Structure and Stability of a Copper(II) Lactate Complex in Alkaline Solution: A Case Study by Energy-Dispersive X-ray Absorption Spectroscopy

Article abstract of DOI:10.1021/acs.inorgchem.7b00553

Energy-dispersive X-ray absorption spectroscopy was applied, aimed at solving the problem of the structure and stability of a copper(II) lactate complex in alkaline solution, used as a precursor for the electrodeposition of Cu₂O. The application of multiple scattering calculations to the simulation of the X-ray absorption near-edge structure part of the spectra allowed an accurate resolution of the structure: the copper(II) cation is surrounded by four lactate ions in a distorted tetrahedral environment, with the lactate anions acting as monodentate ligands. This results in an atomic arrangement where copper is surrounded by four oxygen atoms located at quite a short distance (ca. 1.87 ?) and four oxygen atoms located quite far apart (ca. 3.1-3.2 ?). The complex was finally found to be stable in a wide range of applied potentials.

Full text of DOI:10.1021/acs.inorgchem.7b00553

Article
pubs.acs.org/IC
Structure and Stability of a Copper(II) Lactate Complex in Alkaline
Solution: a Case Study by Energy-Dispersive X‑ray Absorption
Spectroscopy
†
‡
‡
‡
‡
‡
Elisabetta Achilli, Alberto Vertova, Alberto Visibile, Cristina Locatelli, Alessandro Minguzzi,
,†
Sandra Rondinini, and Paolo Ghigna*
†Dipartimento di Chimica, Universita
Italy
̀
degli Studi di Pavia and Unita di Ricerca INSTM di Pavia, Viale Taramelli 16, 27100, Pavia,
‡
Dipartimento di Chimica, Universita
S Supporting Information
̀
degli Studi di Milano and Unita di Ricerca INSTM di Milano, Via Golgi 19, 20133, Milan, Italy
*
ABSTRACT: Energy-dispersive X-ray absorption spectrosco-
py was applied, aimed at solving the problem of the structure
and stability of a copper(II) lactate complex in alkaline
solution, used as a precursor for the electrodeposition of Cu O.
2
The application of multiple scattering calculations to the
simulation of the X-ray absorption near-edge structure part of
the spectra allowed an accurate resolution of the structure: the
copper(II) cation is surrounded by four lactate ions in a
distorted tetrahedral environment, with the lactate anions
acting as monodentate ligands. This results in an atomic
arrangement where copper is surrounded by four oxygen atoms
located at quite a short distance (ca. 1.87 Å) and four oxygen atoms located quite far apart (ca. 3.1−3.2 Å). The complex was
finally found to be stable in a wide range of applied potentials.
10
1. INTRODUCTION
form a complex with copper with 1:2 (Cu/L) stoichiometry,
but no direct information is presently available. In addition,
copper(II) alkalyne solutions with L excess can display
complicated electrochemical behavior, showing different peaks
in cyclic voltammograms. This work is aimed at addressing the
structure and electrochemical stability of copper(II) lactate
complexes in alkaline solution. In this respect, X-ray absorption
spectroscopy (XAS) plays a key role, especially in characterizing
electrocatalytic nanoparticles, as was already demonstrated by
Cu O is a low-cost, nontoxic, and environmentally acceptable
2
material that can be employed in devices for solar energy
1
conversion. The value of the band gap ranges from 1.96 to
2
2
.38 eV, making Cu O one of the most suitable materials
2
among the candidates for the production of hydrogen starting
3
,4
from water and sunlight. One of the most used and cost-
effective procedures for preparing Cu O films is an electro-
2
deposition from an alkaline solution of copper sulfate, where
copper hydroxide (or oxide) precipitation is avoided by adding
11−14
some of us.
The ability of XAS to solve structural
5
problems in solution has been demonstrated since the early
a large excess of lactate ions. The use of electrochemical
15
days of synchrotron light science. A XAS spectrum is typically
divided into two parts, with that extending 50 eV beyond the
absorption threshold being called extended X-ray absorption
fine structure (EXAFS) and that around the absorption
threshold X-ray absorption near-edge structure (XANES).
While the EXAFS part has been widely used to gain
information on the structure of complexes in solution, the
state of the art is quite different for what concerns the XANES.
Indeed, in the study of aqueous complexes, only a small
number of ab initio XANES calculations can be found in the
literature, while qualitative interpretations of XANES spec-
methods allows one to deeply control the deposition
parameters and select the best morphological properties for
6
,7
the final application. Cu O prepared in this way showed
2
exceptionally high activity toward photoelectrochemical water
5
splitting, but also poor stability, which, in turn, can be
8
overcome by the use of protective layers. To explain the
different microstructures found at different temperatures for
9
electrodeposited Cu O layers, de Jongh et al. proposed a
2
complex reaction mechanism for the reduction, involving hole
transfer from the lactate to the growing layer at high
temperatures and electron transfer from the surface defects of
the growing layer to the lactate solution at lower T. In order to
16−19
tra
are still employed. This is at least partially due to the
complicated nature of the systems. Moreover, only complexes
better understand the mechanisms of Cu O electrodeposition,
2
the structure and stability of the copper(II) lactate complex in
aqueous solution have to be investigated. On the basis of solid-
state structural determinations, lactate ions (L) are thought to
Received: March 3, 2017
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.7b00553
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. Illustration of the 3D-printed spectroelectrochemical cell. (A) Schematic representation of the two sides (1 and 4), the working electrode
2), and the O-ring (3). (B) Image of the cell with the three-electrode setup, mounted in the beamline hutch.
