TD-DFT study of the electronic absorption spectra of iron(III) monoisothiocyanate

Polyhedron 90 (2015) 41–46

Contents lists available at ScienceDirect

Polyhedron journal homepage: www.elsevier.com/locate/poly

TD-DFT study of the electronic absorption spectra of iron(III) monoisothiocyanate  naite˙ a, Olegas Eicher-Lorka b,⇑ Erika Elijošiute˙ a, Dalia Janku a b

Department of Environmental Technology, Kaunas University of Technology, Radvile˙nu˛ av. 19, LT-50254 Kaunas, Lithuania Institute of Chemistry, Center for Physical Sciences and Technology, Goštauto 9, LT-01108 Vilnius, Lithuania

a r t i c l e

i n f o

Article history: Received 13 November 2014 Accepted 31 January 2015 Available online 7 February 2015 Keywords: Iron(III) thiocyanate Solvation UV–Vis spectroscopy DFT NTO

a b s t r a c t The electronic structure of the iron(III) monoisothiocyanate complex has been studied using density functional theory (DFT) calculations. Experimental and simulated electronic absorption spectra have been analyzed. The electronic transitions have been identified by means of natural transition orbital (NTO) analysis. The effect of different solvation models (explicit and/or implicit water environment) and position of SO24 ligand versus NCS ligand upon its geometry and electronic absorption spectrum has been evaluated. Obtained results show that TD-PBE1PBE//PBE1PBE calculation when [Fe(NCS)]2+ complex is explicitly solvated with the SO24 ligand beside the NCS ligand is in the best agreement with the experimental [Fe(NCS)]2+ spectrum. NTO analysis indicates that the nature of transitions is mainly characterized as ligand-to-metal (LMCT) charge transfer, where NCS and SO24 ligands are the main ones responsible for the appeared absorbance peaks in the entire spectrum. Ó 2015 Published by Elsevier Ltd.

1. Introduction In the resent studies thiocyanate-containing metal complexes are considered to be the most investigated systems because of their diverse structures and applications in the field of materials science [1]. Due to characteristic features of the iron(III)–thiocyanate complex, it is applicable to a range of quantitative analysis of lipid containing materials which are interesting for biomedicine, food and industry [2]. Currently published papers refer to the application of iron(III)–thiocyanate system for the determination of pharmaceuticals [3,4]. Moreover, iron(III)–thiocyanate complex plays an important role in the solvent extraction of gold from thiocyanate solutions [5,6]. According to van Staden et al. [7], the first investigation of the aspects of iron(III) thiocyanate complexation reaction appeared in 1931. Later studies were mainly focused on the formation kinetics and mechanisms of iron(III) thiocyanate complexes [8,9]. The formation equilibria of the pure iron(III) thiocyanate and the structure of the iron(III) thiocyanate having organic or inorganic part were characterized by various techniques such as nuclear magnetic resonance spectroscopy (NMR), extended X-Ray Absorption Fine Structure (EXAFS) spectroscopy [10,11]. Recent

⇑ Corresponding author. Tel.: +370 52729372. E-mail address: [email protected] (O. Eicher-Lorka). http://dx.doi.org/10.1016/j.poly.2015.01.034 0277-5387/Ó 2015 Published by Elsevier Ltd.

studies combine experimental and theoretical (DFT) studies on the structure of iron(III) complexes with various ligands [12,13]. The formation of iron(III) thiocyanate complexes can also be characterized applying UV–Vis spectroscopic technique. Initial studies employing UV–Vis spectroscopy provided some data on the formation of iron(III) thiocyanate complexes. It was concluded that in the excess of thiocyanate ion, the principal absorbing species is [Fe(NCS)6]3 . In the excess of iron(III) ions the [Fe(NCS)]2+ complex is presented as the only absorbing species [7]. Additionally, the stability of formed [Fe(NCS)n]3-n (where n = 1–6) complexes largely depends on the concentration of reacting component, solution acidity and the presence of a non-aqueous solvent [14]. A red [Fe(NCS)]2+ complex is widely found in the aqueous solutions employed for the determination of chloride, iodine, mercury and etc. ions in clinical, industrial and scientific laboratories [15,16]. It should be emphasized that, despite the extensive experimental studies so far carried out for the [Fe(NCS)n]3 n complexes, the coordination structure of Fe(III) ion in the iron(III) isothiocyanate complex has not yet been determined. Many factors such as: the nature of central atom, steric factors, symbiosis of ‘hard and soft’ ligands in the coordination sphere [17,18], solvent effects [19,20], metal charge, and other coordinated ligands play an important role in the coordination chemistry, especially when metal ion belongs to the group of transition metal [17]. Recent study on the thermodynamics of Fenton reactions [21] has confirmed the introduction of sulfate ligand SO24 into the first coordination sphere of iron(II) and

