Solvation Structure of Zn2+ and Cu2+ Ions in ... - ACS Publications

Feb 24, 2015 - the coordination numbers and polyhedral environment, while the analysis ... existence of an axially elongated square pyramidal coordina...
0 downloads 0 Views 828KB Size
Download PDF
Article pubs.acs.org/JPCB

Solvation Structure of Zn2+ and Cu2+ Ions in Acetonitrile: A Combined EXAFS and XANES Study Paola D’Angelo* and Valentina Migliorati

Downloaded via UNIV OF CALIFORNIA SANTA BARBARA on July 14, 2018 at 18:57:40 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

Dipartimento di Chimica, Università di Roma “La Sapienza”, P.le A. Moro 5, 00185 Roma, Italy ABSTRACT: The solvation structure of Zn2+ and Cu2+ in acetonitrile has been determined by a combined approach using both X-ray absorption near edge structure (XANES) and the extended X-ray absorption fine structure (EXAFS) spectroscopy. For the former cation, an octahedral geometry of the acetonitrile solvate complex has been found with a Zn−N distance of 2.12(1) Å. For the Cu2+ solvates the EXAFS technique has been found to be not able to provide a conclusive determination of the coordination numbers and polyhedral environment, while the analysis of the XANES spectra unambiguously shows the existence of an axially elongated square pyramidal coordination, ruling out the previously proposed octahedral Jahn−Teller (JT) distorted geometry. The Cu−N distances obtained are 2.00(1) and 2.28(2) Å for the equatorial and axial ligands, respectively, and the EXAFS and XANES techniques find values of the bond distances in good agreement. The XANES technique has proven to be extremely powerful in providing a reliable resolution of solution structure for dynamic ion complexes.

■

investigations agreed on this different picture of the Cu2+ hydration structure.7,17−23 The solvation structure of the Cu2+ ion has been also studied in some nonaqueous solvents, and even if the former studies suggested the formation of distorted octahedral complexes in some oxygen-donating solvents, such as methanol (MeOH) and dimethyl sulfoxide (DMSO),2−5 a more recent X-ray absorption near edge structure (XANES) investigation unambiguously showed the presence of an average 5-fold coordination in both the MeOH and DMSO solutions.24 Square pyramidal five-coordinate Cu2+ has also been found to be the predominant structure in aqueous ammonia or imidazole solutions.20,25 As far as the structure of Cu2+ in ACN solution is concerned, an EXAFS and large-angle X-ray scattering investigation carried out by Persson et al.26 suggested the formation of JT distorted octahedral complexes, but no solvate molecules could be detected in the axial positions by any of the techniques used. Recently, a MD study has been carried out on Cu2+ ions in ACN, and the distorted octahedral structure derived from the EXAFS measurements could not be reproduced in the simulations.15 Because an accurate determination of the solvation geometry of Zn2+ and Cu2+ in liquid ACN is still lacking, we decided to use the X-ray absorption spectroscopy (XAS) that has been found to be a reliable tool to provide local structural information around a selected atom in several solutions.27−31 EXAFS is an effective tool able to provide accurate information on bond distances but it is not conclusive in the determination of the coordination numbers and geometries.32 Conversely,

INTRODUCTION After iron, zinc and copper are the most abundant metals in the human body, and they play an essential role in many biological processes. This has led to the publication of several studies on the solvation properties of these ions in water, while investigations on nonaqueous solutions are less abundant.1−5 Acetonitrile (ACN) is an aprotic solvent with good solvating properties for many hydrophobic and hydrophilic compounds, and it also forms stable solvates that are used in coordination and organometallic chemistry. Because most of the chemical processes involving these solvates take place in solution, knowledge about the solvation structure is important to understand their behavior and reactivity. Previous investigations on the solvated structure of Zn2+ cations in oxygen donor solvents pointed to the existence of octahedral complexes.1,6−13 An extended X-ray absorption fine structure (EXAFS) study of the in-solution structure of Zn2+ ACN solvates was not conclusive, and from the experimental results it was not possible to prove the existence of a stable octahedral structure, even if the Zn−N observed bond distance (2.11 Å) was quite reasonable for a 6-coordinate geometry.14 A recent molecular dynamics (MD) study confirmed the existence of a stable octahedral structure for the coordination complex of the Zn2+ ion in ACN solution, but the Zn−N firstshell distance was quite short (1.99 Å).15 For a long time in the past, all studies on the solution structure of Cu2+ ions in water agreed on the existence of Jahn−Teller (JT) distorted complexes because of the d9 electronic configuration of this ion.1 However, in 2001 a neutron diffraction and MD study by Pasquarello et al.16 proposed the presence of a 5-fold coordination for the [Cu(aq)]2+ complex, and ever since this study several other © 2015 American Chemical Society

Received: January 20, 2015 Revised: February 22, 2015 Published: February 24, 2015 4061

DOI: 10.1021/acs.jpcb.5b01634 J. Phys. Chem. B 2015, 119, 4061−4067

Article

The Journal of Physical Chemistry B

ments of the EXAFS function χ(k) and do not include systematic errors of the measurements.

