towards devices,” Phys. Status Solidi B 249, 1487–
1647 (2012).
23. J. Ghijsen, L. H. Tjeng, J. van Elp, H. Eskes, J. Wester-
ink, G. A. Sawatzky, and M. T. Czyzyk, “Electronic
structure of Cu2O and CuO,” Phys. Rev. B 38, 11322–
11330 (1988).
24. F. P. Koffyberg and F. A. Benko, “A photoelectro-
chemical determination of the position of the conduc-
tion and valence band edges of p-type CuO,” J. Appl.
Phys. 53, 1173 –1177 ( 1982).
25. F. Marabelli, G. B. Parravicini, and F. Salghetti-Drioli,
“Optical gap of CuO,” Phys. Rev. B 52, 1433–1436
(1995).
26. J. F. Pierson, A. Thobor-Keck, and A. Billard,
“Cuprite, paramelaconite and tenorite films deposited
by reactive magnetron sputtering,” Appl. Surf. Sci. 210,
359–367 (2003).
27. F. I. Dalidchik, S. A. Kovalevskii, and A. V. Kovytin,
“Atomic and electronic structure of surface nanoscale
graphite structures,” Khim. Fiz. 23 (7), 83–90 (2004).
28. A. di Benedetto, G. Landi, and L. Lisi, “Improved
CO-PROX performance of CuO/CeO2 catalysts by
using nanometric ceria as support,” Int. J. Hydrogen
Energy 42, 12262–12275 (2017).
29. E. A. Goldstein and R. E. Mitchell, “Chemical kinetics
of copper oxide reduction with carbon monoxide,”
Proc. Combust. Inst. 33, 2803–2810 (2011).
30. Y. Bu, S. Er, J. W. Niemantsverdriet, and
H. O. A. Fredriksson, “Preferential oxidation of CO
in H2 on Cu and Cu/CeOx catalysts studied by in situ
UV-vis and mass spectrometry and DFT,” J. Catal.
357, 176–187 (2018).
31. J. Y. Kim, J. A. Rodriguez, J. C. Hanson, A. I. Frenkel,
and P. L. Lee, “Reduction of CuO and Cu2O with H2:
H embedding and kinetic effects in the formation of
suboxides,” J. Am. Chem. Soc. 125, 10684–10692
(2003).
32. J. Harris and A. Liebsch, “On the physisorption inter-
action of H2 with Cu-metal,” Phys. Scr. 4, 14–16
(1983).
33. T. Caputo, L. Lisi, R. Pirone, and G. Russo, “On the
role of redox properties of CuO/CeO2 catalysts in the
preferential oxidation of CO in H2-rich gases,” Appl.
Catal., A 348, 42–53 (2008).
34. L. Rout, A. Kumar, R. S. Dhaka, G. N. Reddy, S. Giri,
and P. Dash, “Bimetallic Au-Cu alloy nanoparticles on
reduced graphene oxide support: synthesis, catalytic
activity and investigation of synergistic effect by DFT
analysis,” Appl. Catal., A 538, 107–122 (2017).
35. J. Han, Y. Zahou, Y.-Q. Chai, L. Mao, Y.-L. Yuan, and
R. Yuan, “Highly conducting gold nanoparticles-
graphene nanohybrid films for ultrasensitive detection
of carcinoembryonic antigen,” Talanta 85, 130–135
(2011).
36. J. Wang, J. Li, A. J. Baca, J. Hu, F. Zhou, W. Yan, and
D. W. Pang, “Amplified voltammetric detection of
DNA hybridization via oxidation of ferrocene caps on
gold nanoparticle/streptavidin conjugates,” Anal.
Chem. 75, 3941–3945 (2003).
37. W. Zhan, J. Wang, H. Wang, J. Zhang, X. Liu,
P.Zhang, M. Chi, Y. Guo, Y. Guo, G. Lu, S. Sun,
S. Dai, and H. Zhu, “Crystal structural effect of AuCu
alloy nanoparticles on catalytic CO oxidation,” J. Am.
Chem. Soc. 139, 8846–8854 (2017).
38. N. V. Dokhlikova, N. N. Kolchenko, M. V. Grishin,
A. K. Gatin, and B. R. Shub, “Substrate effect on
hydrogen adsorption on gold cluster,” Nanotechnol.
Russ. 11, 735–742 (2016).
Translated by V. Kudrinskaya