42

E. Elijošiute˙ et al. / Polyhedron 90 (2015) 41–46

iron(III) aqua complexes in the aqueous solutions of sulfuric acid. In this context, our recently published results [22] of combined experimental and theoretical study on the molecular structure and vibrational spectra of iron(III) monoisothiocyanate complex supplemented the findings of researches [21]. It was demonstrated that at the pH 2, the NCS , SO24 ions and H2O molecules are found as ligands in the first coordination sphere of the Fe3+ ion. Therefore, the goal of this work was to perform a research on the molecular structure and electronic excitations of iron(III) monoisothiocyanate presented in the aqueous solution of sulfuric acid by means of theoretical modeling and experimental UV–Vis spectroscopy. 2. Experiment 2.1. Reagents All chemicals were of analytical-reagent grade, purchased from Sigma Aldrich and used without further purification. Acidified distilled water was used as the solvent throughout the present study in order to avoid the hydrolysis of iron(III) salt. 2.2. Procedure 2.2.1. UV spectroscopic analysis The solution of [Fe(NCS)]2+ complex with the metal(III)–ligand molar ratio of 1:1, in the level of 10 4 M, was prepared by mixing the necessary volumes of the stock solutions of 0.5 M Fe2(SO4)3 and 1 M KSCN. For the dilution of mixture to a known volume distilled water was used. UV–Vis spectra of the sample were acquired on Lambda 35 UV spectrometer (PerkinElmer) in the range of 200– 700 nm using 1 cm quartz cuvette. The pH of the aqueous working solution was adjusted to 2 ± 0.1 by controlled addition of sulfuric acid. pH measurements were performed using the pH meter HI 9321 (HANNA Instruments). 2.3. Quantum chemical calculations All calculations were performed using the GAUSSIAN for Windows 03 (D.01) [23] and 09 (A.02) packages [24]. The geometry optimization calculations were accomplished with the DFT method, using the unrestricted B3LYP [25], CAM-B3LYP [26] and PBE1PBE [27] functionals in the sextet high spin state (S = 5/2) [21]. For all atoms except the Fe, which was treated with pseudopotentials, the 6-31++G(d,p) basis set was used. The Stuttgart relativistic ECP 10MDF basis set for the Fe atom was applied. In order to evaluate the solvent effect, some calculations were performed using the polarizable continuum model (PCM), specifically integral equation formalism (IEF) model, further referenced as IEFPCM.

Additionally, explicit solvation model was taken into account. Geometry optimization calculations with varied quantity (n = 1–4) of explicitly added water molecules were performed. Our recently published results [22] showed that calculated Raman spectra correlates well with the experimental one when the energy minimized structure with explicitly added three H2O molecules was taken into account. In this context, the electronic spectra of energy minimized structure of Fe(III) monoisothiocyanate with explicitly added three H2O molecules (Fig. 1) are further analyzed in this manuscript. The electronic spectra of Fe(III) monoisothiocyanate complex were calculated with the B3LYP, CAM-B3LYP, PBE1PBE and PBEPBE [28] methods using TD-DFT approach on optimized geometries. The virtual and occupied orbitals involved in the electronic transitions were defined by the natural transition orbital (NTO) analysis [29]. For the visualization of electronic spectra the Chemcraft graphical program was employed [30]. Spectroscopy data were processed and managed using the GRAMS/AI spectroscopy software [31].

