Accepted Manuscript Experimental and theoretical investigation of the spectroscopic and electronic properties of pyrazolyl ligands Adebayo A. Adeniyi, Peter A. Ajibade PII: DOI: Reference:

S1386-1425(14)00918-4 http://dx.doi.org/10.1016/j.saa.2014.06.030 SAA 12289

To appear in:

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy

Received Date: Revised Date: Accepted Date:

2 April 2014 25 May 2014 3 June 2014

Please cite this article as: A.A. Adeniyi, P.A. Ajibade, Experimental and theoretical investigation of the spectroscopic and electronic properties of pyrazolyl ligands, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy (2014), doi: http://dx.doi.org/10.1016/j.saa.2014.06.030

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Experimental and theoretical investigation of the spectroscopic and electronic properties of pyrazolyl ligands Adebayo A. Adeniyi and Peter A. Ajibade Department of Chemistry, University of Fort Hare, Private Bag X1314, Alice 5700, South Africa

Abstract The electronic and spectroscopic properties of seven pyrazole derivatives are presented in order to give a clear understanding of their distinguishing features. Four out of the seven ligands are synthesised and are also characterised experimentally. A very high correlation was observed between the experimental and the theoretical IR, 1H-NMR and

13

C-NMR, which help in the

characterisation of the ligands. The excitation properties computed using the TDDFT shows that most of experimentally observed absorptions of the ligands are predominantly form either the HOMO or HOMO-1 to LUMO or LUMO+1. The characteristic features of the *N atoms (i.e. metal available coordinating centre) shows that the carboxylic unit may possibly decrease the metal affinity of the pyrazole unit while the pyridine unit will increase the affinity. The conductivity properties of the seven ligands are found to be in the order of bdmpzpy > bpzpy > bphpza > bdcpzpy > phpz > dcpz. The J-coupling of *N-N can give an insight into the variation in their bond distance, bond stretch and bond strength in the ligands. Also the atomic properties of the *N atoms and their *N-N bonds can help in the molecular characterisation, differentiation and in prediction of the non-linear optical properties of the ligands as conductive materials.

Keywords: Pyrazole derivatives, Raman spectra, Hyperpolarizabilities, NMR shift, Ramsey terms, QTAIM properties .

*Correspondence email: [email protected] Tel: +27 (0) 40 602 2055, Fax: +27(0) 86 518 12225

Introduction The pyrazole is a unique ligand that has received several research attentions since the first successful synthesis of its derivatives called scorpionate ligand by Trofimenko [1, 2]. Since then many of the derivatives of pyrazole have been synthesised that are not linked with boron atom but with different organic linking units like CHCOOH, CH2 and pyridine derivatives that can easily be modified [3, 4]. Many of the pyrazoles derivatives are designed as chelating ligands [5-12]. These ligands have significant application especially in coordination chemistry dues to their strong coordinating affinity with many of the transition metals [4, 13]. In this paper, our interest is to compare the electronic and spectroscopic properties of seven derivatives of pyrazole and their possible application as non-linear optical (NLO) material. Also of interest to us is the possible availability of the nitrogen donor site for metal coordination that is a useful consideration in the design of pyrazole derivatives for the synthesis of metal complexes for various applications. To achieve this, we have presented the electronic, spectroscopic and conductive properties of the seven derivatives of pyrazole (Figure 1) to show the effect of methyl, carboxylic, pyridine and phenyl on the stabilities, conductivity, availability of the nitrogen atom of pyrazole for metal coordination (denoted as *N) and the strength of the *N-N bonds of the pyrazole. The theoretical spectroscopic are also correlated with the experimental spectroscopic in an effort to give at a glance the distinguishing features of their IR, Raman, 1H-NMR, 13C-NMR, 15N-NMR and UV of the derivatives. We strongly believe that the information provided at both experimental and theoretical will be of tremendous help in the design of the organometallic complexes of pyrazole derivatives and their tailoring toward specific uses in biomedical and non-linear optical (NLO) applications.

Experimental and computational methods Physical measurements The Fourier Transform Infra Red (FTIR) spectra of the ligands were measure using on a Brukers Vector 22 spectrophotometer, the Raman spectra were obtained using Bruker Fourier Transform Raman spectrometer (model FRA 106), UV absorption spectroscopic properties were determined using Hewlett–Packard 8452A diode array spectrophotometer while the NMR chemical shifts were determined using Bruker-AVANCE NMR spectrometer at 600 MHz. The samples were dissolved in the deuterorated methanol (CD3OD) or DMSO (DMSO-D6). Both 1H-NMR and 13C-NMR chemical shifts are reported in δ relative to tetramethylsilane (TMS) as reference.

Synthesis of 3,5-diaceticpyrazole (dcpz) The ligand 3,5-dicarboxylpyrazole (dcpz) was synthesised by oxidation of the 3,5dimethylpyrazole. The method used is similar to the methods reported for the oxidation of the methyl group of 4-methyl-1, 10-phenanthroline [14]. The 3,5-dimethylpyrazole (1 g) was refluxed for 4 h with selenium oxide (2.5 g) in dioxane containing 4% water and filtered through Celite 521 while hot. The aldehyde obtained was oxidized with HNO3 to give 3,5-dicarboxylpyrazole (dcpz).

Synthesis of bis(3-phenylpyrazol-1-ly)acetic acid (bphpza) The simple method used in the synthesis of bis(phenylpyrazol-1-ly)acetic is very similar to the reported method in the literature [13]. Both phenylpyrazole (40 mmol, 5.76g) and dibromoacetic (20mmol, 4.355g) was added to THF solvent. The reaction was reflux for 5-6 hours. The product was dry in vacuo, 80ml of water was added and was acidified with half diluted HCl first to a PH of 7.0 and extracted with Et2O (200ml). It was further acidified to PH 1.0 and extracted twice with Et2O (2 x 100ml). The organic solvent was dry with Na2SO4 and evaporated in vacuo. To the oily product was added acetone and dry in vacuo to afford a white powder which was further recrystallized with acetonitrile.

Synthesis of bis(pyrazol-1-ly)pyridine (bpzpy) The method employed here is very simple, it is a combination of different methods reported in the literatures [5, 15] without the need for inert or living the reaction for many days has was reported in the literature. The simple method used here is a bit similar to the method used for the synthesis of bis(pyrazo-1-ly)acetic [13]. Both pyrazole (30 mmol, 2.04375g) and dichloropyridine (5.21mmol, 1.235g) were added to THF with K2CO3 (75.7mmol, 10.45g), KOH (77.7mmol, 4.35g) and BTEA (1.0g) as phase transition. To allow complete replacement of the bromides, a 6:1 mole of pyrazole:bromopyridine was used. The reaction was reflux for 24 hours. The product was dry in vacuo, 80ml of water was added and was acidified with HCl to form precipitate which was filtered and wash with water.

Synthesis of bis(3,5-dimethylpyrazol-1-ly)pyridine (bdmpzpy) The same method for the synthesis of bis(pyrazo-1-ly)pyridine (bpzpy) was employed. Methylpyrazole (30mmol, 2.88g) was reacted with dichloropyridine (5.21mmol, 1.235g).

Computational method The seven pyrazole derivatives are optimized using PBE0 hybrid density functional [16] and basis set

6-31+G(d,p).