XANES determines 3D structures with high accuracy due to the higher structural sensitivity of the low-energy region of the XAS spectrum. Here we report a combined EXAFS and XANES structural study of dissolved Cu2+ and Zn2+ ions in ACN solution that allowed a reliable and conclusive determination of the ion coordination geometries in this solvent. This combined approach has been found to be very effective, and it can be extended to the study of different ions in organic and biological media.

Table 1. Best-Fit Parameters Obtained from the Analysis of the EXAFS Spectra of Zn2+ and Cu2+ in ACN Solutiona NC

σ2(Å2)

β

0.012(3)

0.0(1)

2.00(1) 2.27(2)

0.005(1) 0.018(3)

0.0(1) 0.1(1)

2.00(1) 2.28(2)

0.005(1) 0.008(3)

0.0(1) 0.1(1)

R(Å) Zn

6-Fold Coordination Zn−N 6

■

6-Fold Coordination Cu−Neq 4 Cu−Nax 2 5-Fold Coordination Cu−Neq 4 Cu−Nax 1

MATERIALS AND METHODS X-ray Absorption Measurements. 0.1 M solutions of Cu2+ and Zn2+ in ACN were prepared by dissolving the appropriate amount of Cu(CF3SO3)2 and Zn(CF3SO3)2 in pure ACN (Aldrich). Cu and Zn K-edge X-ray absorption spectra were collected at the EMBL spectrometer at DESY. Spectra were recorded at room temperature in transmission mode using a Si(111) double-crystal monochromator detuned to 30% for harmonic rejection. The spectra were recorded with a 5 eV spacing refining to a minimum of 0.1 eV in the edge and then increasing again with δk = 0.02 Å−1. All spectra were collected up to k = 14 Å−1. The DORIS III storage ring was running at an energy of 4.4 GeV with positron currents between 120 and 90 mA. The solutions were kept in a cell with Kapton film windows and a Teflon spacer of 2 mm. EXAFS Data Analysis. The GNXAS code was used for the EXAFS data analysis.33,34 In this approach, the EXAFS χ(k) signal is decomposed into a summation over γ(n) signals associated with n-body distribution functions calculated by means of the multiple-scattering (MS) theory. The distribution of the ion−solvent distances has been modeled with Γ-like distribution functions that depend on four parameters, namely, the coordination number, NC, the average distance, R, the mean-square variation, σ2, and the skewness, β. Note that β is related to the third cumulant C3 through the relation C3 = σ3β. The EXAFS spectrum of Zn2+ in ACN has been analyzed assuming an octahedral first-shell coordination. The theoretical signal included six Zn−N two-body γ(2) contributions, three N−Zn−N three body γ(3) signals, and six Zn−N−C η(3) contributions. This last term accounts for both the Zn···C two-body γ(2) and the Zn−N−C three-body γ(3) signals, as these contributions have a similar frequency and depend on the same structural parameters, namely the Zn−N distance, the N− C distance, and the Zn−N−C angle. For this reason, these two contributions are added in a total signal that is indicated as η(3). The Cu2+ EXAFS spectrum has been calculated considering four ACN molecules in the equatorial plane and either one or two solvent ligands in the axial positions. The total theoretical signal was calculated including four Cu−Neq two-body γ(2) signals, one or two Cu−Nax γ(2) contributions, two Neq−Cu− Neq three-body γ(3) signals, four Cu−Neq−Ceq η(3) signals, and either one or two Cu−Nax−Cax η(3) signals. In this case, the refined structural parameters are the Cu−Neq and Cu−Nax bond distances and the Cu−Neq−Ceq, Cu−Nax−Cax, and Neq− Cu−Neq bond angles. Least-squares fits of the EXAFS raw experimental data have been performed by minimizing a residual function, Rsq. (See ref 33 for details.) Additional nonstructural parameters were minimized, namely, E0 (core ionization threshold energy) and S20. The standard deviations given for the refined parameters in Table 1 are obtained from k2-weighted least-squares refine-

2+

in ACN

2.12(1) Cu2+ in ACN

NC is the coordination number, R is the interatomic distance, σ2 is the Debye−Waller factor, and β is the asymmetry parameter.

a

XANES Data Analysis. The MXAN code has been used to perform the XANES data analysis, and details of the theoretical framework can be found in refs 35 and 36. The MXAN method uses an empirical approach to account for inelastic losses in which the core hole lifetime, Γc, the plasmon energy onset, Es, and amplitude, As, are refined. In all analyses the Γc value was fixed to 1.55 and 1.67 eV for Cu and Zn, respectively, while the experimental resolution was taken into account by convolution with a Gaussian function. The analysis of the XANES spectrum of Zn2+ in ACN has been carried out starting from an octahedral model with six ACN ligands at the same distance. During the minimization the ACN geometry was kept fixed, while the orientation of the ligands was varied within a preset range of ±10° around the initial geometry. The Zn−N distance was optimized on the basis of the experimental data. In the case of the Cu2+ solution, different possible complexes have been tested and the Cu−Neq and Cu−Nax distances have been optimized together with the Nax−Cu−Neq angle. Hydrogen atoms have been included in all MXAN analyses. The quality of the fits has been estimated with the residual function, Rsq. (See refs 35 and 36.) For both spectra, five nonstructural parameters have been optimized, namely, the Fermi energy level, EF, the experimental resolution, Γexp, the threshold energy, E0, and energy and amplitude of the plasmon, Es and As.

