Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 97 (2012) 1041–1051

Contents lists available at SciVerse ScienceDirect

Spectrochimica Acta Part A: Molecular andBiomolecular Spectroscopy

journal homepage: www.elsevier .com/locate /saa

Molecular structure, natural bond analysis, vibrational, and electronicspectra of aspartateguanidoacetatenickel(II), [Ni(Asp)(GAA)]�H2O: DFT quantummechanical calculations

J.M. Ramos b,c, M.T. de M. Cruz b, A.C. Costa Jr. b,d, G.F. Ondar b,d, Glaucio B. Ferreira b,L. Raniero a, A.A. Martin a, O. Versiane d, C.A. Téllez Soto a,⇑a Laboratory of Biomedical Vibrational Spectroscopy, IP&D, Research and Development Institute-UNIVAP, Av. Shishima Hifumi, 2911 Urbanova,12.224-000 São José dos Campos, SP, Brazilb IQ-UFF, Departamento de Química Inorgânica, Morro de Valonguinho s/n. – Centro, 24210-150 Niterói, RJ, Brazilc IQ-UFRJ, Departamento de Química Inorgânica, Avenida Athos da Silveira Ramos, 149 Bloco A, 6� andar, Cidade Universitária, 21941-909 Rio de Janeiro, RJ, Brazild Instituto Federal de Educação, Ciência e Tecnologia do Rio de Janeiro (IFRJ), Unidade de Rio de Janeiro, Rio de Janeiro, RJ, Brazil

h i g h l i g h t s

" Synthesis ofaspartateguanidoacetatenickel (II)was carried out.

" The FT-IR and the FT-Raman spectraof [Ni(Asp)(GAA)] were recorded andbands were assigned.

" SERS effect is informed." UV–Vis spectrum was assigned and

discussed." The Bond Orbital analysis was

performed thought the DFT method.

1386-1425/$ - see front matter � 2012 Elsevier B.V. Ahttp://dx.doi.org/10.1016/j.saa.2012.07.087

⇑ Corresponding author.E-mail address: tellez@vm.uff.br (C.A. Téllez Soto)

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o

Article history:Received 15 February 2012Received in revised form 24 June 2012Accepted 5 July 2012Available online 3 August 2012

Keywords:DFT:B3LYP/6-311G(d, p)FT-infraredFT-Raman spectraAspartateguanidoacetatenickel(II) complexUV–Vis spectrumSERS

a b s t r a c t

The aspartateguanidoacetatenickel (II) complex, [Ni(Asp)(GAA)], was synthesized and structural analysiswas performed by means of the experimental methods: determination of the C, H, N and O contents, ther-mogravimetry, infrared and Raman spectroscopy. DFT:B3LYP/6-311G(d, p) calculations have been per-formed giving optimized structure and harmonic vibrational wavenumbers. Second derivative of theFT-infrared, FT-Raman and Surface Raman Enhanced Scattering (SERS) spectra, and band deconvolutionanalysis were also performed. Features of the FT-infrared, FT-Raman and SERS confirmed theoreticalstructure prediction. Full assignment of the vibrational spectrum was also supported by a carefully anal-ysis of the distorted geometries generated by the normal modes. The Natural Bond Orbital analysis (NBO)was also carried out as a way to study the Ni (II) hybridization leading to the pseudo planar geometry ofthe framework, and the extension of the atomic N and O hybrid orbital of the different amino acids in thebond formation. Bands of charge transfer and d–d transitions were assigned in the UV–Vis spectrum.

� 2012 Elsevier B.V. All rights reserved.

Introduction systems. Aspartic acid H2NCH2CH(COOH)2 and guanidoacetic acid

The study of metal complexes with biological ligands is impor-tant for a better understanding of the processes that occur in living

ll rights reserved.

.

H2N(NH)2CH2COOH act as bidentate ligands and form complexeswith transition metals [1]. In the complex formation, the –NH2

group of the aspartic acid and the O-bond through the deproto-nated carboxylate groups act as a Lewis base. The guanidoaceticacid acts as a ligand through the –NH group and through thedeprotonated carboxylic acid as an electron donor in the bond for-mation with Ni(II).

1042 J.M. Ramos et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 97 (2012) 1041–1051

In a series of papers [2–4], synthetic and structural aspects of anumber of transition metal complexes of N, O and N, S donor sys-tems have been described. Presently, there is a growing interest inthe coordination chemistry of structurally modified bio-ligands, sotransition metal complexes are the objective of extensive investi-gations [5–8].

Previous works of our group, based in experimental and theo-retical methods from the vibrational spectroscopy point of viewfor complexes [Ni(Ser)2] [9], [Ni(GAA)(Ser)] [10], [Ni(GAA)(Gly)][11] and [Ni(GAA)2] [12] have been reported. Continuing our spec-troscopic research on metal–amino acid complexes, spectral fullvibrational assignments were performed for [Ni(Asp)(OH)(H2O)][13], [Cd(Cys)Cl2]� [14] and [Cd(Cys)(Gly)] [15] based on the FT-infrared spectra, second derivative and band deconvolution analy-sis, and on a DFT study using the hybrid functional B3LYP with 6-311G(d, p) basis set. In the present work the focus was centered onthe geometry of the aspartateguanidoacetatenickel (II) complexand in the vibrational assignment of the infrared, Raman and SERSspectra. Also, the Natural Bond Orbital analysis (NBO) was realized,as a way to study the Ni (II) hybridization leading to the pseudoplanar geometry of the framework, and the extension of the atomicN and O hybrid orbitals of the different amino acids in the bondformation. UV–Vis spectrum also predicts a diamagnetic or spin-paired square planar (dsp2) complex. Due to the bands in the FT-Raman with weak intensity and the sample to present fluorescencein the higher spectral region, measurements were carried out mix-ing the sample with silver nanoparticles (AgNP), in two differentRaman spectrometers resulting in an enhancement of several nor-mal modes.

Experimental

Synthesis of the [Ni(Asp)(GAA)]�H2O complex

Reagents and solvents were used as received without furtherpurification: nickel (II) nitrate hexahidrate P.A., nitric acid 65%P.A., potassium hydroxide P.A., ethanol P.A., methanol P.A., acetoneP.A. and diethyl ether P.A. from VETEC – Química Fina. L-Asparticacid sodium salt monohydrate (Sodium L-aspartate) and guanido-acetic acid from Sigma–Aldrich Co.

In this section the following nomenclature will be used: Aspmeans aspartate; GAA means guanidoacetate, HAps2� meanshydrogen aspartate anion, H3GAA means guanidoacetic acidand GAA3� means guanidoacetate anion. The ternary system[Ni(Asp)(GAA)]�H2O is strongly limited by the low solubility ofthe H3GAA acid. The GAA3� and HAsp2� exist in relevant concen-trations above pH3O+ = 3.0. The best conditions for synthesis aregiving between pH3O+ = 3.0–3.7 to avoid H3GAA precipitation andcrystal formation. A species distribution diagram depicted inFig. 1S of the Supplementary material gives essential informationon the synthesis under controlled pH3O+.

After dissolution of 2 mmol of aspartic acid (0.2662 g) and2 mmol of guanidoacetic acid (0.2342 g), the solution of guanido-acetic acid was slowly added to the aspartic acid solution understrong stirring at 43 �C for three hours. The solution was freezedat 4 �C for 1 h and then kept in a sealed glass tube at room temper-ature for 2 days. Any formed solid was eliminated. When a clearand dense solution was obtained, 2 mmol of Ni(NO3)2 (0.5816 g)were added. The solution was then stirred and heated at 45 �Cfor 3 h. After another 5 h elapsed, a solution of KOH 5 mol/L wasslowly added until a pH3O+ = 3.4 was obtained. The glass tubewas then sealed and stored for 2 days; at this stage the volumeof the solution had been reduced by 50% so the volume was thenfilled to 100% by adding absolute ethanol. The product was washedwith solutions: (a) 60 mL ethanol + 30 mL methanol + 10 mL

acetone and (b) 60 mL ethanol + 5 mL HNO3 + 20 mL diethyl ether.The product was then dried in a sealed vessel with CaCl2. Yieldaround 54%.