Further

computation of properties

like

hyperpolarizability (NLO),

magnetizabilities and nmr shielding and spin-spin coupling constant were done using Becke’s threeparameter exchange [17] and Lee-Yang-Parr’s correlation nonlocal functional denoted as B3LYP with the same basis set. We specifically adopted hybrid functional like B3LYP in computing spinspin coupling constants because of its reported better performance as hybrid functionals [18]. The excitation properties were computed using TDDFT method [19, 20] while the NMR and one-bond NMR spin-spin coupling constants J(A,B) [21, 22] were computed using the GIAO method. The NMR chemical shifts of the ligands were computed theoretically using a direct method and a fitting

equation. In a direct method, the isotropic shielding of the proton, carbon and nitrogen are subtracted directly from that of a reference compound (tetramethylsilane (TMS) for 1H and 13C and CH3NO2 for 15N [23, 24]) while in the fitting method, the following equation is used as reported in the literature [23]: σ1 H= 31.0 – 0.97 *σ1H σ13C= 175.7 – 0.963 *σ13C σ15N= -152.0 – 0.946*σ15N

where the *σ1H, *σ13C and *σ15N represent the computed isotropic shielding tensor of proton,

carbon and nitrogen respectively. All the computation was done using Gaussian 09 (G09) [25] except for the atomic electronic

occupation and energy level that were computed using Gamess Firefly 08 [26]. The QTAIM properties were computed using the wave function files (WFX) files from the G09 output with the

help of AIMAll program package [27]. The Proaim integration approach was applied in computing the intra- and inter-atomic properties as implemented in the AIMAll suite of programs. The orbital contributions were analysed using Gausssum [28] while the orbital surface constructed using Gabedit [29]. The spectra of the ligands are assigned using the method of the potential energy distribution (PED) contributions as implemented in package called Vibrational Energy Distribution Analysis (VEDA4) [30]. This package optimizes the set of internal coordinates for elucidation of IR and Raman experimental/theoretical spectra. The method adopted in for its application is as explained in the literatures [31, 32].

The energy of a system subjected to static electric field can be computed using: 0

E= E − µ i F i −

1 1 1 γ F F F F − ... α F F − β F F F − 2 ij i j 6 ijk i j k 26 ijkl i j k l

where E0 is the energy of the molecule in the absence of an electronic field; µi is the component of

the dipole moment vector; αij is the linear polarizability tensor, βijk and γijkl are first and second

hyperpolarizability tensors where i, j, and k label are the x, y, and z components respectively [33].

The components that were computed for the ligands are dipole (µ=(µ2x + µ2y + µ2z)1/2), anisotropy polarization (= 1/3 (αxx + αyy + αzz); ∆α1= 1/2 (αxx + αyy) – αzz; ∆α2= {1/2 [(αxx - αyy)2 + (αxx αzz)2 - (αyy – αzz)2 + 6 (α2xy + α2 xz + α2yz)]}1/2 and ∆α3= [(∆α2)2 – (∆α1)2]1/2) and the first hyperpolarizability tensors (ß). The single first static hyperpolarizability (ß) are computed using Quasi-Pythagorean equation ßtot= (ßx2 + ßy2 + ßz2)1/2 as reported in the literature [34].

Results and discussions The IR and Raman spectra The experimental and theoretical IR spectra of the ligands are shown in Figure 2. The spectra of the ligands are assigned as shown in Table 1. Besides the OH and NH vibrations, which are theoretically overestimated, there is a very strong correlation between the theoretical and the experimental spectra (Figure 2), which help to establish the successful synthesis of the ligands. The correlation (R2) between the experimental and theoretical IR of the ligands dcpz, bphpza, bpzpy and bdmpzpy are found to be 0.838, 0.906, 0.997 and 0.995 respectively using the comparison by weighted linear regression as implemented in the program package called Specs [35]. Starting from the ligands dcpz, there is an evidence of successfully oxidation of the methyl group of the 3,5dimethylpyrazole with the presence of C=O vibration at 1750. Also another similarity is the two finger peaks of high frequency but weak vibration that are found around 3000 in the experiment and around 3600 in the theoretical which are assigned to both N-H and O-H vibrations. The theoretical O-H vibrations for the ligands dcpz, bphpza and bdcpzpy with carboxylic unit are found above 3600 while the experimental are found around 3000. In bphpza, the distinguishing features theoretically and experimentally is the C=O vibration at 1750 which is observed to be Raman inactive. In the presence of the C=O vibration in ligands like dcpz, bphpza and bdcpzpy, the C=C vibrations around 1600 remain invisible in the experimental and theoretical IR but very strong in the experimental Raman spectra. Other vibrations in the ligands bphpza, bpzpy and bdmpzpy which were found to be

Raman active are CH, HCC, CC, NC, NN, CCN, CCNCout and CCC as shown in Figure 2 for the ligands bphpza, bpzpy and bdmpzpy. A distinguishing feature of the tridentate bpzpy and bdmpzpy from bidentate bphpza is the experimental Raman vibration within 270 to 300 which are assigned to the CCN and CCNCout respectively. In both theoretical and experimental IR, the C-H vibrations around 3100 are disappearing in bpzpy. Some few bands that are experimentally inactive bands are found to be theoretically active bands and also vice versa that according to literature could be as a result of experimental overtones and presence of impurities [36]. However, the qualitative intensity agreement obtained is quite sufficient for good experimental spectra interpretation.

The NMR The experimental proton and

13

C NMR were carried out for the synthesised ligands dcpz, bphpza,

bpzpy and bdmpzpy except for the dcpz that only the proton NMR was considered. The proton, the carbon and nitrogen chemical shift of the seven ligands are computed theoretically using a direct method and a fitting equation. The 13C-NMR shift There is a very high correlation between the 13C-NMR shifts obtained from the experimental and from that of the two theoretical methods (Table 2). In ligand bdmpzpy, the experimental carbon shift of CH3 ranges from 13.724 to 14.892, direct methods gives a range of 15.11 to 15.49 while fitting gives 6.20 to 6.57. The experimental range of 13C-NMR shift of C4 is 105.040 to 109.791, C5 is 125.188 to 138.991 and C3 is 132.232 to 140.791. The theoretical direct method gives 101.91 to 105.71 for C4, 122.80 to 136.64 for C5 and 139.60 to 151.20 for C3 while the fitting method gives the ranges 89.79 to 93.45, 109.91 to 123.24 and 126.08 to 137.20 respectively for same set of carbon shifts. In many of the

13

C-NMR shifts, the direct method gives better values within the

experimental ranges than the fitting method. The carbon shift of CHCOOH in bphpza using the direct (76.65) method is closer to the experimental (74.015) than the fitting (65.47) method while in that of CHCOOH the fitting (151.46) method gives value closer to the experimental (153.219) than

direct (165.95) method. The carbon chemical shift of CH3 in the ligands is the lowest with computational values ranging from 15.11 to 15.49 using the direct method while the experimental ranges from 13.72 to 14.89. In all the ligands, the C4 chemical shift ranges from 98.83 in phpz to 109.84 in bdcpzpy. There is a little shift in the NMR of C4 from phpz (98.83) to dphpzm (100.77) and to dphpza (102.19). The order of the chemical shift of C4 within the ligands is phpz < dphpzm < dphpza < dcpz < bpzpy < bdmpzpy < bdcpzpy. The chemical shifts of C4 are the lowest among the carbon atoms of the monodentate (dcpz and phpz) and tridentate (bpzpy, bdmpzpy and bdcpzpy) except for carbon atom of the CH3 unit in bdmpzpy. The carbon atom in bidentate (bdphpzm and bdphpza) with the lowest chemical shift is the either the linking carbon CH2 in bdphpzm (68.32) or CHCOOH bphpza (76.65). There is significant carbon shift of C25 (157.36) and C24 (116.99) in bdcpzpy as a result of their H-bonding when compared to the C21 (152.92) and C26 (112.41) respectively in the same chemical environment. The order of

13

C-NMR shift in tridentate ligands

bpzpy, bdmpzpy and bdcpzpy which are made up of two pz and one pyr units are C4 < Cpy-m < C5 < Cpy-p < C3 < Cpy-o except in bdmpzpy where Cpy-p of pyr is lower than C5 of pz. The order in the two bidnetate bphpzm and bphpza with phenyl (ph) and two pyrazole (pz) units is CH2 < CHCOOH < C4 < Cph-o < Cph-p < Cph-m < C5 < Cph < C3 < CHCOOH. When considering all the carbon atoms across the seven ligands, CH3 has the lowest chemical shift while CHCOOH has the highest.