3. Results and discussion 3.1. Geometry optimization In order to determine the main features of the electronic absorption spectra of [Fe(NCS)]2+ complex in the acidic aqueous solution as well as to assign the photoactive singlet excited states, firstly the geometry optimization of complex using different solvation models (hypothetical cases) was performed. The explicitly (the water molecules are placed around the simulated solute molecule) or explicitly–implicitly (the solvation effects are presented as interaction between the solute and the solvent through a dielectric continuum along with the placed around water molecules) added water molecules in the first coordination shell were used. Moreover, the influence of SO24 and NCS ligand arrangement on the structural changes and final electronic spectrum of the [Fe(NCS)]2+ complex was evaluated. The different positions of the SO24 versus NCS ligand are shown in Fig. 1. Selected bond lengths and angles of each energy minimized structure along with the values found in literature are listed in the Table 1. All simulated complexes have the octahedral geometry. Iron(III) ion is six-coordinated by one N atom of NCS ligand, two and three O atoms of SO24 and H2O ligands, respectively. Calculated Fe–N bond distances are shorter by 0.02–0.14 Å compared with the experimental values listed in literature (Table 1). It should be noted that all values of experimentally determined bond distances taken from literature are determined for the crystal structures mostly having an organic part. Nevertheless, it is clear that explicit-implicit solvation induces elongation of the bond between Fe and N atoms compared with the explicitly solvated

Fig. 1. Energy-minimized structures of [Fe(NCS)]2+ complex using PBE1PBE method. (a) [Fe(NCS)]2+ explicitly solvated, with the SO24 ligand beside the NCS ligand, (b) [Fe(NCS)]2+ explicitly solvated, with the SO24 ligand on the opposite of the NCS ligand.

43

E. Elijošiute˙ et al. / Polyhedron 90 (2015) 41–46

Table 1 Selected bond lengths (Å) and angles (°) for the [Fe(NCS)]2+ complex in different hypothetical cases using B3LYP, CAM-B3LYP, and PBE1PBE methods: 1 – [Fe(NCS)]2+ explicitly solvated, with the SO24 ligand beside the NCS ligand, 2 – [Fe(NCS)]2+ explicitly-implicitly solvated, with the SO24 ligand beside the NCS ligand, 3 – [Fe(NCS)]2+ explicitly solvated, with the SO24 ligand on the opposite of the NCS ligand, 4 – [Fe(NCS)]2+ explicitly-implicitly solvated, with the SO24 ligand on the opposite of the NCS ligand. Complex

Experimental B3LYP Explicit 1

Bond length (Å) Fe–N 1.90 * Fe–O(SO)4 1.98 * Fe–O(H2O) 2.22 N@C 1.20 C@S 1.59 Angles (°) Fe–N@C 176.0 N@C@S 179.5 * **

CAM-B3LYP Explicit– implicit 2 1.95 2.05 2.14 1.19 1.61 177.6 179.8

Explicit 3 1.90 2.01 2.19 1.20 1.59 178.6 179.9

Explicit– implicit 4 1.95 2.05 2.13 1.19 1.61 176.9 179.9

Explicit 1 1.89 1.97 2.18 1.19 1.59 176.3 179.4

PBE1PBE

Explicit– implicit 2

Explicit 3

1.95 2.01 2.12 1.18 1.61 171.7 179.7

1.89 1.99 2.15 1.19 1.59 178.8 179.9

Explicit– implicit 4 1.94 2.05 2.14 1.18 1.61 176.1 179.9

Explicit 1 1.89 1.97 2.19 1.20 1.59 176.7 179.5

Explicit– implicit 2 1.95 2.03 2.12 1.19 1.61 179.2 179.9

Explicit 3 1.89 1.99 2.16 1.20 1.58 178.8 179.9

Explicit– implicit 4 1.95 2.03 2.11 1.19 1.60 177.4 179.9

2.03 [32], *1.97 [33] 1.94 [36], *2.01 [37] * 2.01 [34], **2.27 [35] 1.15 [32], *1.17 [33] 1.63 [32], *1.62 [33] 176.0 [32], *177.2 [33] 179.6 [32], *178.9 [33]