(
characterized by high symmetry have been examined. However,
XANES contains the finer details of the structure, and it is
capable, in principle, of retrieving the whole 3D structure
around the photoabsorber. The infrequency of structural
determination based on XANES is mainly due to the difficult
theoretical treatment required.
We also wanted to investigate the stability of the CuL
complex versus the electroxidative process. For this, we studied
the copper(II) complex in alkaline solution by linear sweep
voltammetry (LSV), starting from anodic potentials and
salt bridge made of a glass pipet filled with a gelled solution of 0.2 M
aqueous KClO4.
2
.3. XAS Experiment. XAS measurements were performed in the
22
20
transmission mode at the ID24 beamline at the European
Synchrotron Radiation Facility (ESRF), Grenoble, France (the ring
energy is 6.0 GeV, with a ring current ranging from 150 to 200 mA).
Figure 1B provides an image of the electrochemical cell with the three-
electrode setup mounted in the experimental hutch. The X-ray source
consists of two undulators whose gaps were adjusted to tune constant
counts of the first harmonic for energies around the Cu K-edge
spectra. The horizontal-beam focusing was achieved by means of a
curved polychromator [Si(111)] crystal in Bragg geometry, while the
vertical-beam focusing was obtained with a bent silicon mirror at a
glancing angle of 2.5 mrad with respect to the direct beam. The beam
size at the sample was focused at 8 μm full-width at half-maximum
reaching the potential for Cu O deposition (at about −0.4 V
2
vs Ag/AgCl). In this respect, energy-dispersive XAS (EDXAS)
is the technique of choice: as a matter of fact, in the energy-
dispersive mode, a XAS spectrum can be acquired in
milliseconds or less. The main drawbacks are a quite limited
signal-to-noise ratio and a loss in resolution.
(
fwhm) horizontally and defocused at 80 μm fwhm vertically in order
to decrease the photon density for preserving the samples with respect
to beam damage. The curved geometry allows the polychromator to
pass a broad band of energy through the sample and to diverge toward
In this work, we show, however, that fine and reliable details
on the structure of the copper(II) lactate complex in solution
can be obtained also starting form EDXAS data. Moreover,
thanks to the favorable time resolution of EDXAS, we were able
to obtain some insight into the electrochemical behavior of the
compound.
23
a linear photodiode-array detector (FreLonCCD camera ), where the
beam position can be correlated to the energy. XAS data were
collected in operando in order to investigate both the structure of the
copper(II) lactate complex in the starting solution and the pertinent
stability. The integration time chosen for the copper(II) lactate
solution single spectrum was 500 ms. The stability of the solution was
investigated by recording a sequence of spectra, while LSV was
performed on the CuL solution. In this case, we acquired 320 spectra
with an integration time of 500 ms for a total time equal to 160 s.
These values were chosen for a good compromise in order to avoid
saturation and to obtain an adequate signal-to-noise ratio. The LSV
measurements, from +1.2 to −0.4 V (vs Ag/AgCl) at 0.01 V/s, were
triggered by the spectral acquisition system through the trigger option
of the potentiostat (CH Instrument 630D). All measurements were
carried out at room temperature.
2. EXPERIMENTAL SECTION
2
.1. Preparation of the Solution. The copper(II) lactate
1
,5
solution was prepared by mixing 0.2 M CuSO (Sigma-Aldrich), 3
4
M lactic acid (Fischer Scientific), and 0.5 M K HPO (Sigma-Aldrich),
2
4
which acts as a buffer. This solution was subsequently adjusted to pH
=
12 by adding 2 M KOH.
2.2. Spectroelectrochemical Cell for Operando XAS Meas-
urements. In order to perform operando XAS measurements, a
spectroelectrochemical cell was designed and built by means of a 3D
printer. Figure 1A schematically shows the main components of the
electrochemical device. This consists of two parts: a polycarbonate disk
2.4. XAS Data Analysis. Two kinds of measurements were
performed in the work described in this paper: recording of a single
XAS spectrum, in order to gain information about the local structure
of the copper(II) lactate complex in aqueous solution, and acquisition
of a sequence of spectra during a LSV, from +1.2 to −0.4 V versus Ag/
AgCl, in order to investigate the electrochemical oxidation stability of
the copper lactate deposition bath.