The 1H-NMR shifts There is very high correlation between the experimental and theoretical (Table 3) proton shifts. The theoretical direct method gives a range of 2.07 to 2.52 for the proton shift of CH3, the fitting gives the range 2.53 to 2.96 while the experimental values obtained is within 2.317 to 2.673 for ligand bdmpzpy. The direct method shift for the COOH unit of bphpza is 6.56, the fitting method gives 6.89 while the experimental is 6.452. The proton shift at C4-H from the experimental analysis of bphpza, bpzpy and bdmpzpy are within the range of 6.040 to 7.174, the direct method

range is 6.22 to 6.84 while the fitting method ranges from 6.55 to 7.15. Interestingly also, both the experimental and theoretical methods indicates C4-H in bdmpzpy as lowest and the one in bphpza as the highest. Though in all the proton shifts both direct and fitting methods give values that are very closed to the experimental yet the direct method gives better close values. The highest proton shift is observed in the monodentate dcpz and phpz with N-H having 1H-NMR ranging from 9.32 to 10.74. The order of the proton shift does not directly follow the order of the corresponding carbon atoms. Considering the level of proton across the ligands, the lowest shifts are observed in CH3 of bdmpzpy followed by those from CH2 linking unit. The proton shift of the COOH is lower than that of C4-H but higher than that of CH2. It what pointing out also that the proton shift of CHCOO as another linking unit is higher than that of COOH. The proton shift in the bidentate bphpzm and bphpza follows the order CH2 < COOH < C4-H < CHCOO < Cph-p-H < Cph-m-H < fCph-or-H < C5-H < cCph-o-H except for C5-H that is lower than Cph-p-H and C5-H' lower than Cph-o-H in bphpzm (the superscript “c” and “f” on Cph-o mean close to and far from the nitrogen which is the coordination centre of the pyrazole). In tridentates bpzpy, bdmpzpy and bdcpzpy, the proton shift follow the order CH3 < COOH < C4-H < Cpy-m-H < Cpy-p-H < C5-H < C3-H except in bdcpzpy where the Cpy-pH is lower than one of the Cpy-m-H that is closed to the coordinating centre nitrogen atom due to twisting of one of the pyrazole units (Figure 3).

The 15N-NMR shift The nitrogen shift of the nitrogen atom that is not available for metal coordination (N) is far higher than that of the coordinating centre that is available for metal coordination (N2) as shown in Table 4 which indicates that the N are more shielded than the *N atoms. In the tridentate which contain an extra coordinating centre on pyridine, the nitrogen shift of the coordinating centre on pyridine (Npy) is higher than other two in the pyrazole units. The two *N coordination centre on two pz units of bidentate and tridentate ligands have equal shift only bpzpy and bdmpzpy but different

shift in other ligands with the highest difference observed in bdcpzpy. The *N atom of the pz units in bdcpzpy with the higher shift from the reference is the one that face the pyridine nitrogen while the one twisted away is lower. Across the ligands, the nitrogen shift of N2 atoms is in the order of bdcpzpy < dcpz < bpzpy < bdmpzpy < bphpza < bphpzm < phpz. This order shows that carboxylic and pyridine units have significant de-shielding effect on the N2 atoms of the pyrazole units.

The nuclear spin-spin coupling The total nuclear spin-spin coupling (J-coupling) was determined from the four Ramsey terms which are Fermi Contact (FC), spin dipole (SD), diamagnetic spin orbit (DSO) and paramagnetic spin orbit (PSO)) [37] as shown in Table 5. In all the ligands, the FC is the most significant Ramsey term that determines the J-coupling of the *N-N bonds. The two *N-N bonds in bidentates ligands bphpzm and bphpza have significantly different Ramsey terms and J-coupling compare to those in tridentates bpzpy, bdmpzpy, bdcpzpy. The FC and J-coupling of *N-N in the ligands are in the order of bpzpy > bdmpzpy > bphpza > bphpzm > bdcpzpy > phpz > dcpz if the highest value is considered in each ligand. The level of coupling for the non-bonding *N···*N and Npy-*N are also considered. In this case, the FC and J-coupling have positive values. The FC term and the J-coupling of Npy-*N are far higher than *N···*N. The highest values of FC and J-coupling for *N···*N is observed in bphpza while the lowest is observed in bdcpzpy. In bonding *N-N coupling, the FC and PSO are the most significant Ramsey terms while the SD is the lowest. The DSO is also very low which agree with the literatures report that it is least interesting and also the least investigated mechanism [38]. However, in non-bonding *N···*N or Npy-*N, the DSO is more significant than the PSO though still very low. The H-bonding N2···C interactions in bdcpzpy have J-coupling of -1.41E-001 while O···H H-bonding in bphpza have the J-coupling of -2.18E-001 which are higher in magnitude than the non-bonding *N···*N J-coupling. The corresponding strength of interactions of the three H-bonding in the bdcpzpy and bphpza are predicted to be 0.051 and 0.038 from the QTAIM properties (Figure 3). The J-coupling and the corresponding strength of

H-bonding interactions are not in the same order. The J-coupling of the *N-N bonds is found to be highly correlated with their bond distance, laplacian of electron density (∇2(ρ)), ellipticity (ε) and magnitude of V/G as shown in Figure 4. The bond strength of *N-N which is represented as the laplacian of the bond (∇2(ρ)) is also found to be highly correlated with the FC and PSO. As shown in Figure 4, ligands 5 and 6 are found to have the highest magnitude of bond length, bond strength and FC but lowest magnitude of V/G, ellipticity and PSO. Ligands 1 and 2 are found mostly to have the reverse of what was observed for ligands 5 and 6. As was observed in the seven ligands (Figure 4), the magnitude of *N-N J-coupling is found to be inversely proportional to the bond length and bond strength (∇2(ρ)) while it is directly proportional to the bond ellipticity and magnitude of V/G. Also the magnitude of their ∇2(ρ) is inversely proportional to the FC but directly proportional to its PSO. The correlation of other Ramsey terms with the bond properties of the *N-N are shown in Table 6. The implication is that the difference in the experimentally measurable parameters like *N-N J-coupling can give an insight into the variation in the *N-N bond distance, bond stretch and bond strength of the ligands.