Average value. Calculated value.

complex. Calculated length of Fe–O(SO4) and Fe–O(H2O) bonds of each optimized structure is in the range of experimentally identified bond length given by other authors (Table 1). Comparing the calculated Fe–O bond distances it was observed that different solvation models cause the opposite effect on the changes of Fe–O(SO4) and Fe–O(H2O) bonds. The Fe–O(SO4) bond becomes weaker when the explicit–implicit solvation is used, while the Fe–O(H2O) bond always experiences bond strengthening. The tendency is the same for the N@C and C@S bonds. The explicit– implicit solvation slightly strengthens the N@C bond, while the C@S bond is slightly weakened. Calculated values of the N@C and C@S bond distances are close to the experimental (Table 1). Moreover, it can be seen that the calculated Fe–N@C and N@C@S angles correlate well with the listed experimental values. Based on the results of calculation the Fe–NCS group is essentially linear. The Fe–N@C angle is in the range of 176.0–179.2°. The more linear structure is characteristic for the NCS group in all hypothetical cases. The bond angles of N@C@S in each case also do not fluctuate markedly and is in the range of 179.4–179.9° (by all three B3LYP, CAM-B3LYP, and PBE1PBE calculations). Generally, it can be stated that the presence of the linear N@C@S group, strong bond between the N and C atoms, and weak bond between the C and S atoms confirm the high-spin ground state of [Fe(NCS)]2+ complex [31]. Generally, it was found that neither used B3LYP, CAM-B3LYP, and PBE1PBE methods nor different SO24 ligand arrangement do not have significant influence on bond distances. Only in the case of Fe–N@C angle data some differences could be observed (Table 1). Comparing the changes of bond distances between the central Fe atom and the surrounding O(SO4), O(H2O) and N atoms of each hypothetical case, it can be seen that the molecule is more compact when the SO24 ligand is beside the NCS ligand, especially using only explicit solvation model. 3.2. Electronic spectra In this study the performance of the different classes of functionals on the calculation of the UV–Vis spectra along with different DFT methods used for the geometry optimization was evaluated. Moreover, the effect of different solvation models and different ligand arrangement on the final electronic spectrum was also considered. The results of performed calculations revealed that TD-PBEPBE approach applied on the optimizations using B3LYP, CAM-B3LYP, and PBE1PBE methods was in the worst agreement with the

experimentally observed spectrum. As far as the visible region is considered, the calculations using the TD-PBEPBE functional displayed several intensive absorption bands. Furthermore, in the higher wavelengths region (over 550 nm), the high intensity band appeared which was not characteristic for the experimental spectrum. Moreover, it was found that the effect of simultaneous application of explicit and implicit solvation model does not meet spectral features of the experimental spectrum. The calculated maximum in the visible region of the simulated spectrum usually was the most intensive peak in the entire spectrum and exhibited the shift towards longer wavelengths region (red-shift) compared with the experimental one. On the contrary, blue-shifted absorption maximum was usually observed in the UV region. Based on the simulation data it was also found that the theoretical absorption spectra when the SO24 ligand is on the opposite site of the NCS ligand did not correspond to the experimental. In most cases the maximum of main absorption band in the visible region shifted to the longer wavelengths up to 70 nm. Besides, simulated spectra with such ligand arrangement were not able to reproduce spectral features of the experimental spectrum in the region below 280 nm. Based on these findings and omitting the results which engendered in quite big discrepancies, more detailed comparison between experimental and calculated electronic spectra when SO24 ligand is beside the NCS ligand with explicit solvation model was performed. The experimental UV spectrum together with TD-DFT calculated ones for the complex 1 are presented in the Figs. 2–4. Experimental spectrum gave one absorption band in the visible region with the maximum at 453 nm (Fig. 2). The absorption band of much higher intensity was observed in the near UV part with the peak maximum at 301 nm (Fig. 2). Additionally, experimental spectrum presents an intense band with the shoulder in the region of the spectrum extending from 210 nm to 270 nm with the approximate maximum at 216 nm. Comparing all TD-DFT simulated and experimental absorption spectra (Figs. 2–4), it is clear that regardless of the method used for the geometry optimization, the TD-B3LYP functional performed the worst agreement with the experimental spectrum. One can see that the absorption maximum in the visible part is quite red-shifted with the deviation from the experimental maximum of about 25–40 nm (Fig. 2). Moreover, this absorption band tends to have a double-peak character with the second peak maximum in the range of 490–530 nm, which is not characteristic neither for experimental spectrum nor for the TD-CAM-B3LYP and