■

RESULTS EXAFS Analysis. Zn2+ in Acetonitrile Solution. Starting from previous results, the analysis of the EXAFS spectrum of Zn2+ in ACN solution has been carried out assuming the existence of an octahedral geometry.14,15 Least-squares fits of the EXAFS spectrum were performed in the range k = 3.2−13.2 Å−1, improving, as far as possible, the agreement between calculated signal and experimental spectrum. In particular, only three structural parameters were optimized, namely, the Zn−N distance and the Zn−N−C and N−Zn−N angles, while the geometry of the ACN molecules was kept fixed. The results of the minimization procedure are shown in the upper panel of Figure 1. The first three curves from the top are the Zn−N twobody signal and the MS signals associated with the Zn−N−C and N−Zn−N three-body configurations. The reminder of the Figure shows the comparison between the total theoretical contribution and the experimental data and the resulting residuals. The EXAFS spectrum is dominated by the Zn−N 4062

DOI: 10.1021/acs.jpcb.5b01634 J. Phys. Chem. B 2015, 119, 4061−4067

Article

The Journal of Physical Chemistry B

as the nonstructural parameters are concerned, E0 was found 7 eV above the first inflection point of the experimental spectrum, while S20 was equal to 1. Cu2+ in Acetonitrile Solution. The EXAFS analysis of the Cu2+ solution is more challenging because no information is available on the geometry of the solvated complexes. In the first step, a variable number of ligands in the axial sites and in the equatorial plane has been used to assess the sensitivity of EXAFS toward the coordination numbers and distances. In particular, a statistical analysis using 2D contour plots to selected parameters of the fit was carried out to establish error limits. Details of this procedure can be found in ref 33. Contour plots of the Cu−Neq distance versus E0, together with the Cu− Neq and Cu−Nax coordination numbers and Debye−Waller (DW) factors, are shown in Figure 2, where the innermost contour refers to the 95% error confidence interval. From this Figure, the picometer accuracy of the EXAFS technique in determining the first coordination shell distances is evident. Conversely, the correlation between coordination numbers and the DW factors is quite high with an error of ±0.5 around the best-fit value. Looking at the results of Figure 2, it is clear that while the 4-fold coordination in the equatorial plane can be evinced from the EXAFS analysis, the error on the number of ACN molecules bound to the Cu2+ ion in the axial sites is too big to provide a reliable determination of the solvate complex geometry. Starting from these results the analysis of the EXAFS spectrum of Cu2+ in ACN has been carried out using two different models with four ACN molecules in the equatorial plane and either one or two axial ligands, reproducing an elongated square pyramidal or a JT distorted octahedral configuration, respectively. The best-fit analysis of the EXAFS spectrum obtained with the square pyramidal model is shown in Figure 3, as an example, and the results obtained from the JT distorted octahedral geometry are almost identical. During the minimization the coordination numbers and the metrics of the ACN molecules were kept fixed. In particular, the analysis of the spectrum has been carried out including the following contributions: the Cu−Neq and Cu−Nax two-body signals, the Cu−Neq−Ceq and Cu−Nax−Cax three-body contributions, and the Neq−Cu−Neq three-body signal. The MS contribution associated with the Nax−Cu−Nax configuration in the JT distorted octahedral model has been found to provide a negligible contribution. All of the theoretical signals are shown in the upper panel of Figure 3 together with the comparison between the total theoretical spectrum and the experimental data and the residual curve. The minimizations have been carried out in the k range 3.8−13.2 Å−1. The lower panel of Figure 3 shows the k2-weighted FT calculated with no phase

Figure 1. Upper panel: Experimental Zn K-edge EXAFS spectrum of Zn2+ in ACN solution and theoretical signal calculated for an octahedral geometry. The following curves are shown: the Zn−N firstshell signal, the Zn−N−C, and the N−Zn−N three-body signals, the experimental data (red, dotted line) compared with the theoretical spectrum (blue, solid line), and the residuals. Lower panel: Fourier transform (not corrected for phase shift) of the experimental (red, dotted line) and theoretical (blue, solid line) k2 EXAFS signals.

first-shell contribution and by the MS signals associated with the Zn−N−C configuration that contains also the Zn···C two body signal, while the amplitude of the N−Zn−N MS contribution is quite low. The agreement between the fitted and experimental spectra is good, proving the validity of the octahedral model. The lower panel of Figure 1 shows the k2weighted Fourier transform (FT) calculated with no applied phase-shift correction in the k range of 3.2−13.2 Å−1. The FT spectrum shows a prominent first-shell peak that is mainly due to the Zn−N first-shell distances and a separated peak at ∼3 Å that is mainly due to the MS contribution associated with the six Zn−N−C distributions of the ACN ligands. Refined structural values of parameters obtained from this analysis are listed in Table 1. The Zn−N distance and σ2 values are in perfect agreement with previous EXAFS results (2.11 and 0.0152 Å2),14 while the determination obtained from MD simulations is quite shorter (Zn−N distance of 1.99 Å).15 As far