Characterization

Elemental analysis

All the analysis was carried out through a CHNS-O EA 1110model analyzer of CE Instruments, in the same conditions and withthe same CHN column. After the stabilization stage, it was ob-served that any change of column in the analysis of differentsamples produced small variations, of approximately 2%. After3 weeks, the samples were analyzed again and the final results re-flected the arithmetic mean of the results with fluctuations lowerthan 2%. For the [Ni(Asp)(GAA)]�H2O complex the following valueswere found (theoretical composition between parenthesis): C:25.82% (25.87%); H: 4.39% (4.30%); N: 17.10% (17.25%); O: 34.67%(34.50%). Ni was analyzed by atomic absorption spectrometryyielding the experimental value of 18.00% (18.08%).

Thermo-gravimetric analysis

The thermo-gravimetric analysis was performed in an inertN2(g) atmosphere. The results of the thermo-gravimetric analysisin the temperature range of 100–315 �C showed a mass loss of40.8%, which corresponds to an experimental value of 133 g(132.4 g calculated). This loss of mass corresponds to the frag-ments: C3H4O2, CH3N2, and H2O of the outer coordination shell.At temperatures above 315 �C, the mass loss was 59.2% corre-sponding to the experimental value of 191.6 g (192.3 g calculated).This mass loss is attributable to CO, NH2, C2H3NO2 and NiO. Fig. 2Sof the Supplementary material shows the experimental TGA curve.

FT-IR and FT-Raman spectra

The FT-IR spectra of solid nickel (II) complex were recorded on aPerkin Elmer 2000 FT-IR spectrometer. Data were collected with aresolution of 4 cm�1. The scanning speed was held at 0.2 cm�1s�1

and 120 scans were performed. The solid sample was measuredas a KBr pellet in the 4000–370 cm�1 spectral range, and in the700–30 cm�1 region as a polyethylene pellet. Experimental infra-red spectra for both regions are shown in Fig. 1, and a comparisonbetween experimental and calculated spectra is presented inFig. 3S. The FT-Raman spectra were measured at room temperatureusing a Bruker Spectrometer (model RFS 100/S), in the region of theO–H, N–H and C–H stretching with a resolution of 4 cm�1 and 120scans. The samples were measured in a solid state. The excitationlight source was a Nd:YAG laser at 1064 nm. The laser spot sizewas 200 lm in diameter and the power was kept at 30 mW to pre-vent any degradation, but the sample had presented higher fluo-rescence. To avoid this difficulty between 1800 and 400 cm�1,the Raman spectrum was recorded with a Confocal Raman fromRivers Diagnostics (Model 3510) coupling to a laser at 785 nmaccording to the instrument specifications. Mixing the sample withsilver nanoparticles (AgNP) new measurement were carried outwith both Raman spectrometers, maintaining the measurementsconditions and the spectra had showing SERS effect. Comparisonsbetween the experimental Raman spectra obtaining with andwithout AgNP are shown in Figs. 2 and 3. The experimental spectracorrespond to the anharmonic spectrum. The calculated harmonicspectrum does not consider the natural perturbations that takeplace in the solid state of the sample.

Fig. 1. FT-infrared spectra of [Ni(Asp)(GAA)]�H2O: (a) in the region of 4000 to 370 cm�1 and, (b) in the region of 700 to 50 cm�1.

Fig. 2. Raman spectra obtained with the 1064 nm excitation line laser with andwithout AgNP.

Fig. 3. Raman spectra obtained with the 785 nm excitation laser with and withoutAgNP.

J.M. Ramos et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 97 (2012) 1041–1051 1043

UV–Vis spectrum

The UV–Vis spectrum of [Ni(Asp)(GAA)] was carried out for thesolid sample due to his low solubility and decomposition in solu-tion. The solid state UV–Vis spectrum was acquired between 200and 1000 nm, using a Varian Cary 5000 spectrometer. The UV–Vis spectrum is shown in Fig. 4. The solid state spectrum presentsan appreciable enhancement of intensity in the region between450 and 800 nm. Fig. 5(a and b) shows the deconvolution bandanalysis (DBA) of the UV–Vis spectra in the solid state.

A more detailed analysis of spectroscopic information of thecomplex was carried out through a theoretical and experimentalassignment, considering structural parameters calculated by theproposal obtained by geometry optimization.

Calculations

The calculations were carried out for the neutral complex,[Ni(Asp)(GAA)], considered it as no interacting independent units.For geometry optimization, the density functional theory method(B3LYP) was used in the Gaussian 03 program [33]. For all calcula-tions, we used the 6-311G (d, p) basis set in carbon, sulfur, nitro-gen, hydrogen and nickel atoms. All calculations have beenoptimized from several initial geometries, in order to guarantee

Fig. 4. UV–Vis spectra in the solid state of [Ni(Asp)(GAA)]�H2O.

Fig. 5. Band deconvolution analysis of the UV–Vis spectra in the solid state of [Ni(Asp)(GAA)]�H2O.

1044 J.M. Ramos et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 97 (2012) 1041–1051

the global minima energy structures. After this procedure, thevibrational calculations were performed. No imaginary mode wasobserved. Characteristic normal stretching and bending modesfrom the –CH2 groups were visualized using the graphical Chem-craft program [27]. The skeletal or framework normal modes weredeterminate using the percentage deviation of the geometricalparameters (PDPG), from its equilibrium position [9–15].

Results and discussions

Structural DFT calculations

Ab-initio optimization of the geometric parameters, and hence astructural analysis for the aspartateguanidoacetatenickel (II) com-plex, [Ni(Asp)(GAA)]�H2O, has been done using density functionaltheory methods [16], employing the gradient-corrected hybriddensity functional B3LYP procedure. The 6-311G(d, p) standard ba-sis set [17–20] was selected for all atoms. The calculated bondlengths for the Ni–N bond with the aspartate ligand was 1.921 Å,this value can be compared with the experimental Ni–N bondlength of 2.047 Å found in the [Ni(L-aspO)(H2O)2]�H2O complex[21], and with the calculated values for the aspartatehydroxo-aquaNi(II) complex, [Ni(Asp)(OH)(H2O)], using the DFT procedure withB3LYP/6-31G and B3LYP-6-311G levels which are: 1.824 and

Table 1DFT:B3LYP/6-311G(d, p) bond length and inter-bond angles in the

Bond lengths (Å), Atoms i, j Bond angles (�),

1–2 1822; 1.824 2–1–3 174.341–3 1856; 1.861 2–1–9 87.371–9 1.929; 1.978 2–1–17 85.401–17 2.047; 1.892; 1.921 1–2–4 116.53; 12–4 1.342; 1.344 3–1–9 92.333–29 1.260; 1.328; 1.332 3–1–17 94.834–5 1.234 1–3–29 133.284–6 1536; 1.517 9–1–17 172.766–9 1.498; 1.510 1–9–6 106.22; 111–12 1.280; 1.281 1–9–11 107.04;17–20 1.475; 1.484; 1.484 1–17–20 113.6720–26 1.536; 1.545; 1.542 2–4–5 124.3521–22 1.223 2–4–6 112.21; 123–24 0.975 3–29–30 117.426–29 1.512; 1.537; 1.529 5–4–6 122.02; 129–30 1.265; 1.241; 1.241 4–6–9 109.49; 19–11 1.491; 1.48511–14 1.368; 1.367

In italic characters calculated DFT/B3LYP:6-311G(d, p) bondp)(OH)(H2O)] complex according to Refs. [10,12]. In Negrito boldcomplex according to Ref. [20].