The bond and atomic properties The geometrical features with the strength of selected bonds and the amount of localized electrons one selected atoms that are obtained through QTAIM analysis are shown in Figure 3. The strength of the *N-N bonds of the pyrazole unit in the ligands is in the order of dcpz > bdcpzpy > phpz > bphpzm > bphpza > bpzpy > bdmpzpy (Figure 3). The stronger *N-N bonds are also associated with shorter bond distance (Bond length), bond stretch (Bond stretch) and higher electron density (ρ(r)). The values of the Laplacian of the electron-nuclear attractive contribution to virial field (∇2Ven) in *N-N bonds does not directly follow the order of their strength as bdcpzpy is found to have the highest magnitude and phpz the lowest. This order shows that the carboxylic group on pyrazole unit results to a stronger *N-N bonds but the methyl group results to weaker *N-N bond. Considering the Laplacian of the electron density on the pyrazole ring in the ligands, the order is

bdcpzpy > bphpza > bphpzm > bdmpzpy > bpzpy > dcpz > phpz. One intramolecular H-bonding is found in both bdcpzpy and bphpza while others have none. The strength of the H-bonding ranges from 0.038 to 0.051, which is within the suggested range of H-bonding in the literature [39]. The ligand dcpz and bdcpzpy with carboxylic on the pyrazole have the lowest localized electrons (lower than 6.00) on the coordinating centre N atom while others are little higher (6.05) with the highest found in bdmpzpy (6.08) that has methyl group on its pyrazole unit (Figure 3). There is therefore a possibility that the ligands with the carboxylic group on its pyrazole will have lesser coordinating affinity to metal than those with methyl group. Also, the

py

N atom of the

pyridine group in bpzpy and bdmpzpy is characterised with higher localized electron (6.50) and will possibly have the strongest affinity for metal coordination. Generally, the coordinating centre N2 atoms are distinguished from the other N atom with the characteristic features of higher magnitude of intra-atomic dipole (µintra(A)), bonding dipole (µbond(A)), total dipole (µ(A)), atomic volume (Vol(A)), intra-atomic magnetizability (χIntra_Iso(A)), intra-atomic shielding tensor (σIso(A,A)), total isotropic shielding tensor (σIso(A)), Anisotropy, total electron occupation (Occ) but lower atomic charge q(A), bonding magnetizability (χbond_Iso(A)), total magnetizability (χIso(A)), bonding shielding tensor (σIso(A',A)), energy level (Table 7). The out of plane magnetizability (χzz(A)) does not follow the specified order. The few exceptions to the observed properties of N atoms are the lower intra-atomic magnetizability (χIntra_Iso(A)) in dcpz, phpz and bdmpzm; lower σIso(A,A) and σIso(A) in phpz, lower Occ and higher energy level in bphpza. The sum total of the computed atomic QTAIM properties over all the atoms in each of the ligands shows that bdcpzpy and bphpza are having the highest dipole properties (µintra(A), µbond(A) and µ(A)) while phpz have the lowest. The atomic properties of the *N atom are found to be highly correlated with their *N-N and *N···*N bond properties (Table 7 and Figure 5). The *N-N J-coupling is found to be highly correlated with the intra-atomic dipole of the *N atom (Figure 5). The non-bonding distance of the *N···*N in the ligands are also found to be highly correlated with the *N chemical shift, isotropic shielding, intra-atomic magnetizability, atomic dipole and the atomic charge. The correlations of the

*N···*N distances are found to be directly proportional with the magnitude of the *N atomic parameters in Figure 5 except the isotropic shielding and atomic dipole. As shown in Figure 5, ligands 3 and 4 have the highest magnitude of intra-atomic magnetizability, atomic dipole and atomic charge which consequentially resulted to their lowest magnitude of the bond length, *N chemical shift and isotropic shielding. Ligand 7 is found to have the reverse of the observations in ligands 3 and 4. These interesting relationships can be of significant help in the characterisation and differentiation of these ligands.

The UV and excitation properties The experimental and theoretical UV absorption for the ligands are shown in Figure 6. A very broad experimental absorption was observed for bpzpy but found theoretically to be two distinct peaks (at 283.79 and 327.49) of almost the same strength within the range of the broad experimental spectra. Experimentally, we observed a bathochromic (blue) shift from bphpza to bpzpy but red shift from bpzpy to bdmpzpy. In the theoretical absorption, there is a blue shift in the λmax in the order of bdmpzpy > bphpza> bdcpzpy >bpzpy > bphpzm > phpz > dcpz. The experimental λmax observed for dcpz, bphpza, bpzpy and bdmpzpy at 276.00, 286.00, 308.00 and 288.00 respectively can be ascribed to the theoretical λmax at 257.58, 299.56, 283.79 and 331.82 (supplementary Table S1). These experimental λmax absorption are found to be dominated with H2→LUMO, HOMO→L+1, HOMO→L+1 and HOMO→LUMO for the bphpza, bpzpy and bdmpzpy respectively (Table S1). Also, in ligand phpz, bphpzm and bdcpzpy which are considered only theoretically, the λmax absorption are predominately HOMO→LUMO, H-1→LUMO and HOMO→L+1 respectively. The level of the contribution of the N atoms to the energy level is also considered for each of the ligands (Table S2). The orbital contribution of non-available N atom is considered only in monodentate dcpz and phpz while only the available coordinating centre *N atoms are considered in the bidentate and tridentates. The *N atoms contributes significantly to the HOMO levels than

the N atoms. Most of the orbital contributions of the *N atoms are found at the HOMO-4 or HOMO-5 except in ligand bdcpzpy where the highest *N atoms contribution is found at HOMO-1. It is also interesting to point out that the coordinating centre of the pyridine group (Npy) contributes more to the LUMO than the *N atoms in the pyrazole units of the tridentate bpzpyr, bdmpzpy and bdcpzpy. The features of the high contribution of the N2 atoms to HOMO and Npy to the LUMO are shown in Figure 7 for ligand bpzpy. It is obvious from Figure 7 that the HOMO-5 are predominantly the *N atoms of the pyrazole while the LUMO is predominantly the pyridine unit. The orbitals where there is negative contributions of the *N atoms are shown to have characteristic features of big red isodensity surface around the molecule as shown for LUMO+2 in Figure 7 (Table S2).

The molecular properties in relation to the properties of *N atoms. The conductive properties of the ligands are shown in Table 8. The ligand bdmpzpy have the highest molecular dipole moment while bphpza have the lowest. The trend of the conductive properties of the ligands in terms of the hyperpolarizability is bdmpzpy > bpzpy > bphpza > phpz> bdcpzpy > bphpzm > dcpz. It is obvious that the tridentate bdmpzpy and bpzpy have more potential for the application as non-linear optical (NLO) materials than the rest of the ligands. The ligands bpzpy and bdmpzpy which are found to be better NLO material are having the highest dipole and the ligands dcpz which have the poorest conductive properties is characterised also with the widest energy band gap. Other properties computed are the polarizability exaltation index (Γ = α(mol) – Σii) and hardness (η=1/2 (IE - EA) ≈ 1/2(εHOMO – εLUMO). The values of the Γ have been used to estimate the relative aromaticity of furan homologues and stabilities of atomic clusters [40]. A large negative Γ value denotes a very stable structure while the reactivity of series of compounds can be predicted from the molecular property of their hardness (η) [40, 41]. The value of η is simply the half of the value of the band gap. The most stable ligand is phpz that is characterised with the highest magnitude of  but it is not the hardest ligand since its η is lower than that of dcpz

contrary to the expectation [40]. As shown in Table 9 and Figure 8, the molecular properties of the ligands are found to be highly correlated with the atomic properties of *N atoms and their *N-N bonds. A very high correlations were observed between the hyperpolarizabilities of the ligands and their *N-N Jcoupling, bond length, ∇2(ρ), ellipticity, FC and *N intra-atomic dipole. The magnitude of the *N-N J-coupling, FC and ellipticity are inversely proportional to the hyperpolarizabilities while others properties in Figure 8 are directly proportional. Ligands 5 and 6 which are found to have the highest hyperpolarizabilities values are also found to have the highest magnitude of *N-N J-coupling, bond length, FC and *N intra-atomic dipole but lowest magnitude of Laplacian and ellipticity (Figure 8). The properties of ligand 1 are reverse of what were observed for ligands 5 and 6. These high correlations between the hyperpolarizabilities of the ligands and their *N atom and *N-N bonds can be of significant important in the prediction of their non-linear optical properties as conductive materials. It is interesting to point out that the LUMO values of the ligands which can be of significant important in DNA binding [42, 43] as potential anticancer are found to be highly correlated with the *N atomic

15

*N-NMR shift, q(A), µbond(A), µ(A), Vol(A), χIntra_Iso(A), χIso(A),

σIso(A,A), σIso(A), Anisotropy and *N-N SD and PSO (Table 9).