Figure 2. Contour plots obtained from the EXAFS analysis of the Cu K-edge spectrum of Cu2+ in ACN solution. (A) Cu−Neq distance versus E0. (B) Cu−Neq coordination number versus Debye−Waller factor. (C) Cu−Nax coordination number versus Debye−Waller factor. 4063

DOI: 10.1021/acs.jpcb.5b01634 J. Phys. Chem. B 2015, 119, 4061−4067

Article

The Journal of Physical Chemistry B

Figure 4. Zn K-edge XANES experimental signal of Zn2+ in ACN solution (black, dotted line) compared with the theoretical curve (red, full line) calculated with an octahedral model. The best-fit geometry is also depicted, where zinc, nitrogen, carbon, and hydrogen atoms are in gray, blue, orange, and white, respectively.

octahedron with six nitrogen atoms at 2.13(2) Å. This value is in good agreement with the EXAFS determination and with previous results.14,15 The calculated model matches the experimental data very well, and the Rsq value is 1.2; a full list of the nonstructural parameters obtained from the analysis is reported in Table 2. The agreement between the XANES and

Figure 3. Upper panel: Experimental Cu K-edge EXAFS spectrum of Cu2+ in ACN solution and theoretical signal calculated for a square pyramidal geometry. The following curves are shown: the Cu−Neq and Cu−Nax first-shell signals, the Cu−Neq−Ceq, Cu−Nax−Cax, and Neq− Cu−Neq three-body signals, the experimental data (red, dotted line) compared with the theoretical spectrum (blue, solid line), and the residuals. Lower panel: Fourier transform (not corrected for phase shift) of the experimental (red, dotted line) and theoretical (blue, solid line) k2 EXAFS signals.

Table 2. Nonstructural Parameters Obtained from the MXAN Analysis of Zn and Cu K-Edge XANES Spectra of Zn2+ and Cu2+ in ACN Solutiona

shift correction applied in the k range 3.8−13.2 Å−1. From the two analyses a similar agreement between the experimental and theoretical spectra was obtained and the Ri values were 3.9 × 10−5 and 4.0 × 10−5 for the 5-fold and 6-fold coordination, respectively. The structural parameters obtained from the fitting procedures are reported in Table 1. In both analyses the Cu−Neq distance is 2.00(1) Å, in agreement with previous results,26 while the DW factor of the Cu−O axial ligands increases in going from the 5-fold to the 6-fold geometry, due to the correlation between the coordination numbers and the bond length variance that was evidenced by the correlation maps reported in Figure 2. E0 was found 5 eV above the first inflection point of the experimental spectrum in both cases, while S20 was equal to 1 and 0.8 for the 5-fold and 6-fold models, respectively. This finding confirms the insensitivity of EXAFS toward the number of solvate molecules in the Cu2+ axial positions.26 XANES Analysis. Quantitative analysis of the XANES spectra of Zn2+ and Cu2+ in ACN solution has been carried out with the aim of performing a conclusive determination of the ACN solvate complex structures. We resorted to the use of XANES because this technique is especially suited to determine the geometric environment of an atom embedded in a disordered system. Zn2+ in Acetonitrile Solution. To definitely prove that the Zn2+ ion in ACN solution forms stable octahedral complexes we have carried out the analysis of the XANES data starting from an octahedral coordination geometry. The results of the minimizations are depicted in Figure 4. The best-fit structure obtained from the minimization procedure corresponds to an

E0 (eV) octahedron JT distorted octahedron elongated square pyramid

EF (eV)

Zn2+ in ACN 4.7 −5.8 Cu2+ in ACN 1.6 −2.8 1.0 −2.8

Γexp (eV)

Es (eV)

As

2.8

19.1

15.1

2.9 3.2

9.2 22.1

15.0 13.1

E0 is the threshold energy, EF is the Fermi energy level, Γexp is the experimental resolution, and Es and As are the energy and amplitude of the plasmon. a

EXAFS first-shell bond lengths and the very good reproduction of the experimental spectrum suggest that the Zn2+ ion forms an octahedral complex in ACN solution. Cu2+ in Acetonitrile Solution. As previously shown, the EXAFS data analysis is inconclusive in the determination of the coordination structure of the Cu2+ in ACN solution. In particular, the EXAFS spectrum can be reproduced with the same accuracy using both 5-fold and 6-fold models. To shed light on the geometry of the Cu2+ solvated complex, we analyzed the XANES spectrum that is expected to be more sensitive to the number of ligand molecules in the axial sites. In the first step the compatibility of the XANES spectrum with the existence of a JT distorted octahedral complex in ACN has been assessed by performing a minimization of the experimental data while imposing an octahedral geometry. In the minimization procedures the Cu−N eq and Cu−N ax distances and the Nax−Cu−Neq angle were moved, while the ACN geometry has been kept fixed. In Figure 5 we report the best-fit results where the theoretical curve corresponds to a JT 4064

DOI: 10.1021/acs.jpcb.5b01634 J. Phys. Chem. B 2015, 119, 4061−4067

Article

The Journal of Physical Chemistry B

Figure 6. Cu K-edge XANES experimental signal of Cu2+ in ACN solution (black, dotted line) compared with the theoretical curve (red, full line) calculated with an elongated square pyramidal model. The best-fit geometry is also depicted where copper, nitrogen, carbon, and hydrogen atoms are in green, blue, orange, and white, respectively.