21
(
1) and a Teflon hollow cylinder (21 mm diameter, 6 mm thickness,
and 7 mm cavity; 2). The working electrode (3) is a screen-printed
carbon electrode (Dropsens DRP-P-C11XX) loaded with gold
nanoparticles because gold is an effective substrate for Cu O
2
5
,8
deposition. The electrode is held between the main body of the
polycarbonate cell and the Teflon plate, which also includes a hole for
the X-ray beam through the use of an O-ring (4). The cell can contain
Analysis of a single spectrum is based on three main steps: (i)
preliminary processing and data extraction, (ii) EXAFS fitting, and (iii)
XANES theory calculation. The investigation of the sequence of XAS
spectra is constituted of (i) preliminary processing and data extraction
and (ii) automatic XANES fitting by linear combinations of spectra of
model compounds. For both measurements, the first step was carried
2
mL of electrolyte solution and can host a platinum wire counter
electrode and Ag/AgCl reference electrode. The platinum counter
electrode was shaped to form a hook just above the Dropsens working
electrode. The reference electrode is separated from the solution by a
2
4
25
out by means of the PyMCA and PRESTOPRONTO codes. It starts
B
DOI: 10.1021/acs.inorgchem.7b00553
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 2. Cu K-edge EXAFS signal (A and C) and the corresponding FT (B and D) of the starting solution of copper(II) lactate. The experimental
signal is represented by the blue spots, whereas the fit according to the models described in the text is represented by the red line. A and B refer to
the 4 + 2 model, while C and D refer to the 4 + 4 model.
from the evaluation of a preedge function and a postedge baseline
function that were applied to the raw signal. The former is associated
with additional excitations of the photoabsorber or to the possible
ment is employed (Figure S1). These results, which also show that the
XANES signal is very sensitive to the local geometry, are in good
3
0
agreement with recent literature, where the distorted octahedral and
square-pyramidal geometries have been identified with the most
probable local structures for the copper(II) hydrated ion. The best-
fitting parameters that have been obtained are illustrated in Tables S1
and S2. The results are in quite good agreement with those found in
excitation of any other element present (μ x) and was modeled with a
n
straight line calculated by interpolating the preedge data and then
extrapolated to the postedge XAS signal. The latter can be described as
the absorption profile that could be observed if the photoabsorber
3
1−36
were isolated [usually defined as μ (E)] and was obtained with a
the literature.
It should be noted that adding a skewness
0
second-order spline using the first inflection point of μ(E) as E0.
Thanks to these preliminary analyses, it was possible to normalize XAS
spectra and perform EXAFS extraction. For the sequence of spectra
acquired during LSV, data extraction was carried out by observing the
first spectrum and applying the same parameters to all of the spectra of
the series. EXAFS fitting on the extracted signals was carried out by
parameter (β) for some of the coordination shells was important for
improving the GOF. This is reasonable in light of the high fluxionality
3
6−38
claimed by several authors for hydrated copper(II).
We finally
note that the agreement between the experiment and calculations
shown here is by far better than any other found in the current
literature. As for the stability investigation, the variations along a
sequence of XAS spectra were evaluated by means of a XANES linear
26
means of the EXCURVE code. Calculation of the atomic scattering
parameters (phases and amplitudes) employed the muffin-tin
approximation, in the framework of the Hedin-Lundquist and Von
25
combination performed automatically with the software LinComb.
This is based on a comparison of the experimental spectrum with the
spectra of reference compounds. The experimental spectra μexp of a
data set are fitted by a theoretical curve μth calculated as a linear
combination of normalized reference compound spectra:
2
6
Bart approximations for the exchange and ground-state potentials,
respectively. This includes the effects of losses due to electron inelastic
scattering. The fittings were made using a full multiple scattering
(
(
FMS) calculation within the assumed clusters. As a goodness of fit
GOF), we used the F factor:
μ
= ∑ αμth
exp
j
i
(2)
χi,exp − χi,calc]2
n
[
The two reference compounds are the copper(II) lactate complex in
F = 100 ∑
the starting solution and standard Cu O. The time evolution of the
2
σi
(1)
i
parameters α allowed us to evaluate the percentage of copper(II)
i
lactate while performing the LSV measurements.
2
The fittings were made in the k space, using a k weighting scheme.
XANES calculations by means of EXCURVE software are based on
2
7
3. RESULTS AND DISCUSSION
a FMS approach to generate theoretical signals. The simulation is
obtained through the use of the exact calculation of the scattering path
operator. In order to validate the XANES calculation procedure
applied in this work, a calculation was performed to simulate the
XANES spectrum recorded on a 0.1 M solution of CuSO ·6H O
Figure 2 shows the Cu K-edge EXAFS signal of the starting
solution of copper(II) lactate (A) and the corresponding
Fourier transform (FT) (B). Oscillations above the noise level
are well apparent in the spectrum up to k = 8 Å . This allows
one to obtain reasonable quantitative parameters for the first
4
2
−1
standard. We think that the copper(II) hydrated ion is a very good test
system for our XANES fitting approach. In fact, because of the high
exchange rate of water molecules, it is not simple to obtain good
Very good simulations are obtained by
means of the distorted octahedral and square-pyramidal arrangements,
whereas large discrepancies arise if the trigonal-bipyramidal arrange-
coordination shells, within ca. 4 Å. Copper lactate is known to
28,29
yield square-planar complex CuL in the solid state, with lactate
theoretical simulations.
2
10
acting as a bidentate ligand. This structure is assumed for the
complex in solution as well. For what concerns the data
C
DOI: 10.1021/acs.inorgchem.7b00553
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
presented here, the presence of a clear beating in the EXAFS
−1
between 5 and 6 Å clearly indicates that a single frequency
cannot account for the experimental signal. Therefore, the
cluster employed for the initial fit is composed of two shells of
oxygen atoms: the first shell is made up of four oxygen atoms
situated at ca. 1.9 Å, which are provided by the carboxylate and
hydroxyl groups of lactate in a bidentate arrangement, and the
second shell is composed of two oxygen atoms situated at ca. 3
Å, provided by the carboxylate group. This large distance
implies that these two oxygen atoms are not directly bonded to
the central copper(II). The values obtained from the best-
fitting procedure are shown in Table 1, and the fit is reported in
Figure 2A,B as a full line.