44

E. Elijošiute˙ et al. / Polyhedron 90 (2015) 41–46

Fig. 2. The TD-B3LYP calculated absorption spectrum of explicitly solvated [Fe(NCS)]2+, with the SO24 ligand beside the NCS ligand, using different DFT methods on geometry optimizations. The inset shows magnified view of experimental band in the visible region.

Fig. 3. The TD-CAM-B3LYP calculated absorption spectrum of explicitly solvated [Fe(NCS)]2+, with the SO24 ligand beside the NCS ligand, using different DFT methods on geometry optimizations. The inset shows magnified view of experimental band in the visible region.

TD-PBE1PBE simulated spectra. If we analyze the UV part of the spectrum, we can see that this part also does not match with the experimental absorption spectrum. The simulated absorption spectrum using TD-CAM-B3LYP approach on all three optimized structures experiences a less pronounced deviation (in the range of 15–25 nm) from the experimental maximum in visible region compared with TD-B3LYP (Fig. 3). Unlike the apparent red-shift in the visible part, caused by the used TD-B3LYP functional, the blue-shift using the TD-CAM-B3LYP was always identified in the calculated spectra. Analyzing the UV region, we can see that the TD-CAM-B3LYP functional gives absorption bands with a general shape and tendency that are in a reasonable agreement with the experimental spectrum. Actually, the third TD-PBE1PBE functional used for all analyzed DFT methods was in the best agreement with the experimentally obtained spectrum, even if the reproduction of the band shape and intensities was not ideal (Fig. 4). The deviation is 1 nm and 10 nm for PBE1PBE and B3LYP, respectively, and no deviation was observed for CAM-B3LYP compared with the experimental maximum at 453 nm. It was noticed that all TD-DFT computed absorption spectra (Figs. 2–4) have relatively higher intensity at the longer wavelengths (low-energy) and relatively lower intensity at the shorter wavelengths (high-energy). That is opposite to the experimentally observed spectral features. Simulated electronic spectra show blue-shifted maximum in the range of 260–280 nm compared with the peak of 301 nm in the experimental spectrum. Nevertheless, the characteristic spectral feature of experimental spectrum in the 200–230 nm could be identified in each TD-PBE1PBE calculated spectra (Fig. 4). It is interesting to note that although it was found that explicit-implicit solvation model failed to reproduce the absorption in the visible part and UV part lower than 280 nm, it was able reproduce correctly the large and intense experimental band at the 301 nm. In this respect, the valuable information was obtained from the excited state analysis. It was found out that the main electronic transition at about 300 nm using explicitimplicit solvation model corresponded to the water-to-metal charge transfer. It could be assumed that this transition is sensitive to the implicit solvation environment and as a consequence a good compliance with experimentally observed band was obtained. Generally, results obtained in our study show that the choice of the functional used for TD-DFT calculations is much more important than the choice of the method used for the geometry optimization. The PBE1PBE and CAM-B3LYP methods used for geometry optimization gave very similar absorption spectra, while slightly different spectral tendencies using B3LYP were observed (Figs. 2–4). 3.3. Analysis of excited states

Fig. 4. TD-PBE1PBE calculated absorption spectrum of explicitly solvated [Fe(NCS)]2+, with the SO24 ligand beside the NCS ligand, using different DFT methods on geometry optimizations. The inset shows magnified view of experimental band in the visible region.