2+

Figure 5. Cu K-edge XANES experimental signal of Cu in ACN solution (black, dotted line) compared with the theoretical curve (red, full line) calculated with a JT distorted octahedral model. The best-fit geometry is also depicted where copper, nitrogen, carbon, and hydrogen atoms are in green, blue, orange, and white, respectively.

EXAFS determinations. The XANES analysis shows that the four ACN molecules are not coplanar and the Nax−Cu−Neq angles are >90°. In this case, because of the lack of an axial ligand the equatorial ACN molecules undergo a displacement from the mean equatorial plane, as previously also found in water.25 The nonstructural parameters are reported in Table 2, and they are quite similar to the values obtained from the previous analysis. All together these results demonstrate that in ACN solution the Cu2+ ion adopts a 5-fold coordination with a distorted square pyramidal configuration. By comparing the theoretical XANES spectra of the octahedral model in Figure 5 with the 5-fold one in Figure 6, it is evident that the two spectra are quite different in the energy region between 12 and 40 eV. This finding indicates that at variance with the EXAFS spectroscopy XANES is sensitive to the solvent molecules placed in the axial sites. This result is very important because no other experimental techniques are able to detect the axial ligands in solution.

distorted octahedral model with four ACN ligands in the equatorial plane at 1.98(2) Å and two ACN molecules in the axial site at 2.13(4) Å. (See Table 3.) As shown in Figure 5 the Table 3. Cu K-Edge XANES Structural Parameters of Cu2+ in ACN Solutiona

JT distorted octahedron elongated square pyramid

Cu−Neq (Å)

Cu−Nax (Å)

Nax−Cu−Neq (deg)

Rsq

1.98(2) 2.03(2)

2.13(4) 2.32(4)

91(3) 108(3)

4.2 1.9

a

Cu−Neq and Cu−Nax are the distances between the ion and the equatorial and axial nitrogen atoms, respectively, Nax−Cu−Neq is the angle among the axial nitrogen, copper, and equatorial nitrogen atoms, and Rsq is the residual function.

■

agreement between the theoretical curve and the experimental spectrum is quite poor, displaying a clear mismatch in the energy region between 12 and 40 eV from the edge. Note that in this case the residual function value is quite high (Rsq = 4.2) while the nonstructural parameters obtained from this analysis are listed in Table 2. It is important to note that the Cu−Nax distance obtained is quite shorter as compared with the EXAFS determination. These findings suggest that, as previously found in water and other organic solvents,16−20,24,25 the Cu2+ ion does not adopt a 6-fold coordination in ACN solution. A second coordination scheme was considered to analyze the XANES data in which the ACN ligands adopted an elongated square pyramidal configuration around the Cu2+ ion. During the minimization, four Cu−Neq and one Cu−Nax distances and the Nax−Cu-Neq angles were optimized, while the ACN geometry was kept fixed. The results of this analysis are shown in Figure 6, and in this case the agreement between the theoretical and experimental spectrum is very good in the whole energy range (Rsq = 1.9). The best-fit structure is a square pyramidal 5-fold complex with the axial nitrogen atom at 2.32(4) Å and four nitrogen atoms in the equatorial pane at 2.03(2) Å. The best-fit structural model is reported in Figure 6, while the obtained structural parameters are listed in Table 3 and are in good agreement, within the statistical errors, with the

DISCUSSION AND CONCLUSIONS We have studied the solvation structure of Zn2+ and Cu2+ in ACN solution by combining the EXAFS and XANES techniques. In the case of the Zn2+ ion an octahedral geometry was hypothesized by previous investigations, and this result has been confirmed by the present analysis. The structure of the solvation complex has been found to be very similar to the one previously determined by EXAFS, while the Zn−N distance obtained by a recent MD simulation is too short.14,15 As far as the Cu2+ ion is concerned, the results obtained here are different from those previously published.15,26 In particular, the existence of a JT distorted octahedral geometry for the [Cu(ACN)]2+ complex could be discarded on the basis of the XANES analysis, while the EXAFS analysis has been found to be not conclusive. The XANES experimental spectrum could be properly reproduced using an axial elongated square pyramidal model, while the theoretical curve obtained with a JT distorted octahedral complex did not match the experimental data in the energy region close to the edge. This result is justified by the fact that the low-energy region of the spectrum is dominated by MS signals whose amplitude is very large when dealing with three-body atomic configurations with angles close to 180°, 4065