Table 1. EXAFS Fitting Parameters for Spectra Shown in
a
Figure 2
2
2
shell
N
atom
r (Å)
σ (Å )
GOF F = 23.9%
1
2
4
2
oxygen
oxygen
1.87(2)
3.33(7)
10(2) × 10−
0(1) × 10−
3
2
GOF F = 23.1%
10(2) × 10−
8(9) × 10−
3
3
1
2
4
4
oxygen
oxygen
1.87(2)
3.35(6)
a
2
r = distances; σ = distance variances. Coordination numbers N were
kept constant to 4 + 2 and 4 + 4 as described in the text.
In the subsequent step of data analysis, represented by the
XANES simulation, the models tested for the aqueous complex
were constructed on the basis of the previous results.
Therefore, the XANES simulation was started by assuming a
square-planar coordination around copper(II) and two lactate
molecules acting as bidentate ligands (model 0, Figure 3B).
The results are shown in Figure 3A, showing the Cu K-edge
XANES calculations for the solution of copper(II) lactate:
although the overall shape of the XANES manifold is quite
satisfactorily interpreted, there are, however, large discrepancies
in the finer structure, leading to the conclusion that a different
structural model is required. Therefore, we constructed three
different models for the coordination geometry of copper in the
lactate solution: (i) model 1 (Figure 3C) comprises two lactate
ions acting as bidentate ligands and is characterized by a
distorted tetrahedral geometry; (ii) model 2 (Figure 3D) is
made up by two lactate ions acting as bidentate ligands in a
square-planar arrangement with two additional water molecules
in apical positions to yield an overall distorted octahedral
geometry; (iii) model 3 (Figure 3E) is composed of four lactate
ions acting as monodentate ligands to yield a coordination
number of 4 in a geometrical arrangement characterized by
point symmetry S4.
These models can be tested against the experimental data by
fitting the XANES. The first attempt concerns model 1, where
we removed the requirement of a strict square-planar geometry.
In this case, the XANES simulation (orange line) leads to better
but still unsatisfactory results. This led us to consider a model
in which two ligands act as bidentate ions in a square-planar
coordination and two oxygen atoms (water molecules) are
added in the apical positions to yield an overall distorted
octahedral geometry (model 2), thus removing the requirement
of having two oxygen atoms in the second coordination shell.
This model gives a little improvement in the XANES
simulation, indicated as a green line in Figure 3, compared to
that obtained for model 1, but the results are still far from an
Figure 3. (A) Cu K-edge XANES simulations of a copper(II) lactate
starting solution. The blue spots represent the experimental signal,
while the colored lines are the theoretical calculations according to the
models shown in parts B (pink line, model 0), C (orange line, model
1
), D (green line, model 2), and E (red line, model 3). The arrows
mark the energy regions where the largest differences between the
different models are found. An insight illustrating the derivative of
both the experimental and calculated spectra is included in order to
better visualize the most important discrepancies between the models.
optimal theoretical simulation. The blue arrows indicate the
energy regions where major discrepancies between the
theoretical and experimental signals are observed. These
discrepancies can be even better evidenced by looking at the
corresponding derivatives shown in Figure 3A. Thus, we
remove the constraint imposing the lactate anions to act as
bidentate ligands. Taking into account that even though lactate
anion is 15 times more concentrated than copper sulfate, steric
hindrance limits the maximum number of ligands around
copper in a single complex, and assuming the presence of four
lactate anions acting as monodentate ligands (model 3), we
obtain that the agreement between the experimental signal and
theoretical simulation (red line) is largely improved and all of
the relevant spectral features of the XANES manifold are
correctly simulated. We then propose model 3 for the structure
of copper(II) lactate in alkaline solution. Therefore, the
structure of copper(II) lactate in alkaline solution is different
10
from that assumed by copper(II) lactate in the solid state.
The best-fit results obtained by XANES calculation are
shown in Table 2. For each atom of the cluster used to calculate
the XANES theoretical spectrum, the distance from the
photoabsorber and the corresponding Debye−Waller factors
are illustrated. The distance of the nearest oxygen atoms and
D
DOI: 10.1021/acs.inorgchem.7b00553
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 2. XANES Calculation Parameters Employed To
Simulate the Experimental Signal Shown in Figure 3A
According to the Model Illustrated in Figure 3E
2
2
a
atom number type of atom distance from copper(II) (Å)
σ (Å )
GOF F = 1.3%
a
−3
1
2
3
4
5
6
7
8
9
1
1
oxygen
1.855(4)
2.977
3.000
3.234
3.720
3.919
4.351
4.704
4.871
5.492
5.583
6(2) × 10
6(2) × 10−
6(2) × 10−
6(2) × 10−
6(2) × 10−
3
3
3
3
3
3
3
3
3
3
hydrogen
carbon
oxygen
carbon
−
oxygen
6(2) × 10
hydrogen
hydrogen
carbon
6(2) × 10−
6(2) × 10−
Figure 4. Comparison of the simulations of copper(II) lactate XANES
according to the model shown in Figure 3E with (red line) and
without (green line) the hydrogen atoms. The experimental signal is
represented by the blue spots, and the experimental calculated
differences are illustrated in the bottom of the figure for the two cases.