Transition metal chemistry often involves open-shell systems and excited states. Electronically excited states are of key importance for the photochemistry and for a variety of optical materials applied in the development of new technologies. Hence, the TD-DFT theory emerges as one of the most practical tools that can be used to predict the electronic properties of transition metal complexes. In order to demonstrate the role of different ligand in the complex, the NTO analysis is a convenient method. This method offers the most compact representation of the transition density between the ground state and the excited state of the orbital involved in the transition. Moreover, the nature of transitions that can be classified as ligand-to-metal (LMCT), ligand-to-ligand (LLCT), or metal-to-ligand (MLCT) charge transfer is defined by means of NTO. Here we present the nature of the excited states obtained from TD-PBE1PBE calculation using PBE1PBE level of theory since the latter functionals showed a good overall compliance, compared

45

E. Elijošiute˙ et al. / Polyhedron 90 (2015) 41–46

with the experimental [Fe(NCS)]2+ spectrum. The wavelengths, singlet states, and molecular orbital (MO) interactions of the most important vertical transitions in terms of oscillator strength (greater than 0.01) are listed in the Table 2. The classification in terms of transition character was based on the composition of the occupied and virtual MOs of the dominant excitation (according to the transition coefficients) of the excited state considered for the [Fe(NCS)]2+ complex. Some characteristic CTs (charge transfer) are visually shown in Fig. 5. By analyzing the main electronic transitions summarized in Table 2, it was noticed that the visible part of the spectrum refers to the two LMCT. Obtained transition coefficients for HOMOs (highest occupied molecular orbitals) of the transition 6 indicate an extremely important role of the NCS ligand where isothiocyanate definitively acts as the principal

electron donor group (Fig. 5). However, due to the participation of a small number of occupied MOs, at the same time the contribution of the SO24 ligand based LMCT transition was identified. Unlike the transition 6, the second excitation transition in the visible region (transition 7) involves lower-energy occupied MOs which are mainly centered on the SO24 ligand (Fig. 5). Based on the results of NTO analysis it is clear that the electron donors are either three or four O atoms. Nevertheless, a small contribution from occupied MOs of the NCS ligand was also identified. It can be assumed that both NCS and SO24 ligands are the main ones responsible for the occurrence of absorbance peak in the visible part. Based on our results, all low-lying transitions in the near UV region (200–380 nm) were still characterized as LMCT. It was observed that at the intersection point of visible and near UV

Table 2 Main electronic transition of the benchmark [Fe(NCS)]2+ complex calculated with TD-PBE1PBE//PBE1PBE.

*

Transition

MO interaction*

Transition type

Wavelength (nm)

Oscillator strength (f )

6 7 17 18 27 28 32 34 39 41 56 58 66 68 71

HOMO?LUMO HOMO-2?LUMO HOMO-1?LUMO+4 HOMO?LUMO+4 HOMO-5?LUMO+2 HOMO-5?LUMO+2 HOMO-7?LUMO+1 HOMO-8?LUMO HOMO-5?LUMO+3 HOMO-5?LUMO+3 HOMO-10?LUMO+1 HOMO-1?LUMO+6 HOMO-13?LUMO HOMO?LUMO+8 HOMO-11?LUMO+2

LMCT LMCT LMCT LMCT LMCT LMCT LMCT LMCT LMCT LMCT LMCT LLCT LMCT LLCT LMCT

455 444 372 368 270 268 252 248 237 232 212 210 202 201 200

0.1857 0.0623 0.0215 0.0143 0.0279 0.0168 0.0719 0.0338 0.0144 0.0160 0.0205 0.0268 0.0113 0.0221 0.0238

According to the highest transition coefficient.

Fig. 5. The TD-PBE1PBE/PBE1PBE calculated singlet electron transition (illustrations) for selected transitions.