DOI: 10.1021/acs.jpcb.5b01634 J. Phys. Chem. B 2015, 119, 4061−4067

Article

The Journal of Physical Chemistry B owing to the focusing effect. For this reason the presence of the Nax−Cu−Nax configuration in the octahedral structure gives rise to a modification of the XANES spectrum as compared with the 5-fold one, which allows one to discriminate between the two models. Note that the Nax−Cu−Nax configuration was found to provide a negligible contribution to the EXAFS region of the spectrum, thus explaining why this technique is not able to detect the number of ligands coordinated in the axial sites. In the EXAFS data analysis the configurational average has to be treated carefully for linear geometries. In particular, in the collinear case also if the average positions of the atoms are aligned by definition, in a real vibrating system the probability of finding exactly a linear configuration is zero, as it vanishes like sin(θ) as the spherical volume element. In the case of a P− A−B linear configuration, where P is the photoabsorber and A and B are two scattering atoms, in the GNXAS method the γ(3) signal depends on the P−A and A−B distances, on the intervening angle, θ, and on four covariance matrix parameters, namely, σ2A, σ2B, ρAB, and σ2θ.33 σ2θ provides information on the angular fluctuations, and for disordered linear configurations, linear fluctuations can be confused with a bent average configuration. In the case of Zn2+ in ACN, a σ2θ = 6°2 has been obtained for the Zn−N−C configuration corresponding to an angular fluctuation of the solvent molecules of ∼2.5° around the linear configuration. Such a fluctuation is accounted for in the XANES analysis by the slightly bent best-fit geometry obtained, as shown in Figure 4. For the Cu2+ ion the σ2θ values obtained from the EXAFS minimization are 6 and 12°2 for the Cu−Nax−Cax and Cu−Neq−Ceq configurations, respectively. This finding is in agreement with the XANES analysis that shows the presence of a best-fit geometry, with the ACN molecules undergoing a displacement from the equatorial plane. It is well known that Zn2+ and Cu2+ have different coordinating properties, and while the former ion forms stable complexes, the latter gives rise to more labile species due to the weakly bounded axial ligands in rapid fluctionality with the bulk molecules. This is a direct consequence of the 3d9 electronic structure of the Cu2+ atomic shell, which causes a departure from octahedral coordination because of the JT effect. Moreover, XANES detects the so-called diffusion-averaged structure, that is, the structure averaged with respect to the whole time of the experiment. Therefore, the Cu2+ 5-fold coordination has to be considered in the framework of a dynamical picture in which the axial ligands undergo a fast exchange with the second-shell solvent molecules. The present results are in line with the Cu2+ coordination determined in water, methanol, and DMSO but also in nitrogen donor media such aqueous ammonia or imidazole solutions.16−20,24,25 As a consequence, the axially elongated square pyramidal motif seems to be the preferred solvation structure of Cu2+. Flexure of the equatorial ligands has also been detected in water, providing a structural picture of the known fluxional dynamics of the dissolved Cu2+ ion.25 Lastly, XANES has proven to be extremely powerful in resolving the structure of cations in solution, and this is important because neither other experimental techniques nor computational methods have been able, up to now, to provide a reliable resolution of solution structure for dynamic ion complexes. Note that a recent MD simulation of Cu2+ in ACN was unable to reproduce the 3d9-induced distortion, and an unrealistic regular octahedral geometry has been obtained with the developed Lennard-Jones parameters.15

The present results can stimulate further experimental and theoretical work to better rationalize the solution structure of ions in nonaqueous solutions.

■

AUTHOR INFORMATION

Corresponding Author

*Tel: +39 0649913751. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Federico Riccitelli for his invaluable help. This work was supported by the University of Rome ”La Sapienza” (Progetto ateneo 2014, n.C26A14L7CX) and by the CINECA supercomputing centers through the grant IsC23_GEDI (n.HP10CLG9ZC).