The arrows mark the energy regions where the largest differences
between the models with and without hydrogen atoms are found.
−
6(2) × 10
0
1
hydrogen
hydrogen
6(2) × 10−
6(2) × 10−
a
Fitted values.
the Debye−Waller factors were fitted by treating the lactate
anions as rigid units. In addition, all of the Debye−Waller
factors have been treated as equal. This procedure is somewhat
arbitrary but has been used in order to keep the number of
fitted parameters to a minimum value to avoid correlations. The
fitting results are illustrated in Table 2, and all of the nearest
distances between carbon and oxygen atoms, in each lactate,
obtained by the theoretical calculation were checked by a
comparison with the literature. No significant differences are
detected, as illustrated by Table 3. Finally, the coordinates of all
of the atoms comprised in the cluster of model 3 are displayed
in Table S3.
lines) and absence (green lines) of hydrogen atoms in the
cluster. The inclusion of hydrogen atoms determines a
substantial variation in the simulated XANES spectrum, in
particular in the energy regions indicated by the arrows.
Looking at the difference signals between the theoretical
spectrum and the experimental one, we note that the
calculation carried out including hydrogen atoms leads to a
better fitting. This has the obvious meaning that multiple
scattering paths involving hydrogen atoms cannot be neglected
at least up to 40 eV above the absorption edge.
The electrochemical stability of the lactate solution was
investigated by LSV measurements from +1.2 to −0.4 V (vs
Ag/AgCl at 0.01 V/s), while acquiring the XAS spectra with a
time resolution of 500 ms. The total time for one LSV
measurement was 160 s.
Table 3. Nearest Distances between Carbon and Oxygen
Atoms in Each Lactate Molecule Present in Model E (Figure
3E): A Comparison between Fitted and Literature Values
Figure 5A shows some spectra of the sequence separated by
time intervals of ca. 25 s. A spectral weight increase of the
prepeak indicated by the arrows is clearly detected. A
comparison with the standard spectra presented in Figure 5B
Lactate Interatomic Distances
atoms
distance in the model (Å)
literature value (Å)
1
3
O −C3
1.241
1.507
1.219
1.429
1.507
1.530
1.219
1.507
1.219
1.429
1.507
1.530
C −C5
shows that, in this energy region, Cu O has a sizable preedge
2
C −O6
3
structure. Therefore, the increase of the preedge is attributed to
a partial reduction of copper(II) to copper(I). At a given time,
the XANES spectrum can be simulated with linear combina-
O −C5
4
C −C7
6
C −C9
5
tions of standard XANES. In particular, we used Cu O as a
2
standard for copper(I).
The fitting according to the structural model is good, as
The fitting range selected was 8950−9030 eV. The results
obtained from the fitting procedure by means of LinComb
software are reported in Figure 6. This shows the time
evolution of the coefficients related to the two standards used
for the linear combination. At time zero, the amount of
copper(II) lactate is equal to 100% and no copper(I) is present.
A small amount of copper(I) (10%) appears at a potential of ca.
1.1 V, increasing up to ca. 20% at 1 V. Then it remains almost
constant during the rest of the linear sweep. The automatic
linear combination fitting procedure was checked by manually
fitting four spectra at four potential values along LSV, finding
excellent agreement (Figure S2).
The initial formation of copper(I) is quite surprising because
the current flow is anodic up to an applied potential of ca. 0.4
V. Radiolysis phenomena generated by the X-ray beam are
irrelevant, as demonstrated by preliminary tests in the absence
of an applied potential, showing that the spectral shape remains
the same along the whole sequence. The rationale we propose
supported by a GOF F that is ca. 1.3%. We note that the
parameters obtained are in perfect agreement with the values
obtained by EXAFS fitting for the same oxygen atoms (see
Table 1).
It is also worth noting that the EXAFS can be fitted with a
structural model made up by a first- and second- coordination
shell of oxygen, both containing four atoms, with a slight
improvement in the fit quality (see Figure 2C,D) and with a
more reasonable value for the Debye−Waller factor of the
second shell (see Table 1). This is in agreement with the results
of the XANES fitting, and it is due to the quite poor accuracy of
EXAFS in the determination of the coordination numbers.
An important remark concerns the inclusion of hydrogen
atoms in the simulation of XANES. In XAS spectroscopy,
hydrogen atoms are intrinsically not well-defined because of
their small scattering factor. Figure 4 shows a comparison of
XANES calculations for the CuL complex in the presence (red
4
E
DOI: 10.1021/acs.inorgchem.7b00553
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
oxidation of the lactate ion and can act as strong reducing
agents on CuO, thus producing Cu O. However, this scenario
2
seems to be less probable. In fact, the anodic stability of a
lactate-based ionic liquid at room temperature was demon-
strated up to 1.0−1.3 V versus saturated calomel electrode
41
depending on the cations. In conclusion, the CuL complex
4
remains largely stable in a wide range of applied potentials.