Conclusion The electronic and spectroscopic properties of seven derivatives of pyrazole have been reported. The experimental spectroscopic properties are in good agreement with the theoretical methods can be of help in establishing the successful synthesis of the ligands. Some of the vibrations that are found to be strongly Raman active are C-H, N-C-N, C-H, which were found around 3050, 1450 and 1000 respectively. The *N atoms are much more shielded than the N atoms and the order of the shift in the *N atoms across the ligands is bdcpzpy < dcpz < bpzpy < bdmpzpy < bphpza < bphpzm < phpz. The highest values of FC and J-coupling for *N···*N is observed in bphpza while the lowest is observed in bdcpzpy. The values of *N-N J-coupling can give an insight

into the variation in the *N-N bond distance, bond stretch and bond strength of the ligands. Also, the interesting relationship between the non-bonding *N···*N distance and the atomic properties of the ligands can be of significant help in the characterisation and differentiation of these ligands. The *N atoms contributes significantly to the HOMO levels than the N atoms. The experimental λmax absorption is found to be dominated with H-2→LUMO, HOMO→L+1, HOMO→L+1 and HOMO→LUMO for the dcpz, bphpza, bpzpy and bdmpzpy respectively. There is evidences based on number of localized electrons and the level of deshielding of *N atoms that the carboxylic unit may result to lower affinity of the *N for metal coordination. The trend of the conductive properties of the ligands in terms of the hyperpolarizability is bdmpzpy > bpzpy > bphpza > phpz> bdcpzpy > bphpzm > dcpz. These high correlations between the hyperpolarizabilities of the ligands and their *N atom and *N-N bonds can be of significant important in the prediction of their non-linear optical properties as conductive materials. Acknowledgements The authors gratefully acknowledged the financial support of Govan Mbeki Research and Development Centre, University of Fort Hare, South Africa. The CHPC in Republic of South Africa is gracefully acknowledged for providing the computing facilities and some of the software’s that were used for the computation.

References 1. S. Trofimenko, J. Am. Chem. Soc. 89 (1967) 3165–3170.

2. S. Trofimenko, Chem. Rev., 93 (1993) 943-980. 3. L-F. Tang, W.-L. Jia, X.-M. Zhao, P. Yang, J.-T. Wang, J. Organomet. Chem. 658 (2002) 198203. 4. A. Otero, J. Fernández-Baeza, A. Antiñolo, J. Tejeda, A. Lara-Sánchez, Dalton Trans. (2004) 1499–1510. 5. C. Klein, E. Baranoff, M. Grätzel, Md.K. Nazeeruddin, Tetrahedron Let. 52 (2011) 584–587. 6. M. Zhao, Z. Yu, S. Yan, Y. Li, J. Organomet. Chem. 694 (2009) 3068-3075. 7. F. Blasberg, J.W. Bats, M. Bolte, H-W. Lerner, M. Wagner, Inorg. Chem. 49 (2010) 7435–7445. 8. G. Türkoglu, S. Tampier, F. Strinitz, F.W. Heinemann, E. Hübner, N. Burzlaff, Organomet. 31 (2012) 2166–2174. 9. I.D. Burns, A.F. Hill, A.J.P. White, D.J. Williams, J.D.E.T. Wilton-Ely, Organomet. 17 (1998) 1552–1557. 10. S. Tampier, R. Müller, A. Thorn, E. Hübner, N. Burzlaff, Inorg. Chem. 47 (2008) 9624–9641. 11. H. Kopf, C. Pietraszuk, E. Hübner, N. Burzlaff, Organomet. 25 (2006) 2533–2546. 12. H. Kopf, B. Holzberger, C. Pietraszuk, E. Hübner, N. Burzlaff, Organomet. 27 (2008) 5894– 5905. 13. N. Burzlaff, I. Hegelmann, B. Weibert, J. Organomet. Chem. 626 (2001) 16 – 23. 14. K. Hara, H. Sugihara, Y. Tachibana, A. Islam, M. Yanagida, K. Sayama, H. Arakawa, Langmuir 17 (2001) 5992-5999. 15. E. Díez-Barra, J.C. García-Martínez, J. Guerra, V. Hornillos, S. Merino, R. del Rey, R.I. Rodríguez-Curiel, J. Rodríguez-López, P. Sánchez-Verdú, J. Tejeda, J. Tolosa, ARKIVOC v (2002) 17-25. 16. C. Adamo, V. Barone, J. Chem. Phys. 110 (1999) 6158–6170. 17. A.D. Becke, J. Chem. Phys. 98 (1993) 5648-5652. 18. T. Helgaker, M. Jaszunski, M. Pecul, Prog. Nucl. Mag. Reson. Spec. 53 (2008) 249–268. 19. M.E. Casida, C. Jamorski, K.C. Casida, D.R. Salahub, J. Chem. Phys. 108 (1998) 4439-4449. 20. R. Bauernschmitt, R. Ahlrichs, Chem. Phys. Lett. 256 (1996) 454-464. 21. T. Helgaker, M. Watson, N.C. Handy, J. Chem. Phys. 113 (2000) 9402-9409. 22. V. Sychrovsky, J. Grafenstein, D. Cremer, J. Chem. Phys. 113 (2000) 3530-3547. 23. D. Sanz, R.M. Claramunt, I. Alkorta, G. Sánchez-Sanz, J. Elguero, Magn. Reson. Chem. 2012, 50, 246–255. 24. L. Pazderski, J. Tousek, J. Sitkowski, L. Kozerski, E. Szłyk, Magn. Reson. Chem. 2007, 45, 1045–1058. 25. 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, O. Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian,

Inc., Wallingford CT, 2009. 26. A.A. Granovsky, Firefly version 8.0.G, www http://classic.chem.msu.su/gran/firefly/index.html 27. T.A. Keith, AIMAll (Version 12.06.03), TK Gristmill Software, Overland Park KS, USA, 2012 (aim.tkgristmill.com) 28. N.M. O'Boyle, A.L. Tenderholt, K.M. Langner, J. Comp. Chem. 29 (2008) 839-845. 29. A.R. Allouche, Gabedit - A graphical user interface for computational chemistry softwares, J. Comp. Chem. 32 (2011) 174-182. 30. M.H. Jamróz, Vibrational Energy Distribution Analysis: VEDA 4, program,Warsaw, 2004–2010. 31. M.H. Jamróz, J.Cz. Dobrowolski, R. Brzozowski, Mol. Struc. 787 (2006) 172–183. 32. M.H. Jamróz, Spectrochimica Acta Part A: Mol. and Biomol. Spec. 114 (2013) 220–230. 33. Y. Liu, C-G. Liu, S-L. Sun, G-C. Yang, Y-Q. Qiu, Comp. and Theoret. Chem. 979 (2012) 112– 118. 34. P.S. Liyanage, R.M. de Silva, K.M.N. de Silva, J. Mol. Struc. (Theochem) 639 (2003) 195–201. 35. Jamroz, M.H. Spesca, http://smmg.pl/software/sowtware-spesca.html, Oct 2013. 36. M.H. Jamroz, J.C. Dobrowolski, R. Brzozowski, J. Mol. Struc. 787 (2006) 172–183. 37. D. Cremer, J. Grafenstein, Phys. Chem. Chem. Phys. 9 (2007) 2791–2816. 38. R. Gaur, L. Mishra, Inorg. Chem. 51 (2012) 3059−3070. 39. I. Alkorta, J. Elguero, J.E.D. Bene, Chem. Phys. Let. 489 (2010) 159–163. 40. A. Alparone, J. Fluorine Chem. 144 (2012) 94–101. 41. M. Karabacak, M. Cinar, Spectrochimica Acta Part A 86 (2012) 590–599. 42. X. Chen, F. Gao, Z.-X. Zhou, W.-Y. Yang, L.-T. Guo, L.-N. Ji, J. Inorg.Biochem. 104 (2010) 576–582. 43. L.-F. Tan, H. Chao, K.-C. Zhen, J.-J. Fei, F. Wang, Y.-F. Zhou, L.-N. Ji, Polyhedron 26 (2007) 5458–5468.