■

REFERENCES

(1) Ohtaki, H.; Radnai, T. Structure and Dynamics of Hydrated Ions. Chem. Rev. 1993, 93, 1157−1204. (2) Inada, Y.; Sugimoto, K.; Ozutsumi, K.; Funahashi, S. Solvation Structures of Manganese(II), Iron(II), Cobalt(II), Nickel(II), Copper(II), Zinc(II), Cadmium(II), and Indium(III) Ions in 1,1,3,3Tetramethylurea As Studied by EXAFS and Electronic Spectroscopy. Variation of Coordination Number. Inorg. Chem. 1994, 33, 1875− 1880. (3) Ozutsumi, K.; Koide, M.; Suzuki, H.; Ishiguro, S. Solvation Structure of Divalent Transition-Metal Ions in N,N-dimethylformamide and N,N-dimethylacetamide. J. Phys. Chem. 1993, 97, 500−502. (4) Ozutsumi, K.; Abe, Y.; Takahashi, R.; Ishiguro, S. Chloro and Bromo Complexation of the Manganese(II) Ion and Solvation Structure of the Manganese(II), Iron(II), Cobalt(II), Nickel(II), Copper(II), and Zinc(II) Ions in Hexamethylphosphoric Triamide. J. Phys. Chem. 1994, 98, 9894−9899. (5) Inada, Y.; Hayashi, H.; Sugimoto, K.; Funahashi, S. Solvation Structures of Manganese(II), Iron(II), Cobalt(II), Nickel(II), Copper(II), Zinc(II), and Gallium(III) Ions in Methanol, Ethanol, Dimethyl Sulfoxide, and Trimethyl Phosphate As Studied by EXAFS and Electronic Spectroscopies. J. Phys. Chem. A 1999, 103, 1401−1406. (6) D’Angelo, P.; Barone, V.; Chillemi, G.; Sanna, N.; MeyerKlaucke, W.; Pavel, N. V. Hydrogen and Higher Shell Contributions in Zn2+, Ni2+, and Co2+ Aqueous Solutions: An X-ray Absorption Fine Structure and Molecular Dynamics Study. J. Am. Chem. Soc. 2002, 124, 1958−1967. (7) D’Angelo, P.; Benfatto, M.; Della Longa, S.; Pavel, N. V. Combined XANES and EXAFS Analysis of Co2+, Ni2+, and Zn2+ Aqueous Solutions. Phys. Rev. B 2002, 66, 064209−7. (8) Migliorati, V.; Mancini, G.; Chillemi, G.; Zitolo, A.; D’Angelo, P. Effect of the Zn2+ and Hg2+ Ions on the Structure of Liquid Water. J. Phys. Chem. A 2011, 115, 4798−4803. (9) Migliorati, V.; Zitolo, A.; Chillemi, G.; D’Angelo, P. Influence of the Second Coordination Shell on the XANES Spectra of the Zn2+ Ion in Water and Methanol. ChemPlusChem. 2012, 77, 234−239. (10) Migliorati, V.; Mancini, G.; Tatoli, S.; Zitolo, A.; Filipponi, A.; De Panfilis, S.; Di Cicco, A.; D’Angelo, P. Hydration Properties of the Zn2+ Ion in Water at High Pressure. Inorg. Chem. 2013, 52, 1141− 1150. (11) Cauet, E.; Bogatko, S.; Weare, J. H.; Fulton, J. L.; Schenter, G. K.; Bylaska, E. J. Structure and Dynamics of the Hydration Shells of the Zn2+ Ion from Ab Initio Molecular Dynamics and Combined Ab Initio and Classical Molecular Dynamics Simulations. J. Chem. Phys. 2010, 132, 194502. (12) Migliorati, V.; Chillemi, G.; D’Angelo, P. On the Solvation of the Zn2+ Ion in Methanol: A Combined Quantum Mechanics, Molecular Dynamics, and EXAFS Approach. Inorg. Chem. 2011, 50, 8509−8515. 4066

DOI: 10.1021/acs.jpcb.5b01634 J. Phys. Chem. B 2015, 119, 4061−4067

Article

The Journal of Physical Chemistry B

(31) Mancini, G.; Sanna, N.; Barone, V.; Migliorati, V.; D’Angelo, P.; Chillemi, G. Structural and Dynamical Properties of The Hg2+ Aqua Ion: A Molecular Dynamics Study. J. Phys. Chem. B 2005, 109, 9186− 9193. (32) Hayakawa, K.; Hatada, K.; D’Angelo, P.; Della Longa, S.; Natoli, C. R.; Benfatto, M. Full Quantitative Multiple-Scattering Analysis of Xray Absorption Spectra: Application to Potassium Hexacyanoferrat(II) and -(III) Complexes. J. Am. Chem. Soc. 2004, 126, 15618−15623. (33) Filipponi, A.; Di Cicco, A. X-Ray-Absorption Spectroscopy and N-Body Distribution Functions in Condensed Matter. II. Data Analysis and Applications. Phys. Rev. B 1995, 52, 15135−15149. (34) Filipponi, A.; Di Cicco, A.; Natoli, C. R. X-Ray-Absorption Spectroscopy and N-Body Distribution Functions in Condensed Matter. I. Theory. Phys. Rev. B 1995, 52, 15122−15134. (35) Benfatto, M.; Della Longa, S. Geometrical Fitting of Experimental XANES Spectra by a Full Multiple-Scattering Procedure. J. Synchrotron Radiat. 2001, 8, 1087−1094. (36) Benfatto, M.; Della Longa, S.; Natoli, C. R. The MXAN Procedure: A New Method for Analysing the XANES Spectra of Metalloproteins to Obtain Structural Quantitative Information. J. Synchrotron Radiat. 2003, 10, 51−57.