4
. CONCLUSIONS
In this work, we used EDXAS to determine the structure and
electrochemical properties of the copper lactate complex in
alkaline solution. For what concerns the structure, our approach
consists of (i) first fitting the EXAFS in order to obtain
reasonable coordination numbers, (ii) formulating structural
models accounting for the coordination numbers obtained, and
finally (iii) testing the structural models against a fit of the
XANES.
We found that the best fit is obtained by a model where four
lactate ions act as monodentate ligands, 1:4 (Cu/L), in a
distorted tetrahedral geometry. This is in contrast with the
10
structural determination of the complex in the solid state,
according to which the stoichiometry is 1:2, but is quite
reasonable considering the large excess (15:1 L/Cu) of lactate
in the solution.
Figure 5. (A) Seven spectra selected from the sequence recorded on
the copper(II) lactate solution during the application of LSV and
separated by a time interval of ca. 25 s. (B) Cu K-edge XANES spectra
Concerning the electrochemical properties of the solution,
the sequence of spectra recorded during application of LSV
from +1.2 to −0.4 V (vs Ag/AgCl) showed that the modest
of the standards of copper(II) lactate (blue line), Cu O (brown line),
2
and metallic copper (black line).
production of Cu O is due to the copresence of metallic copper
2
II
on the counter electrode and Cu L in the solution. In any case,
4
we can confirm the electrochemical stability of the copper
lactate complex in the potential range scanned, 80% still present
at the end of LSV.
ASSOCIATED CONTENT
■
*
S
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.7b00553.
XANES fits for the copper(II) hydrated ion, manual
linear combination fits for the EDXAS data of the lactate
complex, and extended tables of coordinates and atomic
positions (PDF)
Figure 6. Results of the automatic linear combination fitting of the
XANES spectra of the sequence recorded during the application of
LSV. The blue dots represent the percentage of copper(II) lactate, the
red dots represent the percentage of Cu O, and the orange line is the
current recorded during LSV. In order to reduce the scattering of the
results, only every fifth spectrum was selected.
2
AUTHOR INFORMATION
■
Corresponding Author
E-mail: paolo.ghigna@unipv.it.
*
ORCID
for this complex behavior is as follows. At the beginning of LSV,
the potential is sufficiently anodic to oxidize water to oxygen, at
the working electrode, and this accounts for the anodic current.
The counter electrode acts therefore as a cathode, thus
reducing the amount of copper(II) in the solution to
copper(0). The presence of copper(0) on the counter electrode
Paolo Ghigna: 0000-0002-8680-7272
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
II
and of Cu L in the solution can then produce Cu O, as
4
2
39
Notes
described by Bijani for a similar complex, thus explaining the
XANES spectra of Figure 5. The amount of copper(I) then
remains constant during the rest of LSV because no reaction
involving copper species occurs at both cathode and anode
because of the smaller applied potentials. An alternative
explanation involves the possible formation of radicals because
of the presence of lactate anions, as was already observed by
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The ESRF is kindly acknowledged for the provision of
beamtime (Exp. CH-4409). We also acknowledge the ID24
beamline staff, and, in particular, Dr. Giovanni Agostini, for
considerable help during data collection and analysis.
40
Rubinstein and Bard. These radicals are generated upon
F
DOI: 10.1021/acs.inorgchem.7b00553
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
18) Frenkel, A. I.; Korshin, G. V.; Ankudinov, A. L. XANES study of
REFERENCES
■
2+
Cu -binding sites in aquatic humic substances. Environ. Sci. Technol.
000, 34, 2138−2142.
19) Brugger, J.; Etschmann, B.; Liu, W.; Testemale, D.; Hazemann,
(
1) Golden, T. D.; Shumsky, M. G.; Zhou, Y.; VanderWerf, R. A.;
Van Leeuwen, R. A.; Switzer, J. A. Electrochemical deposition of
copper (I) oxide films. Chem. Mater. 1996, 8, 2499−2504.
2) Katayama, J.; Ito, K.; Matsuoka, M.; Tamaki, J. Performance of
Cu2O/ZnO solar cell prepared by two-step electrodeposition. J. Appl.
Electrochem. 2004, 34, 687−692.
3) Siripala, W.; Perera, L. D. R. D.; De Silva, K. T. L.; Jayanetto, J. K.
D. S.; Dharmadasa, I. M. Study of annealing effects of cuprous oxide
grown by electrodeposition technique. Sol. Energy Mater. Sol. Cells
2
(
J. L.; Emerich, H.; van Beek, W.; Proux, O. An XAS study of the
structure and thermodynamics of Cu (I) chloride complexes in brines
up to high temperature (400 C, 600bar). Geochim. Cosmochim. Acta
(
2
(
007, 71, 4920−4941.
(
20) Benfatto, M.; Natoli, C. R.; Bianconi, A.; Garcia, J.; Marcelli, A.;
Fanfoni, M.; Davoli, I. Multiple-scattering regime and higher-order
correlations in x-ray-absorption spectra of liquid solutions. Phys. Rev.
B: Condens. Matter Mater. Phys. 1986, 34, 5774−5781.
1
(
996, 44, 251−260.