Figure 1: Optimized geometries of 3,5-dicarboxylpyrazole (dcpz), 3-phenylpyrazole (phpz), bis(3phenylpyrazol-1-yl)methane (bphpzm), bis(3-phenylpyrazol-1-yl)acetic (bphpza), bis(pyrazol-1yl)pyridine (bpzpy), bis(3,5-dimethylpyrazol-1-yl)pyridine (bdmpzpy), bis(3,5-dicarboxylpyrazol1-yl)pyridine (bdcpzpy). Dihedral angle of the four nitrogen atoms of the bispyrazol are shown in orange and some selected bond angle in blue. Bond angles N-C-N of the bphpzm and bphpza are shown at the centre.

Figure 2: The experimental IR and Raman (in red) with the theoretical IR (in black) spectral of the ligands

1. dcpz

2. phpz

3. bphpzm

4. bphpza

5. bpzpy 6. bdmpzpy

7. bdcpzpy Figure 3: The features of the QTAIM properties of the ligands showing the atomic localized electrons, Laplacian of electron density of the bonds and the ring electron density.

Figure 4: The relation of the Ramsey terms of *N-N bonds to the QTAIM properties of the same bond.

Fig ure 5: The relat ion bet wee n the ato mic pro pert ies of the ligands and their *N-N and *N···*N bond properties (the highly correlated values are in bold)

Figure 6: The experimental and theoretical UV absorption of the ligands

Figure 7: Electronic isodensity surface of six levels of HOMO and LUMO of bpzpy

Figure 8: Relating the molecular hyperpolarizabilities with the *N atomic and bond properties of the ligands

Table 1: Assignment of the prominent IR picks with weight of their contribution using the PED method dcpz Inten Freq. . Assigned vib. 3648. 54 82.42 υ(OH) 100 3540. 131.4 35 9 υ(NH) 100 1768. 305.0 92 6 υ(OC) 80

phpz Freq.

Inten.

Assigned vib.

3593.44 98.31 υ(NH) 100 3138.03 20.34 υ(CH) 75; υ(CH) 20 3128.31 27.42 υ(CH) -92

bphpzm Freq. 3129.0 9 3000.6 4 1494.6 5

Inten.

Assigned vib.

25.59 υ(CH) 88 31.33 υ(CH) 16; υ(CH) 84 41.23 υ(CC) 20

υ(CC) 15; β(HCC) 33; 1757. 84 373.2 υ(OC) 82 1450.73 28.8 β(CNN) 15; υ(CC) 47; β(HNN) 1537. 16 42.29 12; υ(NC) -12; 718.12 133.94 τ(HCCC) 80 υ(CC) -11; β(CNN) τ(CCCC) -10; τ(HCCC) 1442. 47 69.35 27; υ(NN) -15; 676.43 29.4 -42; τ(CCCC) 21; 1390. β(HNN) -11; υ(NC) 89 83.55 38 500.13 20.06 τ(HNNC) 74 1325. 153.8 35 2 υ(NC) 62 1282. υ(OC) 12; β(HOC) 98 192.1 32 1222. 175.3 β(HOC) 35; β(HOC) 05 6 -13; υ(NN) -33;

1458.9 β(HCC) 15; τ(HCNC) 2 29.35 16

υ(CC) 10; β(HOC) 33; β(HCC) -20; 1141. 69 81.89 υ(NN) 15

υ(NN) 15; β(HCC) 14; 1173.9 β(HCN) 16; τ(HCNC) 1 33.52 -16

υ(OC) 28; β(HOC) 1102. 265.5 23; β(HCC) 11; 96 9 υ(NN) 12 υ(OC) -20; β(HCC) 1071. 28 48.81 36; υ(NC) 14; υ(OC) -34; β(CNN) 991.9 7 58.75 -11; β(CNN) 38;

1396.4 7 25.73 υ(NC) -18; β(HCH) 11 1331.1 9 1323.9 8 1269.8 8 1221.4 4

β(HCN) 19; τ(HCNC) 51.15 -13

υ(NN) 34; τ(HCNC) 179.74 18 31.67 υ(NN) 29

β(HCC) -25; β(HCC) 65.93 28

1211.5 υ(CC) -13; β(HCC) 5 38.26 43

υ(CC) 10; β(HCC) 38; 1078.5 2 28.76 β(HCC) 13; 1076.9 1 27.27 υ(CC) 26; β(HCC) 37

750.4 9 44.99

1037.5 β(HCC) 13; β(HCC) 2 44.66 22 υ(NC) -15; β(CNN) 769.06 56.23 14; β(NCN) 11;

ο(OCOC) -71; 728.5 40.05 τ(HNNC) 10

735.51 60.47 υ(NN) -17; β(CNN) 24

υ(OC) 26; β(CCO) 715.8 36.14 25

τ(HCCC) 63; τ(HCCC) 723.81 72.56 -11 τ(HCCC) -12; 718.87 99.7 τ(HCCC) -70 τ(CCCC) 17; τ(HCCC) 675.37 27.06 -11; τ(HCCC) 29;

689.6 6 42.42 τ(HNNC) 75 τ(HOCC) 68; 558.0 4 84.82 τ(CNNC) 20 432.8 76.95 τ(HOCC) -90 bphpza 3616. 33 77.78 υ(OH) 100

bpzpy 1592.97 209.02 υ(CC) 70

bmpzpy 2979.1 8

43.5 υ(CH) 94

1760. 205.6 17 1 υ(OC) 86

υ(CC) 37; β(HCC) -11; 1585.64 121.33 β(CNC) 10;

2974.4 2 61.58 υ(CH) 77; υ(CH) 16

1495. υ(CC) -21; β(HCC) 18 44.25 10

1518.25 138.22 υ(CC) 65; β(HCN) 10

υ(CC) -36; υ(CC) -11; 1586.8 8 164.64 β(HCC) -11;

υ(NC) -15; υ(OC) 1360. 97 31.16 11; β(HCC) 14;

1466.34 53.07 υ(NC) -55; β(HCC) -10

1585.8 3 139.1 υ(CC) 70

υ(NC) -21; β(HCC) 35; 1319. 161.6 υ(NC) 11; β(HCC) 14 9 17; τ(HCCO) 25; 1443.46 372.02 β(NCC) -13; υ(CC) -18; υ(NC) -15; 1294. υ(NN) 25; β(HCC) 65 31.53 11 1408.63 60.09 β(HCC) 19; β(CNN) 11 υ(NC) -10; β(CCN) -30; 1218. 120.2 υ(NC) 30; β(HCC) 14 9 21 1408 55.38 β(HCC) 11; υ(NN) 16; β(HOC) 1192. 55 70.32 17; β(HCC) 16; 1398.39 120.44 υ(NC) -51; β(HCN) 21 υ(NN) 42; τ(HCCO) 1183. 41 78.99 16 1384.99 28.62 υ(NC) 22; β(HCC) -19

1562.8 9 142.13 υ(CC) -11; υ(CC) -49 υ(NC) -31; υ(CC) 10; 1472.6 9 62.89 β(HCH) 17; β(HCH) -36; β(CNN) 1448.2 9 56.14 11; τ(HCCC) -13; 1445.3 β(HCH) 13; β(HCH) 7 177.85 47 β(HCH) 46; τ(HCCN) 1434.6 1 203.98 12