1.892 Å, respectively [13]. The Ni–N bond with the guanidoacetateligand was 1.978 Å; this value can be compared with the calculatedNi–N bond of 1.966 Å in the bis-guanidoacetatenickel (II) complex[12]. The Ni–O bond length with the aspartate ligand was 1.863 Å,and with the guanidoacetate ligand was 1.824 Å; these values areclose to the DFT/B3LYP-6-31G Ni–O calculated bond lengths of1.852 and 1.852 Å for the [Ni(Gly)(GAA)] complex [11]. Also theDFT/B3LYP-6-31G calculated Ni–O bond lengths for the [Ni(Ser)2]complex are 1.844 and 1.853 Å. For the [Ni(Asp)(OH)(H2O)] com-plex the DFT values for the Ni–O bonds are: 1.751; 1.818 and1.854; 1.906 Å, with the same basis as indicated above. Experimen-tal values for Ni–O bonds for [Ni(L-AspO)(H2O)2]�H2O were: 2.063and 2.064 Å. Concerning the calculated bond angles, the mean va-lue of the ratio between the NNiN (172.756�) and ONiO (174.342�)divided by two gives the factor b = 86.77� [22], indicating a devia-tion of 3.22� of the planar framework structure. Selected bondlengths and bond angles are given in Table 1. Full [Ni(Asp)(GAA)](without the hydration water) complex structure is shown in Fig. 6.

Natural Bond Orbital analysis

NBO analysis [23–25] at the B3LPY/6-311G(d, p) level was car-ried out to rationalize the factors contributing to the total confor-mational energy. The NBO analysis for the isolated complex shows

[Ni(Asp)(GAA)] complex⁄.

Atoms i, j, k Bond angles (�), Atoms i, j, k

9–11–12 126.33;117.6212–11–14 123.32;130.9111–12–13 115.521

18.03 17–20–26 110.38520–21–22 123.24720–26–29 115.32922–21–23 120.38926–29–30 119.748

08.22 9–11–14 110.43; 111.49114.07 6–9-11 115.72; 113.71;115.20 2–4–5 125.71; 124.36

3–29–25 116.10; 117.4613.69 25–20–17 110.57; 107.25

622.7611.36

lengths and bond angles for [Ni(GAA)(Ser)] and [Ni(As-characters: equivalent bonds in the [Ni(L-aspO)(H2O)2]�H2O

Fig. 6. DFT/B3LYP:6-311G(d, p) calculated structure for the [Ni(Asp)(GAA)] com-plex. Numbering atoms: Ni(1); O(2, 3, 5, 22, 23, 30); N(9, 12, 14 17); C(4, 6, 11,20,21, 26,29); H(7, 8, 10, 13, 15, 16, 18,19, 24, 25, 27, 28).

J.M. Ramos et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 97 (2012) 1041–1051 1045

that the nickel (II) ion interacts more strongly with the oxygenthan with the nitrogen atom. The second-order perturbation ener-gies for the stronger interaction involved in the formation of eachNi–O bond is about 76.0 kcal mol�1. This energy is 6.7 kcal mol�1

larger than the highest energy obtained for all the interactions be-tween Ni and nitrogen (69.30 kcal mol�1). However, unlike the twoNi–O interactions (Ni1�O3 and Ni1–O2), that are almost isoener-getic; the interaction between Ni and the nitrogen atoms differsin energy. The interaction energy between the nickel and the nitro-gen bounded to the guanidine group (N9) is about 10.00 kcal mol�1

higher than the corresponding for the nitrogen close to the car-boxyl group (N17) (69.3 kcal mol�1 against 59.2 kcal mol�1). Ananalysis of the electronic delocalization effect in the stability ofthe complex, due the presence of groups containing p-electrons(C@O and C@N), can be performed by removing the Fock matrixelements corresponding to the second-order charge transfer of

Fig. 7. Natural Bond Orbital analysis; (a) natural bond orbital between nickel and oxygebetween a p⁄ C@N anti-bond orbital and a sp4.63 orbital on the nitrogen from the Ni–N9anti-bond orbital and sp4.63 orbital on N9.

the specific interactions. As a result, we found that the total elec-tronic transfer involving the pC@O anti-bond on the C5@O4 andC29@O30 carbonyls contributed each one with 52.0 kcal mol�1 tothe complex stability. This energy is 15.5 kcal mol�1 larger thanthe total stabilization energy calculated for all interactions involv-ing the pC@N anti-bond orbital present in the guanidine group(C11@N12, 36.5 kcal mol�1). Nevertheless, the pC@O anti-bond ofthe carboxyl group contributes with only 12.0 kcal mol�1 to thecomplex stability, 40.0 kcal mol�1 lesser than the pC@O anti-bondof C5@O4 and C29@O30 carbonyl bonded to the O3 and O2 directlycoordinated to the nickel (II) ion. The NBO calculation shows anelectronic transfer from N9 via sp4.63(17.76% s and 82.24% p) hy-brid orbital to the pC@N anti-bond orbital in the guanidine groupinvolving energy of 12.63 kcal mol�1, in Fig. 7c and d. This reso-nance should be the principal cause of the higher interaction en-ergy of the N9 with the nickel ion relative to the N17 atom withthe nickel ion. In addition, the carbonyl C21@O22, is distant toeffectively interact with the nitrogen from the Ni�N17 bond tocompensate for this difference of energy. The NBO results showalso that the Ni1–O2 bond is formed by the interaction betweena sp2.61d1.29 (20.41% s, 53.25% p and 26.33% d) orbital on the nickelion and a sp5.04 (16.55% s and 83.45% p) orbital on the oxygen atom(Fig. 7a). Indeed, the Ni1–O3 bond is formed by a similar hybridorbital on the nickel (sp2.12d1.02, 24.15% s, 51.25% p and 24.60%d), in Fig. 7a. However, the orbital on oxygen O3 is hybridized witha lesser p character (74.15% in O3 against 83.45% in O2) and conse-quently with a superior s character (25.85% in O3 against 16.55% inO2).

For the Ni–N bond calculations, the NBO results show thatthe Ni1–N9 bond is built by the interaction between asp1.7d0.87(28.01% s, 47.66% p and 24.33% d) hybrid orbital on thenickel and a sp3.76 (21.01% s and 78.99% p) one on the N9, whilefor the Ni1–N17 bond, a sp2.83d1.48(18.84% s, 53.30% p and27.86% d) orbital on the nickel ion interacts with a sp3.75(21.06%s and 78.94% p) orbital on the N17 atom to form the bonding, inFig. 7b. According to these results, a small variation is observed

n atom and (b) natural bond orbital between nickel and nitrogen atom; (c) overlapbond and (d) other view of (c) showing the resonance effect between the p⁄ Ni@N9

1046 J.M. Ramos et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 97 (2012) 1041–1051

on the hybridization of the orbital for the nickel ion in the bondingformation when the two Ni1–N9 and Ni1–N17 bonds are compared(sp1.7d0.87 versus sp2.83d1.48, respectively), but the participation oforbital s and p on both nitrogen atoms are very similar (sp3.76

and sp3.75).The participation coefficient in the formation of the Ni–O2 bond

is 12.50% on nickel ion and 87.50% on oxygen atom, and for the Ni–O3 bond, 10.42% on nickel ion and 89.58% on oxygen atom. In theformation of both Ni1–N9 and Ni1–N17 bonds these coefficientsare very close (11.12% and 11.44% on Ni and 88.88% and 88.56%on nitrogen, respectively). These results show a high polarizationfor the Ni–O and Ni–N bonds, indicating a strong ionic characterfor both bonds.