46

E. Elijošiute˙ et al. / Polyhedron 90 (2015) 41–46

wavelengths, the electron donation from NCS ligand is prevalent (transitions 17 and 18). At the wavelengths lower than 370 nm, the vast majority of occupied MOs are essentially orbitals of the SO24 ligand with the charge transfer delocalized on the metal. It should be emphasized here that depending on the respective excited state, the additional occupied MOs of the H2O and NCS ligands, respectively, were also determined in the transitions. Based on the transition coefficients of pure SO24 ligand character the highest occupied MOs are presented in the transitions 27 (Fig. 5), 28, 39, 41. Concerning particular MO involved in these transitions, electrons are transferred to Fe atom from all four or three O atoms of SO24 ligand. More diverse electronic transitions were noticed in the range of 200–250 nm. This region exhibits two main absorption peaks. The absorption band at the higher wavelengths encompasses two main transitions – 32 and 34. According to transition coefficients, both these excited states have NCS ? Fe, SO4 ? Fe, SO4 and H2O ? Fe LMCT character. The latter transition is visualized in Fig. 5. However, based on the transition coefficients, the tendency that both transitions being much more delocalized over the NCS ligand was observed. Transitions 56, 66 and 71 are typical for LMCT state (Fig. 5 transition 56 c), but the contribution of ligand-to-ligand charge transfer (LLCT) was also fixed. Here we identified the electron transfer from S atom to C atom in the NCS ligand and from the NCS ligand to H2O ligand (Fig. 5 transition 56 a and b). By analyzing the LMCT state of the transitions 56, 66 and 71 it is clear that the excited electron density is mainly delocalized over the H2O ligand. Until nearly far UV region is approached, electronic transitions 58 (Fig. 5) and 68 were assigned as LLCT transitions, in which electrons are mainly promoted from the NCS-based MO, to the empty MOs of the H2O ligand. Generally, the excited state analysis allowed us to determine that both NCS and SO24 ligands are the most active substituents governing spectral properties of the aqueous [Fe(NCS)]2+ complex. 4. Conclusions In the present work theoretical electronic spectra of the aqueous [Fe(NCS)]2+ complex were described by means of DFT calculations. The influence of methods and functionals used on the geometry optimizations and electronic spectra calculations of the [Fe(NCS)]2+ complex, respectively, was investigated. Additionally, the influence of different solvation models and different position of SO24 ligand vs. NCS ligand upon [Fe(NCS)]2+ electronic spectra was evaluated. In order to estimate the nature of electronic transitions, the NTO analysis has been employed. It was observed that the choice of the functional used for TD-DFT calculations is much more important than the choice of the method used for the geometry optimization. From the simulated electronic spectra analysis, it was concluded that the TD-PBEPBE approach applied for optimizations using different methods was in the worst agreement with experimentally observed spectrum. It was found that the TD-PBE1PBE calculation using PBE1PBE level of theory is the most appropriate for the reproduction of the main features of the experimental [Fe(NCS)]2+ spectrum. In this case, it has been shown that the main electronic transitions are characterized as LMCT in the entire spectrum. Moreover, it has been demonstrated that the electron donors are mainly NCS and SO24 ligands in the entire spectrum. Acknowledgement Dr. Juozas Šulskus is acknowledged for the theoretical calculations using CAM-B3LYP method.