4) De Jongh, P. E.; Vanmaekelbergh, D.; Kelly, J. Photo-
electrochemistry of Electrodeposited Cu O. J. Electrochem. Soc. 2000,
(
21) Achilli, E.; Minguzzi, A.; Visibile, A.; Locatelli, C.; Vertova, A.;
2
Naldoni, A.; Rondinini, S.; Auricchio, F.; Marconi, S.; Fracchia, M.;
Ghigna, P. 3D-printed photo-spectroelectrochemical devices for in situ
and in operando X-ray absorption spectroscopy investigation. J.
Synchrotron Radiat. 2016, 23, 622−628.
(22) Pascarelli, S.; Mathon, O.; Munoz, M.; Mairs, T.; Susini, J.
Energy-dispersive absorption spectroscopy for hard-X-ray micro-XAS
applications. J. Synchrotron Radiat. 2006, 13, 351−358.
(23) Labiche, J. C.; Mathon, O.; Pascarelli, S.; Newton, M. A.; Ferre,
G. G. C.; Curfs, C.; Vaughan, G.; Homs, A.; Carreiras, D. F. The fast
readout low noise camera as a versatile x-ray detector for time resolved
dispersive extended x-ray absorption fine structure and diffraction
studies of dynamic problems in materials science, chemistry, and
catalysis. Rev. Sci. Instrum. 2007, 78, 091301.
1
(
47, 486−489.
5) Paracchino, A.; Brauer, J. C.; Moser, J.-E.; Thimsen, E.; Graetzel,
M. Synthesis and characterization of high-photoactivity electro-
deposited Cu O solar absorber by photoelectrochemistry and ultrafast
spectroscopy. J. Phys. Chem. C 2012, 116, 7341−7350.
(
2
6) Yang, Y.; Li, Y.; Pritzker, M. Control of Cu O Film Morphology
2
Using Potentiostatic Pulsed Electrodeposition. Electrochim. Acta 2016,
13, 225−235.
7) Zhao, W.; Fu, W.; Yang, H.; Tian, C.; Li, M.; Li, Y.; Zhang, L.;
2
(
Sui, Y.; Zhou, X.; Chen, H.; Zou, G. Electrodeposition of Cu O films
and their photoelectrochemical properties. CrystEngComm 2011, 13,
2
2
(
871−2877.
8) Paracchino, A.; Laporte, V.; Sivula, K.; Gratzel, M.; Thimsen, E.
̆
Highly active oxide photocathode for photoelectrochemical water
reduction. Nat. Mater. 2011, 10, 456−461.
(
(24) Sole, V. A.; Papillon, E.; Cotte, M.; Walter, Ph; Susini, J. A
́
multiplatform code for the analysis of energy-dispersive X-ray
fluorescence spectra. Spectrochim. Acta, Part B 2007, 62, 63−68.
9) de Jongh, P. E.; Vanmaekelbergh, D.; Kelly, J. J. Cu O:
2
Electrodeposition and Characterization. Chem. Mater. 1999, 11, 3512−
517.
10) Prout, C. K.; Armstrong, R. A.; Carruthers, J. R.; Forrest, J. G.;
(25) Figueroa, S. J. A.; Prestipino, C. PrestoPronto: a code devoted
3
(
to handling large data sets. J. Phys.: Conf. Ser. 2016, 712, 012012.
26) Binsted, N.; Hasnain, S. S. State-of-the-art analysis of whole X-
ray absorption spectra. J. Synchrotron Radiat. 1996, 3, 185−196.
27) Fujikawa, T. Basic features of the short-range-order multiple
scattering XANES theory. J. Phys. Soc. Jpn. 1993, 62, 2155−2165.
28) La Penna, G.; Minicozzi, V.; Morante, S.; Rossi, G. C.; Stellato,
(
Murray-Rust, P.; Rossotti, F. J. C. Structure and stability of carboxylate
complexes. Part I. The crystal and molecular structures of copper(II)
glycollate, DL-lactate, 2-hydroxy-2-methylpropionate, methoxyacetate,
and phenoxyacetate. J. Chem. Soc. A 1968, 2791−2813.
(
(
(
11) Baran, T.; Fracchia, M.; Vertova, A.; Achilli, E.; Naldoni, A.;
2+
F. A first-principle calculation of the XANES spectrum of Cu in
Malara, F.; Rossi, G.; Rondinini, S.; Ghigna, P.; Minguzzi, A.;
D’Acapito, F. Operando and Time-Resolved X-Ray Absorption
Spectroscopy for the Study of Photoelectrode Architectures. Electro-
chim. Acta 2016, 207, 16−21.
water. J. Chem. Phys. 2015, 143, 124508.
(
29) Chaboy, J.; Munoz-Paez, A.; Merkling, P. J.; Sanchez Marcosa,
̃
́
́
2+
E. The hydration of Cu : can the Jahn-Teller effect be detected in
liquid solution? J. Chem. Phys. 2006, 124, 064509.
(
12) Baran, T.; Wojtyła, S.; Lenardi, C.; Vertova, A.; Ghigna, P.;
(30) Manceau, A.; Matynia, A. The nature of Cu bonding to natural
Achilli, E.; Fracchia, M.; Rondinini, S.; Minguzzi, A. An Efficient CuxO
Photocathode for Hydrogen Production at Neutral pH: New Insights
from Combined Spectroscopy and Electrochemistry. ACS Appl. Mater.
Interfaces 2016, 8, 21250−21260.
organic matter. Geochim. Cosmochim. Acta 2010, 74, 2556−2580.