1109. 186.7 υ(OC) 37; β(HOC) 1430.8 45 6 16; β(HCC) -12; 1347.29 22.45 υ(NC) -14; β(CCN) -13 5 88.06 υ(CC) -10; β(HCH) 34 υ(NC) -31; β(HCH) 1088. 1405.6 β(HCC) 12; β(NCN) 7 60.09 10 1319.78 24.08 υ(NC) 44; β(CCN) -13 9 34.13 17; β(CNN) -10; υ(CC) 15; β(HCC) υ(NN) 42; β(HCC) 20; 1375.6 1080. 2 31.23 21; β(HCC) -10; 1261.65 60.87 β(NCN) -12; 4 34.59 υ(NC) 12; β(HCH) 41 1078. 1371.0 24 43.6 β(HCC) 25 1153.82 46.74 β(HCC) 68 4 62.88 β(HCH) -80 1036. 02 59.94 β(HCC) 49 ο(OCOC) 17; 817.0 8 35.99 τ(HCCC) 23 τ(HCCC) -10; 735.4 132.7 τ(HCCC) -51; 4 6 τ(HCCC) 13; 722.1 1 113.5 τ(HCCC) 71 709.3 139.5 ο(OCOC) 32; 2 6 τ(HOCC) -10 675.5 1 44.08 τ(HCCC) 16

υ(CC) -12; υ(NN) -12; 1031.61 91.19 β(HCC) -49;

υ(NC) 24; β(HCH) 1368.1 7 83.05 13; β(HCH) 25;

υ(NN) -35; β(CCN) 24; 1350.1 υ(NC) 15; υ(NC) -12; 944.02 67.32 β(HCC) 18; 1 48.42 β(HCH) 48; 1345.3 799.06 22.34 τ(HCNN) 74 6 36.03 υ(NC) 62 τ(HCCC) -64; τ(CNCC) 1110.0 778.87 39.66 -11 1 36.86 υ(NC) 10; β(NCN) 31 τ(CNCC) 33; τ(HCCC) 762.8 65.56 β(CCC) 28; β(CNN) 18 790.01 46.22 -24 715.31 131.94 τ(HCCC) 61

760.05 35.83 τ(HCCC) 85

υ(NC) -14; β(CNN) 14; β(CCC) -13; 723.36 57.06 β(NCN) -11

ο(CCNN) 20; 647.5 6 74.3 β(OCO) 32 bdcpzpy 3647. 99 97.65 υ(OH) 100 3646. 101.6 99 5 υ(OH) 100 1784. 237.1 33 4 υ(OC) 86 1770. 184.5 78 1 υ(OC) 77 1764. 328.2 3 2 υ(OC) 76 1761. 380.4 17 6 υ(OC) 81

1423.68 295.15 υ(CC) 13; β(HCC) 20

υ(OC) -21; β(HOC) 1115.6 6 375.57 17; β(HCC) 13;

1291.72 103.15 β(HOC) 11

1098.5 υ(OC) -50; β(OCO) 2 227.46 12 1091.9 4 152.97 υ(CC) -15; β(HCC) 38

υ(OC) 10; β(HOC) 21; 1282.21 186.13 β(HOC) -11;

υ(OC) -38; β(CCN) 1041.4 221.86 11

1407.62 156.69 υ(NC) 34

1268.92 263.03 υ(CC) 14 1258.68 236.55 υ(NN) -38; β(HOC) -10

1463. 102.8 υ(NC) 13; β(HCC) 43 1 15 1247.5 90.14 υ(NN) -49 1428. υ(CC) 24; β(CNN) β(HOC) 19; β(HCC) 08 90.81 28 1130.47 117.12 17 Greek letters υ, β, τ, ο denote stretching, bending, tortion and out of plane modes respectively

Table 2: The experimental and computed 13C-NMR shifts from the direct and fitting methods

C4

C5

C3

Cpxp

Cpxm

Cpxo

Cph CH2 CHC OOH

bphp bphp dcpz phpz zm za Direc Direc Fittin Direc Fittin Direc Fittin t Fitting t g t g Exp. t g 100.3 101.9 4, 88.28 1, 89.79 , 105.0 102.1 , 104.0 100.7 6 91.86 98.83 86.83 7 88.69 4 9 90.06 109.8 124.9 1, 7, 112, 122.7 , 115.0 128.9 125.1 112.1 131.3 118.1 123.3 110.4 5 4 6 5 128.1 2 47 7 9 149.8 136.3 1, 135.9 2, 150.2 2, 132.2 3, 137.2 143.6 130.0 150.3 153.0 8 2 1 136.4 1 139 32 151.2 6 123.3 110.4 123.8 110.9 4, 3, 4, 1, 123.4 110.5 124.1 111.1 126.3 124.0 111.0 6 4 3 9 92 2 8 124.4 111.5 124.5 111.5 124.5 111.5 8, 3, 1649125.1 112.1 125.4 112.4 128.5 125.3 112.3 3 5 5 6 82 7 8 121.2 108.3 121.2 108.4 121.4 108.6 1, 8, 734121.9 109.0 122.8 109.9 126.2 122.7 109.8 1 5 2 3 1 2 3 131.1 130.9 117.7 , 117.9 4, 5, , 131.8 131.1 117.9 132.0 118.8 131.8 5 1 2 118.6 67 8 8 68.32 57.44

153.8 139.7 6COO 156.5 142.4 H 7 3

76.65 65.47 74.01 5 (74.7 0 165.9 151.4 [13]) 5 6

bpzp bdm bdcp y pzpy zpy Direc Fittin Direc Fittin Direc Fittin Exp. t g Exp. t g t g 105.3 109.0 1, 93.06 6, 96.67 , 109.7 105.7 , 108.2 105.3 109.8 7 1 93.07 91 1 93.45 4 97.43 122.8 109.9 131.7 118.5 , 1, 5, 2, 125.1 122.8 109.9 138.9 136.6 123.2 136.2 122.8 88 1 2 91 4 4 3 5 126.0 141.6 128.1 8, 9, , 139.6 , 126.0 140.7 148.6 140.7 143.0 129.4 91 139.6 8 43 5 134.8 5 1

127.1 58 135.4 106.7 2, 110.8 106.7 98 4 151.3 7, 142.9 151.3 26 8

122.0 124.6 133.2 136.2 4 83 8 120 9 122.9 111.8 148.0 134.2 , 99.31 7, 4, 94.42 , 113.8 111.7 , 148.6 134.7 94.44 79 9 99.31 2 7 137.4 112.4 2, 1, 99.9, 137.4 150.6 151.6 137.7 116.9 104.3 3 71 8 2 9 1

152.9 138.9 22157.3 143.1 6 9

13.72 4, 15.11 153.2 14.89 - 6.2CH3 19 2 15.49 6.57 The subscript “px” represents the phenyl (ph) in bphpza and pyridine (pyr) in bpzpy and bdmpzpy. The value in bracket is from the cited reference for similar molecule.