References [1] B. Machura, A. S´witlicka, J. Palion, R. Kruszynski, Struct. Chem. 24 (1) (2013) 89. [2] B. Mihaljevic´, B. Katušin-Razˇem, D. Razˇem, Free Radical Biol. Med. 21 (1) (1996) 53. [3] K. Basavaiah, J.M. Swamy, Farmaco Sci. 56 (8) (2001) 579. [4] W. Misiuk, Farmaco Sci. 60 (1) (2005) 61. [5] J. Li, M.S. Safarzadeh, M.S. Moats, J.D. Miller, K.M. LeVier, M. Dietrich, R.Y. Wan, Hydrometallurgy 113–114 (2012) 31. [6] J. Li, M.S. Safarzadeh, M.S. Moats, J.D. Miller, K.M. LeVier, M. Dietrich, R.Y. Wan, Hydrometallurgy 113–114 (2012) 1. [7] J.F. van Staden, C. Saling, D. Malan, R.E. Taljaard, Anal. Chim. Acta 350 (1–2) (1997) 37. [8] S.M. Sultan, E. Bishop, Analyst 107 (1982) 1060. [9] M.J. Capitan, E. Muñoz, M.M. Graciani, R. Jiménez, I. Tejera, F. Sánchez, J. Chem. Soc., Faraday Trans. 85 (1989) 4193. [10] I. Korondán, I. Nagypál, Inorg. Chim. Acta 73 (1983) 131. [11] K. Ozutsumi, E. Hasegawa, M. Kurihara, T. Kawashima, Polyhedron 12 (1993) 2185. [12] C.R. Bhattacharjee, P. Goswami, P. Mondal, J. Coord. Chem. 63 (2010) 2002. [13] V. Ramesh, A. Shihabudeen Syed, K. Jagannathan, K. Rajarajan, Spectrochim. Acta Part A Mol. Biomol. Spectrosc. 108 (2013) 236. [14] A.G. Ivšic´, B. Tamhina, Croat. Chem. Acta 76 (2003) 323. [15] W.J. Kirsten, I. Lindholm-Franzén, Microchem. J. 25 (1980) 240. [16] X.-G. Li, Q. Dou, M.-R. Huang, Monatsh. Chem. 139 (2008) 1157. [17] J.S. Gancheff, A. Acosta, D. Armentano, G. De Munno, R. Chiozzone, R. Gonzalez, J. Inorg. Chim. Acta 387 (2012) 314. [18] Q.-M. Wang, G.-C. Guo, T.C.W. Mak, Polyhedron 20 (2001) 2683. [19] O. Kohle, S. Ruile, M. Grätzel, Inorg. Chem. 35 (1996) 4779. [20] E. Shurdha, S.H. Lapidus, P.W. Stephens, C.E. Moore, A.L. Rheingold, J.S. Miller, Inorg. Chem. 51 (2012) 9655. [21] I.I. Zakharov, K.Yu. Kudjukov, V.V. Bondar, N.F. Tyupalo, B.F. Minaev, Comp. Theor. Chem. 964 (2011) 94.  naite˙, G. Denafas, [22] E. Elijošiute˙, O. Eicher-Lorka, E. Griškonis, Z. Kuodis, D. Janku Spectrochim. Acta, A: Mol. Biomol. Spectrosc. 129 (2014) 43. [23] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukada, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Know, H.P. Hratchian, J.B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J. Dannenburg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, A. Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A. Pople, GAUSSIAN 03, Revision D.01, Gaussian, Inc., Wallingford, CT, 2004. [24] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö. Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, GAUSSIAN 09, Revision A.02, Gaussian, Inc., Wallingford CT, 2009. [25] A.D. Becke, J. Chem. Phys. 98 (7) (1993) 5648. [26] T. Yanai, D. Tew, N. Handy, Chem. Phys. Lett. 393 (2004) 51–57. 1–3. [27] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 78 (1997) 1396. [28] J.P. Perdew, K. Burke, M. Ernzerhof, Chem. Phys. Lett. 77 (1996) 3865. [29] R.L. Martin, J. Chem. Phys. 118 (11) (2003) 4775. [30] www.chemcraftprog.com. [31] www.thermoscientific.com. [32] W.R. Scheidt, Y.J. Lee, D.K. Geiger, K. Taylor, K. Hatano, J. Am. Chem. Soc. 104 (12) (1982) 3367. [33] C.-H. Hsieh, S.M. Brothers, J.H. Reibenspies, M.B. Hall, C.V. Popescu, M.Y. Darensbourg, Inorg. Chem. 52 (4) (2013) 2119. [34] D. Armentano, G. de Munno, J. Faus, F. Lloret, M. Julve, Inorg. Chem. 40 (4) (2001) 655. [35] A. Stefánsson, K.H. Lemke, T.M. Seward, In: Proceedings of ICPWS XV, 7 p. www.15icpws.de/papers/09_Geo-05_stefansson.pdf. [36] J.Y. Zhang, L. Wang, Y. Liu, Acta Crystallogr., Sect. E: Struct. Rep. Online E67 (2011) m1568, http://dx.doi.org/10.1107/S1600536811042723. [37] A. Horn Jr., I. Vencato, A.J. Bortoluzzi, V. Drago, M.A. Noyak, A. Neves, J. Braz. Chem. Soc. 17 (8) (2006) 1584.