+
2+
aq
(
31) Blumberger, J. Cu /Cu
redox reaction exhibits strong
aq
nonlinear solvent response due to change in coordination number. J.
Am. Chem. Soc. 2008, 130, 16065−16068.
(
13) Rondinini, S.; Minguzzi, A.; Achilli, E.; Locatelli, C.; Agostini,
(
32) Liu, X.; Lu, X.; Jan Meijer, E.; Wang, R. Hydration mechanisms
G.; Pascarelli, S.; Spinolo, G.; Vertova, A.; Ghigna, P. The dynamics of
pseudocapacitive phenomena studied by Energy Dispersive X-Ray
Absorption Spectroscopy on hydrous iridium oxide electrodes in
alkaline media. Electrochim. Acta 2016, 212, 247−253.
2+
of Cu : tetra-, penta-or hexa-coordinated? Phys. Chem. Chem. Phys.
010, 12, 10801−10804.
33) Liu, X.; Cheng, J.; Sprik, M. Aqueous Transition-Metal Cations
2
(
2+
+
2+
+
as Impurities in a Wide Gap Oxide: The Cu /Cu and Ag /Ag
(
14) Minguzzi, A.; Locatelli, C.; Cappelletti, G.; Scavini, M.; Vertova,
Redox Couples Revisited. J. Phys. Chem. B 2015, 119, 1152−1163.
A.; Ghigna, P.; Rondinini, S. IrO -based Disperse-Phase Electro-
2
(34) Frank, P.; Benfatto, M.; Qayyam, M.; Hedman, B.; Hodgson, K.
catalysts: a Complementary Study by Means of the Cavity-Micro-
electrode and the ex-Situ X-ray Absorption Spectroscopy The. J. Phys.
Chem. A 2012, 116, 6497−6504.
O. A high-resolution XAS study of aqueous Cu (II) in liquid and
frozen solutions: Pyramidal, polymorphic, and non-centrosymmetric. J.
Chem. Phys. 2015, 142, 084310.
(
15) Garcia, J.; Benfatto, M.; Natoli, C. R.; Bianconi, A.; Fontaine, A.;
2+
(35) Amira, S.; Spangberg, D.; Hermansson, K. Distorted five-fold
Tolentino, H. The quantitative Jahn-Teller distortion of the Cu site
in aqueous solution by XANES spectroscopy. Chem. Phys. 1989, 132,
coordination of Cu2
+
from a Car−Parrinello molecular dynamics
(aq)
simulation. Phys. Chem. Chem. Phys. 2005, 7, 2874−2880.
36) Blumberger, J.; Bernasconi, L.; Tavernelli, I.; Vuilleumier, R.;
2
(
95−302.
(
16) Frank, P.; Benfatto, M.; Qayyam, M.; Hedman, B.; Hodgson, K.
Sprik, M. Electronic structure and solvation of copper and silver ions: a
theoretical picture of a model aqueous redox reaction. J. Am. Chem.
Soc. 2004, 126, 3928−3938.
O. A high-resolution XAS study of aqueous Cu(II) in liquid and frozen
solutions: Pyramidal, polymorphic, and non-centrosymmetric. J. Chem.
Phys. 2015, 142, 084310.
(
17) Benfatto, M.; Solera, J. A.; Chaboy, J.; Proietti, M. G.; Garcìa, J.
(37) Pasquarello, A.; Petri, I.; Salmon, P. S.; Parisel, O.; Car, R.;
Theoretical analysis of X-ray absorption near-edge structure of
transition-metal aqueous complexes in solution at the metal K edge.
Phys. Rev. B: Condens. Matter Mater. Phys. 1997, 56, 2447−2452.
́
Toth, E.; Powell, D. H.; Fischer, H. E.; Helm, L.; Merbach, A. E. First
solvation shell of the Cu(II) aqua ion: evidence for fivefold
coordination. Science 2001, 291, 856−859.
G
DOI: 10.1021/acs.inorgchem.7b00553
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
38) Curtiss, L. A.; Halley, J. W.; Wang, X. R. Jahn-Teller effect in
liquids: General principles and a molecular-dynamics simulation of the
cupric ion in water. Phys. Rev. Lett. 1992, 69, 2435−2438.
(
39) Bijani, S.; Schrebler, R.; Dalchiele, E. A.; Gabas, M.; Martinez,
L.; Ramos-Barrado, J. R. Study of Nucleation and growth mechanism
in the electrodeposition of micro-and nanostructured Cu O thin films.
2
J. Phys. Chem. C 2011, 115, 21373−21382.
(
40) Rubinstein, I.; Bard, A. J. Electrogenerated Chemiluminescence.
2+
3
7. Aqueous Ecl systems based on Ru(2,2′-bippyridine)3 and Oxalate
or organic acids. J. Am. Chem. Soc. 1981, 103, 512.
41) Meenatchi, B.; Renuga, V.; Manikandan, A. Electrodeposition of
(
Nickel on Glassy Carbon Electrode from Protic Ionic Liquids with
Imidazolium Cation. J. Inorg. Organomet. Polym. Mater. 2016, 26, 423.
H
DOI: 10.1021/acs.inorgchem.7b00553
Inorg. Chem. XXXX, XXX, XXX−XXX