Electronic UV–Vis spectra

Calculations of the transition energies and the oscillatorstrengths in the UV–Vis spectra of the optimized structure wereaccomplished using the TD method, using calculated orbitals withB3LYP. Results of main singlet transition energies (eV) and theoscillator strength from the ground state of the [Ni(Asp)(GAA)]complex, with the B3LYP method considering 70 singlet statesare presented in Table 2. Koopmans energy (eV), Mullikenpopulation analysis and assignment for the frontier orbitals of[Ni(Asp)(GAA)], are presented in Table 3. The Koopmans’ energydifference between the highest occupied and the lowest unoccu-pied orbital, HOMO–LUMO gap was 4.54 eV. The HOMO orbitalwas characterized by 3d metal (50%). The LUMO orbital was char-acterized by 3d metal (65%) with the participation of the sulfur(46%). The final analysis of the molecular orbital was obtainedthrough the density of states. These results are shown inTable 1S and Fig. 4S in Supplementary material, where the degreeof positive/negative overlap for frontier orbitals were detached inthe OPDOS diagram. The frontier orbitals are populated by d metalorbitals. This analysis was important, especially to evaluate thenature of the electronic valence transitions LMCT or MLCT.

The analysis of the states and the simulation of the UV–Visspectrum were accomplished using the Chemcraft [27] and Gauss-Sum [30] programs, and the obtained results are presented inFig. 5S, of the Supplementary material.

The literature no presents information about the charge trans-ference spectrum for the bands in the [Ni(Asp)(GAA)] complex, be-cause it is a non-reported new compound. In [Ni(Asp)(GAA)], thecalculated framework structure was square planar with Ni(II)bound to two oxygen and to two nitrogen atoms. The calculatedangles N–Ni–N and O–Ni–O were of 172.67� and 174.15�, respec-tively, giving a C2v symmetry for the NiN2O2 framework. Ni(II) isa d8 system with his eight electrons paired in the four lower energyd orbital’s, and has the possibility to achieve a low spinconfiguration.

The data obtained in the deconvolution of experimental spectrain Fig. 5(a and b) were used in comparison with theoretical data.Thus, the following discussion will be carried: the observed low-intensity bands at 706, 634 and 573 nm in the solid state werecompared with the theoretical data in 695, 616 and 523 nm forB3LYP. By comparison with Ni(II) square planar complexes of D2h

symmetry [31], correlations between complexes with D2h and C2v

point groups can be done, so that we consider assign these bandsas: 706/695 nm to m1 (1A1 ? 1B1, C2v); (1A1g ?

1B2g, D2h); 634/616 nm to m2(1A1 ?

1A2, C2v); (1A1g ?1B1g, D2h); 573/523 nm to

m3 (1A1 ? 1B2, C2v); (1A1g ?1B3g, D2h). Atanassov and Nikolov [32]

in his study on electronic spectra of planar chelate nickel(II) com-plex is of the opinion that the effect of the chelate angle a is morestrongly expressed for the two low energy d–d transitions, and thes–d mixing The s–d mixing affects the dz2 and dx2-y2 and conse-quently the transition energies 1A1g ?

1B1g (dz2, dxy) and

1A1g ?1B1g (dx2 ? dy2, dxy), and gives in a graphical representa-

tion the variation of the transition energies DE (in kK) as a functionof the bite angle, a, which in our case is of 86.77(�).

The bands observed at 486, 389, 278, 231 and 217 nm fromdeconvoluted UV–Vis spectrum in the solid state were comparedwith the theoretical data in 513, 313 289, (241, 237, 237) and207 nm for B3LYP. These states of oscillator strengths between0.01 and 0.08 a.u. were classified as LMCT transitions for bands be-tween 389 and 231 nm. Also, the band in 207 nm was classified asMLCT. The main assignments are summarized in Table 4. Fig. 8shows the orbital representation of the HOMO and LUMO, atB3LYP method of [Ni(Asp)(GAA)]. The contour values of the orbitalsare all 0.035 a.u.

Vibrational spectra

The 3n�6 = 84 normal modes of the aspartateguanidoacetate-nickel (II) complex can be theoretical described by 33 stretching,60 bond angle bending and 21 torsion internal coordinates includ-ing redundant coordinates.

The infrared spectrum in the region of 4000 to 370 cm�1 can besubdivided in four main spectral regions. The first one between4000 to 2500 cm�1 is a strong overlapped region where absorp-tions are observed for one O–H, four N–H from the –RH2 groups,two N–H from the imine groups, four C–H stretching from the –CH2 groups and two C–H stretching from the methine groups,totalizing 12 absorption bands. The second region from 1800 to1500 cm�1 shows a large band containing the –C@O and –C@Nstretching’s and the HNH and HCH scissoring bending absorptions.The third region is characterized by the presence of large bands be-tween 1500 to 1000 cm�1 composed of HNH and HCH bending andring deformations. The fourth region, known as the fingerprint re-gion, has absorptions of the C–C, C–N, C–O stretching and deforma-tion bands as well as the rocking vibrations of the –CH2, –CH, –NH2

and –NH (imine) groups. Skeletal or framework vibrations absorbin the low energy region between 700 and 50 cm�1.

O–H, N–H and C–H stretchingThe wide and large infrared band centered at 3553 cm�1 could

be assigned to the m(OH) stretching vibrational mode. The DFTassignment situates this band in a level lower in energy than theN–H stretching of the –NH2 group. The pair of N–H stretching thatwas observed at: 3384/3343 cm�1 and at 3045/3011 cm�1 followsthe Bellamy–Williams wavenumber relation ms(a0) = 345.53 +0.876m as(a00) [26] and could be assigned to the mas(NH)/ms(NH) nor-mal modes of the –NH2 groups. The other bands found at 3281,3176 cm�1 can be assigned to the m(NH) stretching of the iminegroups present in the complex structure. Concerning the C–Hstretching, five infrared absorption bands were expected, and thesewere found at: 2997 (by deconvolution band analysis, hereafternamed as DBA), 2975 (2959), 2921 (2923), 2875 (2869) and at2855 (2850) cm�1 where the bands observed by DBA are given inparenthesis.

C@N stretchingThe infrared band and the SERS band found at 1654 cm�1 corre-

lates with the DFT calculated value of 1667 cm�1 and could be as-signed to a coupled vibrational mode described mainly by them(C@N) and d(HNH) internal coordinates.

C@O stretching and H–N–H bendingIn the vibrational spectrum of [Ni(Asp)(GAA)] complex it was

expected to observe three d(HNH) and two m(C@O). Five wavenum-bers can be assigned to the following vibrational modes: 1666(IR) cm�1, d(HNH) sciss + m(C@O), indicating the coupling of theH–N–H bending and C@O stretching internal coordinates in the

Table 2Experimental FT-IR, FT-Raman and SERS spectra, and calculated DFT:B3LYP/6-311G(d, p) vibrational wave numbers (wn) for the [Ni(Asp)(GAA)] complex.