(13) Zitolo, A.; D’Angelo, P. X-Ray Absorption Spectroscopy Study of The Solvation Structure of Zinc(II) in Dimethyl Sulfoxide Solution. Chem. Phys. Lett. 2010, 499, 113−116. (14) Inada, Y.; Niwa, Y.; Iwata, K.; Funahashi, S.; Ohtaki, H.; Nomura, M. Solvation Structure of Metal Ions in Nitrogen-Donating Solvents. J. Mol. Liq. 2006, 129, 18−24. (15) Torras, J.; Aleman, C. Determination of New Cu+, Cu2+, and Zn2+ Lennard-Jones Ion Parameters in Acetonitrile. J. Phys. Chem. B 2013, 117, 10513−10522. (16) Pasquarello, A.; Petri, I.; Salmon, P. S.; Parisel, O.; Car, R.; Toth, E.; Powell, D. H.; Fischer, H. E.; Helm, L.; Merbach, A. First Solvation Shell of the Cu(II) Aqua Ion: Evidence for Fivefold Coordination. Science 2001, 291, 856−859. (17) Blumberger, J.; Bernasconi, L.; Tavernelli, I.; Vuilleumier, R.; 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. (18) Frank, P.; Benfatto, M.; Szilagyi, R. K.; D’Angelo, P.; Della Longa, S.; Hodgson, K. O. The Solution Structure of [Cu(aq)]2+ and Its Implications for Rack-Induced Bonding in Blue Copper Protein Active Sites. Inorg. Chem. 2005, 44, 1922−1933. (19) Xia, F. F.; Yi, H.-B.; Zeng, D. Hydrates of Cu2+ and CuCl+ in Dilute Aqueous Solution: A Density Functional Theory and Polarized Continuum Model Investigation. J. Phys. Chem. A 2010, 114, 8406− 8416. (20) Frank, P.; Benfatto, M.; Hedman, B.; Hodgson, K. O. Solution [Cu(amm)]2+ is a Strongly Solvated Square Pyramid: A Full Account of the Copper K-edge XAS Spectrum Within Single-Electron Theory. Inorg. Chem. 2008, 47, 4126−4139. (21) Benfatto, M.; D’Angelo, P.; della Longa, S.; Pavel, N. V. Evidence of Distorted Fivefold Coordination of The Cu2+ Aqua Ion From an X-Ray-Absorption Spectroscopy Quantitative Analysis. Phys. Rev. B 2002, 65, 1742051−1742055. (22) Bordiga, S.; Groppo, E.; Agostini, G.; van Bokhoven, J. A.; Lamberti, C. Reactivity of Surface Species in Heterogeneous Catalysts Probed by In Situ X-ray Absorption Techniques. Chem. Rev. 2013, 113, 1736−1850. (23) Garino, C.; Borfecchia, E.; Gobetto, R.; van Bokhoven, J. A.; Lamberti, C. Determination of The Electronic and Structural Configuration of Coordination Compounds by Synchrotron-radiation Techniques. Coord. Chem. Rev. 2014, 277−278, 130−186. (24) Zitolo, A.; Chillemi, G.; D’Angelo, P. X-ray Absorption Study of the Solvation Structure of Cu2+ in Methanol and Dimethyl Sulfoxide. Inorg. Chem. 2012, 51, 8827−8833. (25) Frank, P.; Benfatto, M.; Hedman, B.; Hodgson, K. O. The X-ray Absorption Spectroscopic Model of the Copper(II) Imidazole Complex Ion in Liquid Aqueous Solution: A Strongly Solvated Square Pyramid. Inorg. Chem. 2012, 51, 2086−2096. (26) Persson, I.; Penner-Hahn, J. E.; Hodgson, K. O. An EXAFS Spectroscopic Study of Solvates of Copper(I) and Copper(II) in Acetonitrile, Dimethyl Sulfoxide, Pyridine, and Tetrahydrothiophene Solutions and A Large-angle X-ray Scattering Study of the Copper(II) Acetonitrile Solvate in Solution. Inorg. Chem. 1993, 32, 2497−2501. (27) Chillemi, G.; Mancini, G.; Sanna, N.; Barone, V.; Della Longa, S.; Benfatto, M.; Pavel, N. V.; D’Angelo, P. Evidence for Sevenfold Coordination in The First Solvation Shell of Hg(II) Aqua Ion. J. Am. Chem. Soc. 2007, 129, 5430−5436. (28) D’ Angelo, P.; Zitolo, a.; Migliorati, V.; Persson, I. Analysis of The Detailed Configuration of Hydrated Lanthanoid(III) Ions in Aqueous Solution and Crystalline Salts by Using K- And L3-Edge XANES Spectroscopy. Chem.Eur. J. 2010, 16, 684−692. (29) D’Angelo, P.; Migliorati, V.; Mancini, G.; Chillemi, G. A Coupled Molecular Dynamics and XANES Data Analysis Investigation of Aqueous Cadmium(II). J. Phys. Chem. A 2008, 112, 11833−11841. (30) Spezia, R.; Duvail, M.; Vitorge, P.; Cartailler, T.; Tortajada, J.; Chillemi, G.; D’Angelo, P.; Gaigeot, M.-P. A Aoupled Car-Parrinello Molecular Dynamics and EXAFS Data Analysis Investigation of Aqueous Co2+. J. Phys. Chem. A 2006, 110, 13081−13088. 4067

DOI: 10.1021/acs.jpcb.5b01634 J. Phys. Chem. B 2015, 119, 4061−4067