DFT:B3LYP/6-311Gwn

DFT � 0.9613 IRintensities

FT-IR wn(%T)

2d derivative//BDAa

FT-Raman (Bruker)wn

Raman confocalwn

Raman/SERSb 2d/BDA Approximate assignment

3672 3350 48.04 3384 (18) 3388/3386 3394 (0.10) mas(NH)3660 3518 44.10 3553 (11) –/3580 m(OH)3488 3353 33.03 3343 (18) 3340/3336 3335 (0.3) mas(NH)3484 3349 8.53 3281 (17) 3282/3271 3258 (0.4) 3247 (0.10) m(NH)imine3459 3325 13.91 3176 (14) 3179/3174 m(NH)imine3421 3289 25.63 3045 3068/3067 3106 (0.3) 3090 (0.20) ms(NH)3368 3238 434,89 3011 3021/– ms(NH)3119 2998 15.83 2997/2979 2999 (0.4) mas(CH)3103 2983 9.85 2975 (16) 2967/2959 2968 (0.3) mas(CH)3064 2945 31.53 2921 2924/2923 2936 (0.3) 2928 (0.50) ms(CH)3052 2934 9.15 2875 (3) 2875/2869 ms(CH) + m(CH)methine3020 2903 2.76 2855 2850/– 2847(0.7) 2845 (0.10) ms(CH)1752 1684 245.96 1666 (23) 1672/1671 d(HNH) + m(C@O) + d(OCC)1734 1667 237.52 1654 1655/– 1654 (0.20) 1653/

1655m(C@N) + d(HNH)

1732 1665 13.01 1633 1638/1639 1640 (0.19) d(HNH) sciss.1723 1656 170.96 –/1626 d(HNH)sciss.1695 1629 483.19 1624 (31) –/1612 –/1619 m(C@O) + d(OCC)1651 1587 373.6 1582 1578/1585 1577 (0.24) 1570/

1567m(C@O) + d(HNH)

1511 1452 14.82 1545 (33) 1544/1552 1544 (0.19) 1545/1545

d(HCH)sciss

1508 1450 10.56 1509 1509/– 1510 (0.06), 1502(0.78)

1502/1501

d(HCH)sciss

1462 1405 55.47 1493 (17) 1491/– 1455/1450

d(CNH) + d(HCH)twist

1428 1373 2.76 1472 1436/1438 1430 (0.24) 1426 (2.00) 1427/1426

d(NCH) + d(HNH)wagg

1397 1343 68.97 1412 (41) 1414/1412 1421(0.20), 1417/1412

d(HCH)wagg + d(HNH)

1376 1323 1.82 1384 (47) 1384/1384 1382 (0.06) d(HCH)wagg + d(CCH)1367 1314 20.83 –/1374 1376 (0.05) d(HCH)wagg + d(HNH)1343 1291 31.76 1332 (27) 1330/1332 1343 (0.05) 1348 (0.02) 1351/

1349d(HCH)twsit + d(CCH) + d(HNH)twist

1319 1268 27.23 1308 (23) 1306/1306 1302 (0.03) 1304 (0.06), 1310(0.28)

1318/1311

d(HCH)wagg + d (CCC) + d(HNH)twist

1300 1250 30.09 1263/1278 1273 (0.06) 1271 (0.06) 1271/1271

d(CNH) + d(HCH)twsit

1293 1243 188.71 1236 –/1226 1243 (0.06) 1239 (0.04), 1246(0.10)

1247/1232

m(CO) + d(CCO) + d(CCH) + d(HCH)wagg

1272 1223 260.36 1226 (9) 1218/1217 1220 (0.10) 1222/1220

d(HNH)wagg + d(OCC) + m(CC)

1245 1197 320.73 1181 (7) 1179/1182 1179 (0.10) 1165 (0.04), 1183(0.14)

1184/1188

d(HNH)wagg

1210 1163 521.46 –/1136 1147 (0.40) 1145/1148

m(CO) + d(COH)

1185 1139 5.68 1124/1125 1128 vw d(HCH)twsit + d (HNH)twsit1177 1131 162.95 –/1112 q(NH) + d(CNH) + d(HCH)twist1151 1106 18.04 –/1100 1103 vw d(COH)1147 1103 23.57 1094 (10) 1095/1092 1099 (0.07), 1093

(1.18)10961096 q(NH) + d(HCH)twist

1100 1057 68.55 1071/1070 –./1078 m(CC) + m(CN) + d(HNH)wagg1077 1035 117.10 1059 vw 1051/1050 1063 vw q(NH)1022 982 61.89 1044 vw 1024 (0.86) 1028/ m(CN)

(continued on next page)

J.M.R

amos

etal./Spectrochim

icaA

ctaPart

A:

Molecular

andBiom

olecularSpectroscopy

97(2012)

1041–1051

1047

Table 2 (continued)

DFT:B3LYP/6-311Gwn

DFT � 0.9613 IRintensities

FT-IR wn(%T)

2d derivative//BDAa

FT-Raman (Bruker)wn

Raman confocalwn

Raman/SERSb 2d/BDA Approximate assignment

10251010 971 21.94 1028 (6) 1007/1007 1010 (0.19) 1008 (0.10), 1005

(1.64)1006/1006

d(HNH)twist + q(CH2)

1000 961 10.25 980/995 q(NH) + q(CH2)962 925 22.50 968 vw 953/955 958 vw m(CC) + m(CO)926 890 26.85 949 vw 944/946 946 vw 922/923 m(CC) + m(CO)910 875 13.66 906 (7) 905/906 895 (0.26) 891 (0.10), 907

(1.65)–/888 d(CCC) + q(CH2) + m(CC)

901 866 15.09 866 (5) 867/866 889 (2.16) 864/875 d(CCN) + q(CH2)857 824 99.08 853 vw 842/838 849 (0.00) 847 (0.08), 840 (0.20) m(CN) + d(CNC)821 789 42.54 827 (13) 823/826 q(NH) + q(NH2)797 766 352.74 782 (3) 793/793 783 (0.03) 782 (0.06) q(NH2)779 749 24.95 741 (5) 742/741 756 (0.04) 754 (0.09) 755/752 q(NH2)768 738 26.01 735 vw –/720 733 (0.08) 736 (0.10) 720/720 q(NH2) + d(CONi) + d(OC@O)745 716 22.42 701 (11) 704/704 714 (0.05) 712 (0.06), 715

(0.46)715/714 Ring distortion

708 681 15.83 684 vw –/687 q(NH) out of plane + d(NCN)694 667 1.26 669 (11) 670/670 671 (0.60) 676/672 q(C@O) out of plane + d(COH) + q (NH2)681 655 15.66 653 (11) 653/654 659 (0.58) 659 (0.10), 658

(0.98)667/660 d(ONiN) 20% + d(CCN) 18% + d(NiNC) 10% + m(CC) 11%

661 635 15.74 630 vw –/643 642 (0.21) 645 (1.58) 646/645 q(C@O) out of plane + d(CNC)599 576 3.78 611 (11) 609/610 614 (0.04) 612 (0.09) m(NiN) 22% + m(NiO) 22% + m(CC) 21%588 565 32.31 587 vw 594/587 601(0.06) d(CCN) 24% + m(NiN) 15% + d(NiNC) 12% + m(NiO) 10%566 544 2.51 564 (8) 579/562 582 (0.01) 582 (0.01) m(NiN)27% + m(CC) 11%533 512 7.81 526 vw 520/522 536 (0.06), 530

(0.56)532/534 Ring distortion 40% + m(CO)8%

507 487 43.19 507 (15) 515/– 525/528 q(NH) of the guanidinoacetic acid492 473 4.56 498 (15) 477/. 488 (0.04) 491 (0.14) 491/492 Ring distortion 38%482 463 62.80 464 (6) 460/467 460 (0.01) 458 (0.11) 438/435 Ring distortion 44%459 441 46.54 443 (2) –/455 449 (0.00) 446 (0.03) m(NiO) 23% + m (NiN) 12% + d(OCC) 11%448 431 73.33 434 (10) 437/434 439 (0.01) 434 (0.09) d(CC@O) 11% + d(OCC) 9%445 428 34.21 421 vw –/427 425 (0.05), 425

(0.17)427/423 d(ONiN) 21% + m(NiN) 14% + d(NiNC) 12% + d(NCC) 12%

400 384 4.37 399 (3) 389/– 382 (0.07) d(ONiN) 20% + m(NiN) + dNiNC) 12% + d (NiOC) 10%366 352 7.55 363 vw 363/359 d(ONiN)19% + m(NiO) 17 + d(CCC)11% + d(NiOC) 11%.360 346 27.76 345 vw 345/344 d(ONiN) 27% + d(NiNC) 15% + r(NiN) 11% + m(NiO) 11%335 322 3.00 318 (25) 318/319 d(CCN) 21% + d(ONiN) 20%316 304 4.67 287 (26) d(ONiN) 28% + m(CC) 14%271 260 1.30 271 vw 270/266 d(ONiO)linear 20% + d(ONiN) 17% + d (NiNC) 12% + d

(NNiN)linear 10%,250 240 0.78 239 vw 240/239 236 (0.80) m(NiO) 22% + d(ONiO) 12%221 212 5.41 215 vw 215/213 d(ONiN) 45%.206 198 12.93 192 vw –/182 d(NiNC) 15% + d(ONiN) 10%173 166 38.96 169 vw 164/160 Torsion140 134 17.76 143(50) 143/140 Torsion116 111 9.85 119 vw –/122 Torsion97 93 0.96 109 vw 111/97 110 (1.0) Torsion89 85 2.20 88/84 Torsion80 77 1.17 66(5) –/60 Torsion51 49 15.96 54 vw 53/– Torsion43 41 2.80 –/45 Torsion34 33 0.77 –/30 Torsion27 26 0.81 –/28 Torsion

a BDA = band deconvolution analysis; In enhanced italic characters = FT-Raman Bruker.b Both spectra were normalized by the band at 889 cm�1 of higher height found with the Raman confocal.

1048J.M

.Ram

oset

al./Spectrochimica

Acta

PartA

:M

olecularand

Biomolecular

Spectroscopy97

(2012)1041–

1051

Table 3Main singlet transition energies (nm) and the oscillator strength from the groundstate of the Ni(GAA)(ASP) complex, with the B3LYP method.

State B3LYP

Dominant configuration⁄ Wavelength (nm) Oscillator strength

A HOMO ? LUMO (85%) 695 0.0001A H-2 ? LUMO (90%) 616 0.0001A H-5 ? LUMO (88%) 523 0.0001A H-8 ? LUMO (50%) 513 0.0011A H-1 ? LUMO (79%) 313 0.0180A H-3 ? LUMO (75%) 289 0.0221A H-4 ? LUMO (80%) 282 0.0205A H-7 ? LUMO (46%)

H-6 ? LUMO (33%) 270 0.0021A HOMO ? L+1 (97%) 253 0.0001A H-7 ? LUMO (43%)

H-6 ? LUMO (43%) 251 0.0008A H-2 ? L+1 (46%)

H-1 ? L+1 (41%) 242 0.0031A H-11 ? LUMO (33%)

H-9 ? LUMO (31%) 241 0.0551A H-2 ? L+1 (37%)

H-1 ? L+1 (52%) 238 0.0007A H-11 ? LUMO (32%)

H-9 ? LUMO (40%) 237 0.0795A H-4 ? L+1 (41%) 234 0.0425A H-4 ? L+3 (41%) 229 0.0007A HOMO ? L+2 (61%) 227 0.0003A H-3 ? L+1 (46%),

H-1 ? L+6 (21%) 221 0.0010A H-3 ? L+1 (26%)

H-1 ? L+5 (15%)H-1 ? L+6 (37%) 221 0.0004

A H-10 ? L+1 (32%)H-4 ? L+1 (16%) 217 0.0167

A H-10 ? L+1 (15%)H-2 ? L+2 (32%) 216 0.0036

A H-10 ? LUMO (26%)H-5 ? L+1 (30%) 213 0.0068

A H-5 ? L+1 (27%)H-1 ? L+2 (40%) 211 0.0024

A H-5 ? L+1 (20%)HOMO ? L+3 (33%) 211 0.0072

A H-1 ? L+2 (20%)HOMO ? L+4 (37%) 210 0.0033

A H-10 ? LUMO (41%) 209 0.0107A HOMO ? L+3 (18%)

HOMO ? L+4 (22%) 207 0.0786A H-4 ? L+2 (19%)

H-3 ? L+2 (15%)H-2 ? L+3 (33%) 205 0.0032

A H-4 ? L+2 (31%)H-3 ? L+2 (17%) 203 0.0143

A H-1 ? L+4 (31%) 202 0.0056A H-4 ? L+2 (21%)

H-3 ? L+2 (25%)H-2 ? L+4 (20%) 200 0.0032

A H-7 ? L+1 (28%)H-6 ? L+1 (62%) 200 0.0006

A H-1 ? L+3 (34%) 199 0.0059A H-1 ? L+3 (29%)

H-1 ? L+4 (36%) 199 0.0003A H-5 ? L+2 (39%)

HOMO ? L+5 (16%) 197 0.0042A H-8 ? L+1 (17%)

H-7 ? L+1 (38%)HOMO ? L+5 (27%) 197 0.0100

A H-5 ? L+2 (16%)HOMO ? L+5 (40%) 197 0.0100

A H-3 ? L+4 (44%) 192 0.0073A H-8 ? L+1 (61%)

H-6 ? L+1 (20%) 192 0.0002A H-3 ? L+3 (20%)

HOMO ? L+6 (22%) 191 0.0219A H-3 ? L+3 (18%)

H-2 ? L+5 (45%) 190 0.0187A H-3 ? L+4 (16%)

HOMO ? L+6 (23%) 190 0.0108

Table 4Koopmans’ energy (eV), Mulliken population analysis and assignment for the frontierorbitals of [Ni(Asp)(GAA)], with the B3LYP method.

Orbital Energy (eV) B3LYP Mulliken population

L+6 0.52 p�C@O [GAA 87%]

L+5 0.38 p�C@N [GAA 71%]

L+4 0.00 r⁄C–H [Ni 23%; GAA 53%]L+3 �0.19 p�

C@O [ASP 89%]

L+2 �0.51 r⁄C–H [ASP 84%]L+1 �1.25 pC@O r⁄ [ASP 97%]LUMO �2.24 Ni3d [Ni 65%; ASP 15%; GAA 20% ]HOMO �6.78 Ni3d + pO [Ni 50%; ASP 34%; GAA 14%]H-1 �6.89 pO + rC–C [GAA 92%]H-2 �7.09 Ni 3d [Ni 87%]H-3 �7.26 pC@N + pN + pO [GAA 67%; ASP 19%]H-4 �7.33 pO + pN [GAA 20%; ASP 62%]H-5 �7.77 Ni3d + pN [Ni 84%; GAA 11%]H-6 �8.00 pC@O + pO + pN + Ni3d [Ni 14%; GAA 71%, ASP 15%]H-7 �8.08 pC@O + pO + pN [GAA 61%, ASP 32%]H-8 �8.29 rNi–O + pO [Ni 45%; GAA 14%, ASP 28%]H-9 �8.41 pO + rNi–O [Ni 11%; GAA 59%, ASP 16%]H-10 �8.80 pO [ASP 94%]H-11 �8.84 pNi–O + pC@O [Ni 31%; GAA 18%, ASP 40%]

Fig. 8. Orbital representation of the HOMO and LUMO, at B3LYP method ofNi[(Asp)(GAA)]. The contour values of the orbitals are all 0.035 a.u.

J.M. Ramos et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 97 (2012) 1041–1051 1049

1050 J.M. Ramos et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 97 (2012) 1041–1051

normal mode description; 1633 (IR)/1640 (SERS) cm�1, d(HNH);1626 cm�1 observed by IR/DBA, d(HNH); 1624 (IR)/1619 (SERS/DBA) cm�1,m(C@O) + d(OCC); and 1582 (IR)/1577 (SERS) cm�1 asm(C@O) + d(HNH) indicating that the internal coordinates C@Ostretching and H–N–H bending participate in the description ofthe normal mode.

Other HNH, HCH bending and rocking vibrationsIn the 545–1250 cm�1 spectral region infrared bands were as-

signed principally to the HCH scissoring, wagging and twistingmodes, and also HNH wagging vibrational modes, all of them beingcoupled modes with considerable participation of the H–C–H andH–N–H bending internal coordinates. The rocking vibrations arenot pure vibrational modes either. The approximate descriptionsof these vibrational modes are given in Table 5.

C–O, C–N and C–C stretchingFor the geometric parameters of the [Ni(Asp)(GAA)] structure

there were defined three C–O, four C–C and four C–N stretchinginternal coordinates. In the vibrational spectra of this complex onlyfive bands were observed which could be assigned to the followingcoupled modes with different internal coordinates: 1236 (IR) cm�1

(1226 cm�1 by DBA) as: m(CO) + d(CCO) + d(CCH) + d(HCH)wagg.;the infrared band observed at 1226 (IR)/1220 (SERS) cm�1

(1217 cm�1, DBA) present some C–C stretching character and canbe considered as a coupled vibrational mode: d(HNH)wagg. + -d(OCC) + m(CC); the observed band at 1044 (IR)/1147 (SERS) cm�1

(1136 cm�1 by DBA) can be assigned as: m(CO) + d(COH); 1059(IR) cm�1(1050 cm�1 by DBA) is probably m(CN); 968(IR) cm�1(955 cm�1 by DBA) can be assigned as the mixturem(CC) + m(CO); 949 (IR)/946 (SERS) cm�1(946 cm�1 by DBA) is alsoprobably m(CC) + m(CO). The very weak band observed at 853 (IR)/847 (SERS) cm�1(838 cm�1 by DBA) can be assigned as the mixturem(CN) + d(CNC).

Framework normal modesFor the inorganic chemist, specially for those which work with

coordination chemistry, it is of great importance to characterizemetal–ligand normal modes, which can be, by example, stretchingmetal–ligand modes and bending metal–ligand modes, so simplesas Ni–O and Ni–N stretching, or in general terms M–O, M–N andM–S stretching modes. Excellent books of relevant scientist de-voted in this research field are devoted to the characterization ofthe normal modes of inorganic and coordination compounds[28,29]. In this vibrational spectroscopy study special attentionwas paid to the low energy region of the infrared and Raman spec-tra where the skeletal or framework normal modes occur. As theGaussian [33] and Gamess [34] packages do not give in matrixform the potential energy distribution (PED); and also do not givethe kinetic energy distribution (more appropriate the kinematicscoefficient distributions) (KED), and the sum know as the total

Table 5Electronic transitions assignment for the [Ni(Asp)(GAA)]complex.

Solid spectra (nm) B3LYP (nm) Assignment

706 695 Ni 3d + pO ? Ni 3d634 616 Ni 3d ? Ni 3d573 523 Ni 3d ? Ni 3d486 513 rNi–O + pO ? r⁄Ni–O + rNi–N

389 313 pO + rC–C ? r⁄Ni–O + r⁄Ni–N

278 289 pC@N + pN + pO ? r⁄Ni–O + r⁄Ni–N

276 282 pO + pN ? r⁄Ni–O + r⁄Ni–N

231 241/237/234 pNi–O + pC@O ? r⁄Ni–O + r⁄Ni–N

pO + rNi–O ? r⁄Ni–O + r⁄Ni–N

pO + pN ? pC@O

217 207 p⁄Ni–O + pO ? p⁄C@O + r⁄C–H

energy distribution (TED), for assisting the vibrational assign-ments, we look here carefully inside the geometrical ‘‘anatomy’’of each normal mode pertaining to the framework vibrations, hav-ing as information the matrix of each normal mode in the Cartesiancoordinate space representation, composed by all the atomic dis-placement resulting from the vibrational movement. Hence wewere able to determine which bond distance or bond angle thatare included in the internal coordinates definitions has more devi-ation from his equilibrium position. Working in this way we avoidassigning normal modes merely or exclusively by a visual ap-proach. We only characterize the ‘‘shape of the vibrations’’ usingthe L matrix obtained through the vibrational calculations usingthe Gaussian package [33]. As we have stated before [11–15], aclear identification of the metal–ligand vibrations are hard to carryout due to the high degree of mixture of the different internal coor-dinates taking part in the description of the normal modes. In thisstudy we selected 11 stretching internal coordinates described as:m(NiO), m(NiN), m(CO), m(CN) and m(CC); 15 bending internal coordi-nates described as: d(NNiO), d(ONiO), d(NNiN), d(NCC), d(CCO) andd(CCC). Two d(NNiO) are localized outside the two five memberring of the complex structure. From a visual approximate descrip-tion [27] of the normal modes we have selected the region 660 to240 cm�1 as the region in which the metal–ligand vibrations havea higher participation.

The infrared observed band at 564 cm�1 (566/544 cm�1 DFT cal-culated values) could be described as a mixture m(NiN)26.6% + m(CC)10.70% considering only the participation of the two internal coor-dinates with higher percentage of variation of the geometric param-eter which described it, among 26 internal co-ordinates selected.The observed band at 507 cm�1 (507/487 cm�1 calculated values)can be assigned as out of plane internal coordinate q(NH)" of theguanidoacetic acid. The band at 443 (IR)/439 (SERS) cm�1

(455 cm�1 by BDA; 459/441 cm�1 DFT calculated values), can be de-scribed as: m(NiO) 22.7% + m(NiN) 12% + d(OCC) 11.1%. Contributionof the m(NiN) stretching appears in the observed band at 399 (IR)/382 (SERS) cm�1 (400/384 cm�1 calculated values), which can beassigned as: d(ONiN) 19.7% + m(NiN)13.6% + d(NiNC) 11.6% + dNiOC)10.4%. The m(NiO) stretching internal coordinate appears with 22%of participation in the normal mode observed by band deconvolu-tion analysis at 239 (IR)/236 (SERS) cm�1. Accentuatedbending character for the observed band at 363 cm�1 can be de-scribed as the principal mixture of: d(ONiN)19.4% + m(NiO)17.3% + d(CCC)11% + d(NiOC) 10.5%. In the same way the bandsobserved at 318, 271, and at 215 cm�1 can be assigned principallyto bending vibrational modes with the following respective descrip-tions: d(CCN) 21.3% + d(ONiN) 20.2%; d(ONiO)linear 20.4% +d(ONiN) 13.6% + d(NiNC) 11.6% + d(NNiN)linear 9.8%, and d(ONiN)45.1%. A full description of the low energy vibrational modes is pre-sented in Table 5.

Conclusions

The synthesis, elemental CHN-O analysis, thermogravimetryand infrared spectroscopy of the [Ni(Asp)(GAA)]�H2O complex havebeen done. The vibrational analysis of the infrared spectrum hasconfirmed the trans Ni–N and Ni–O coordination’s in the complexmixture with the guanidoacetic acid and the aspartic acid that actas ligands. Vibrational assignments of bands in the infrared spec-trum of the [Ni(Asp)(GAA)] complex without the crystallizationwater have been done based on the DFT:B3LYP/6-311G(d, p). Forthe skeletal vibrations the most probable assignment was basedin the interpretation of the distorted geometry of the normalmodes, having as base the study of the percentage of deviationfor the geometric parameters [9–15]. The results suggest the pro-posed structure depicted in Fig. 6 as the most probable, and the full

J.M. Ramos et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 97 (2012) 1041–1051 1051

assignments for the complex are presented in Table 5. Concerningthe NBO analysis, the results indicate that the Ni–O and Ni–Nbonds are formed by the interaction of the sp1.8 (35.76% s and64.24% p) orbital on the oxygen atom, and the sp4.33 (18.77% sand 81.23% p) orbital on the nitrogen atom with the sp2.04d1.04 hy-brid orbital of Ni.

The UV–Vis spectra of the [Ni(Asp)(GAA)] complex were mea-sured in the solid state. The multi-configurationally nature of sev-eral excited states of complex involving intraligand transitions arealso observed. The calculated results are in agreement with theexperimental spectra, confirming the existence of d–d, severalLMCT and MLCT transitions. Analysis of the transition energies,using the TD-B3LYP method, is in excellent agreement with theexperimental results. The CI calculations allowed a more preciseanalysis than the previous one obtained from simple analysis ofthe boundary orbitals.

SERS effect was obtained from a mixture of the [Ni(Asp)(GAA)]complex with AgNP using a laser of 1064 nm from the Bruker spec-trometer and a laser of 785 nm as excitation source from the RiverDiagnostic spectrometer. The fluorescence remains exciting thesample with the 1064 nm laser, and only a few bands hadpresented appreciable enhancement. With the 785 nm laser almostall the bands shows enhanced intensity, principally those observedat: 645 cm�1 and assigned to the mode q(C@O) + d(CNC),enhanced by a factor of 7.0; 889 cm�1 assigned as d(CCN) + q(CH2)and enhanced by 8.0; 907 cm�1 assigned to the moded(CCC) + q(CH2) + m(CC) increasing by 17. The band observed at1006 cm�1 assigned to d(HNH) twist + q(CH2) is 9 times more in-tense, and so on. Assignment of the SERS bands and his intensitiesare given in Table 5. SERS deconvolution band analyses are pre-sented in Fig. 6S.

Acknowledgments

The authors acknowledge the Brazilian agencies CNPq, FAPERJand Capes (PNPD) for financial assistance and research grant. Profes-sor Téllez thanks the FAPESP for grant of Visiting Professor at UNI-VAP – LEVB (Laboratory of Biomedical Vibrational Spectroscopy).

Appendix A. Supplementary data

The interaction between the atoms in [Ni(Asp)(GAA)], by thedegree of overlap, for the B3LYP method, species distribution dia-gram for the [Ni(Asp)(GAA)]�H2O synthesis, TGA analysis, compar-ative experimental and calculated infrared and Raman spectra,molecular orbital diagram, density of states diagram (DOS) andOrbital Overlap Population diagram (OPDOS) of the [Ni(Asp)(GAA)]complex, with the B3LYP method, calculated UV–Vis spectrum andband deconvolution analysis for the SERS spectrum. Supplemen-tary data associated with this article can be found, in the onlineversion, at http://dx.doi.org/10.1016/j.saa.2012.07.087.

References

[1] O.V. Cabral, Síntese e caracterização de compostos de aminoácidos com metaisde transição empregando metodologia gráfica, Tese de Doutorado, PUC-Rio,2005.

[2] J.K. Swearingen, D.X. West, Trans. Met. Chem. 26 (2001) 252–260.[3] N.K. Singh, S.B. Singh, Trans. Met. Chem. 26 (2001) 487–495.[4] S. Talwar, U. Bansal, S. Chandra, R. Singh, P. Veeraswamy, Synth. React. Inorg.

Met. Org. Chem. 21 (1991) 131–142.[5] J.R Lancaster, The Bioinorganic Chemistry of Nickel, First ed., John Wiley &

Sons, 1988.[6] Sulekh Chandra, Smriti Raizada, Monika Tyagi, Archana Gautam, Bioinorganic

Chemistry and Applications, vol. 2007, 2007 (Article ID 51483, 7p). http://dx.doi.org/10.1155/2007/51483.

[7] M. Sung, K. Park, Exp. Mol. Med. 31 (2) (1999) 64–69.[8] M.A. Pearson, I.S. Park, R.A. Schallewr, L.O. Michel, P.A. Karplus, R.P. Hausinger,

Biochemistry 39 (29) (2000) 8575–8584.[9] J.M. Ramos, O. Versiane, A. Claudio, S. Téllez, Spectrochim. Acta A 67 (2007)

1046–1054.[10] J.M. Ramos, O. Versiane, C.A. Téllez Soto, Spectrochim. Acta A 67 (2007) 1037–

1045.[11] J.M. Ramos, O. Versiane, J. Felcman, A. Claudio, S. Téllez, Spectrochim. Acta A 72

(2009) 182–189.[12] J.M. Ramos, M.T.M. Cruz, A.C. Costa Jr., O. Versiane, C.A. Téllez Soto, ScienceAsia

37 (2011) 247–255.[13] O. Versiane Cabral, J. Felcman, A. Claudio, S. Téllez, Tiago Giannerini,

Spectrochim. Acta A 61 (2005) 337–345.[14] G. Faget O, T. Giannerini, J. Felcman, A. Claudio, S. Téllez, Spectrochim. Acta A

61 (2005) 2121–2129.[15] J.M. Ramos, G. Faget, O.J. Felcman, A. Claudio, S. Téllez, Spectrochim. Acta A 71

(2008) 1364–1370.[16] R.G Parr, W. Yang, Density Functional Theory of Atoms and Molecules, Oxford

University Press, New York, 1989.[17] G.A. Petersson, M.A. Al-Laham, J. Chem. Phys. 94 (1991) 6081–6090.[18] A.D. Becke, Phys. Rev. A 38 (1988) 3098–3100.[19] A.D. Becke, J. Chem. Phys. 98 (1993) 1372–1377.[20] A.D. Becke, J. Chem. Phys. 98 (1993) 5648–5652.[21] L. Antoloni, L. Menabue, G.C. Pellacani, J. Chem. Soc., Dalton Trans. (1982)

2541–2543.[22] A.W. Addison, Spectroscopic and redox trends from model copper complexes,

in: K.D. Karlin, Zuibietas (Eds.), Copper Chemistry: Biochemical and InorganicPerspectives, Academic Press, New York, 1983.

[23] J.P. Foster, F. Weinhold, J. Am. Chem. Soc. 102 (1980) 7211–7218.[24] A.E. Redd, F. Weinhold, J. Chem. Phys. 78 (1983) 4066–4073.[25] A.E. Redd, R.B. Weinstock, F. Weinhold, J. Chem. Phys. 83 (1985) 735–746.[26] L.J. Bellamy, R.L. Williams, Spectrochim. Acta 9 (1957) 341–345.[27] Chemcraft, <http://www.chemcraftprog.com>.[28] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination

Compounds: Applications in Coordination, Organometallic, and BioinorganicChemistry, John Wiley & Sons, 2009.

[29] J.R Ferraro, Fourier transform infrared spectroscopy applications to chemicalsystems, Academic Press, 1985.

[30] N.M. O’Boyle, A.L. Tenderholt, K.M. Langner, J. Comput. Chem. 29 (5) (2008)839–845.

[31] L. El-Sayed, M.F. Iskander, A. El-Toukhy, S.E. Zayan, Inorg. Chim. Acta 6 (4)(1972) 663–668.

[32] A. Atanassov, G.ST. Nikolov, Inorg. Chim. Acta 68 (1983) 15–23.[33] 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. Fukuda, J. Hasegawa,M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox,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. Dannenberg, 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, Q. Cui, A.G. Baboul, S. Clifford, J.Cioslowski, B.B. Stefanov, G. Liu, 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 C.02, Gaussian, Inc., Wallingford, CT, 2004.

[34] GAMESS-UK is a package of ab initio programs. See: <http://www.cfs.dl.ac.uk/gamess-uk/index.shtml>, M.F. Guest, I.J. Bush, H.J.J. van Dam, P. Sherwood,J.M.H. Thomas, J.H. van Lenthe, R.W.A Havenith, J. Kendrick, The GAMESS-UKelectronic structure package: algorithms, developments and applications,Molecular Physics, vol. 103, No. 6–8, 20 March–20 April 2005, 719–747.