Table 3: The experimental and computed 1H-NMR shifts from the direct and fitting methods

C4-H C5-H C3-H Cpx-pH Cpx-mH Cpx-oH CH2 CHC OO COO H

bphp bphp bpzp bdm bdcp dcpz phpz zm za y pzpy zpy Direc Fittin Direc Fittin Direc Fittin Direc Fittin Direc Fittin Direc Fittin Direc Fittin t g t g t g Exp. t g Exp. t g Exp. t g t g 6.71, 7.03, 6.8, 7.12, 7.11, 7.42, 6.91 7.23 6.84 7.16 6.88 7.2 7.174 6.84 7.15 6.497 6.59 6.92 6.04 6.22 6.55 7.2 7.51 7, 7.31, 7.69, 7.98, 7.48 7.78 7.68 7.97 7.79 8.43 8.7 7.952 7.91 8.19 8.561 7.92 8.2 7.31, 7.61, 7.32, 7.62, 7.3 7.6 7.33 7.63 7.369 7.32 7.63 7.376 7.75 8.04 7.461 7.75 8.04 7.98 8.26 7.38, 7.69, 7.4- 7.77.2, 7.51, 7.4- 7.77.41 7.71 7.48 7.78 7.429 7.5 7.8 7.292 7.11 7.41 7.309 6.99 7.3 8.38 8.65 7.72, 8.01, 7.69- 7.98- 7.52, 7.67- 7.968.64 8.91 8.63 8.9 7.85 8.67 8.93 7.681 7.647 5.83, 6.18, 6.49 6.82 7.354 7.27 7.57

6.16, 6.5, 6.54 6.87

6.25- 6.596.81 7.12

6.452 6.56 6.89 2.317 2.07- 2.532.673 2.51 2.96

CH3 N-H 10.74 10.94 9.32 9.56 The subscript “px” represents the phenyl (ph) in bphpza and pyridine (pyr) in bpzpy and bdmpzpy. Also, the superscript “c” and “f” on Cph-ortho mean close to and far from the nitrogen that is the coordination centre of the pyrazole.

Table 4: The computed 15N-NMR shifts from the direct and fitting methods dcpz phpz bphpzm bphpza Direc t Fitting Direct Fitting Direct Fitting Direct Fitting

N

173.5 - 193.8 - 171.49, 2 194.15 7 213.40 -178.81

- -79.52, *N 45.78 -73.31 -88.00 113.25 -84.46

Npy

bpzpy Direct

bdmpzpy

Fitting Direct

bdcpzpy

Fitting Direct Fitting 169.5 192.23 8, , - -171.53, -192.27, -154.15, 175.82, - 147.55, 175.0 199.15 -173.01 -193.66 -154.16 -175.83 -161.87 183.13 -153.33 4 105.22 , - -78.61, -104.36, -67.32, -93.68, -32.73, 60.96, 109.90 -79.20 -104.92 -67.33 -93.70 -70.52 -96.71 -65.32 -91.80 133.6 -106.43 -130.68 -84.25 109.70 -109.55 3

Table 5: The four Ramsey terms of the N-N bonds in the ligands

FC

SD

N-*N

-3.947

*N-N

-4.535

dcpz PSO 0.000 phpz -0.053

DSO

J

-1.564

0.025

-5.486

-1.175

0.024

-5.739

bphpzm N-*N *N-*N

-4.832, -4.499 -0.044, -0.051 -1.177, -1.205 0.014 0.000 -0.001

N-*N

-5.00, -4.940 -0.056, -0.060 -1.139, -1.128

0.031, 0.030 -6.022, -5.725 0.000 0.014

bphpza *N-*N

0.025

N-*N

-5.712

0.032, 0.032 -6.163, -6.096

0.000

-0.001

0.004

0.027

bpzpy 0.005

-1.167

0.029

-6.844

0.016, 0.016 -0.002

0.008, 0.008 0.004

0.558, 0.559 0.008

0.031 0.008 0.004

-6.566 0.481 0.005

Npy-*N *N-*N

0.530, 0.531 0.006

0.004, 0.004 0.000

N-*N Npy-*N *N-*N

-5.548 0.478 0.003

-0.015 -0.005 0.000

bdmpzpy

N-*N

-1.034 -0.001 -0.002

bdcpzpy -4.517, -4.608 0.020, -0.008 -1.526, -1.316

Npy-*N *N-*N

0.350, 0.497 0.007, -0.013 -0.001 0.000

0.030, 0.008 0.000

0.033, 0.033 -5.990, -5.900 -0.003, 0.010 -0.001

0.384, 0.502 -0.002

Table 6: The relationship of the computed Ramsey terms of *N-N bonds and their bond properties (the highly correlated values are shown in bold values). FC

J

Bond length

Bond stretch

∇2(ρ)

1.000

0.977

-0.945

0.808

-0.959

SD PSO

0.008 -0.780

-0.185 -0.631

-0.287 0.925

0.460 -0.848

-0.230 0.886

DSO J

-0.435 0.977

-0.487 1.000

0.422 -0.860

-0.600 0.722

0.500 -0.891

Bond length Bond stretch

-0.945 0.808

-0.860 0.722

1.000 -0.898

-0.898 1.000

0.994 -0.913

0.943

0.857

-1.000

0.907

-0.995

-0.959

-0.891

0.994

-0.913

1.000

FC

ρ ∇2(ρ) ε

0.886

0.927

-0.724

0.460

-0.736

V

-0.928

-0.832

0.998

-0.908

0.989

G

0.893

0.779

-0.986

0.879

-0.965

desqVen

0.137

0.048

-0.147

-0.119

-0.063

0.959

|V/G|

0.955

-0.913

0.863

-0.949

Table 7: Correlation of the *N-N and *N-*N bond properties in relation to the atomic properties of *N atom *N-N bond distance FC

*N-*N non-bond distance Bond Bond length stretch

J

∇2(ρ)

distance

FC

J

15

*N-NMR shift

-0.037

-0.197

-0.153

0.062

-0.083

0.961

-0.723

-0.768

0.714

0.555

-0.880

0.865

-0.843

0.872

-0.575

-0.643

-0.914

-0.838

0.909

-0.678

0.883

-0.325

-0.067

0.043

-0.869

-0.748

0.964

-0.895

0.942

-0.732

0.378

0.469

-0.490

-0.300

0.706

-0.711

0.647

-0.929

0.677

0.728

-0.528

-0.376

0.635

-0.481

0.562

-0.727

0.354

0.439

0.531

0.360

-0.684

0.601

-0.617

0.844

-0.523

-0.598

-0.136

0.006

0.417

-0.510

0.396

-0.271

0.301

0.276

0.507

0.376

-0.571

0.453

-0.508

0.738

-0.413

-0.495

-0.352

-0.326

0.453

-0.731

0.501

-0.431

0.744

0.719

-0.401

-0.213

0.609

-0.624

0.548

-0.961

0.722

0.767

0.758

0.656

-0.893

0.948

-0.904

0.807

-0.574

-0.617

-0.386

-0.198

0.594

-0.608

0.531

-0.961

0.723

0.768

Anisotropy

0.625

0.469

-0.840

0.880

-0.813

0.768

-0.608

-0.655

Occupation

0.091

0.041

-0.198

0.499

-0.214

0.573

-0.849

-0.852

-0.377

-0.286

0.508

-0.714

0.506

-0.714

0.857

0.893

q(A) µintra(A) µbond(A) µintra(A) Vol(A)

χIntra_Iso(A) χbond_Iso(A) χIso(A) χzz(A)

σIso(A,A) σIso(A',A) σIso(A)

Energy

Table 8: The non-linear conductive properties of the ligands

dcpz phpz

Nonß in esu Valence (1x10Gap in Lewis Dipole 30) HOMO LUMO KJ/Mol Orbitals ∆α1 ∆α2 ∆α3 Γ 3.55 92.52 -3.31 75.16 2819.12 -49.97 1.84 -0.3015 -0.0870 563.27 2.32% 2.49 124.37 -90.01 102.87 1240.26 -65.22 3.40 -0.2250 -0.0341 500.97 2.79%

bphpzm bphpza

3.21 2.35

273.88 296.55

-9.61 -48.38

172.79 14881.42 -8.31 133.87 7790.76 -11.62

2.71 -0.2240 -0.0436 4.19 -0.2273 -0.0569

473.61 447.56

2.73% 2.84%

bpzpy

5.16

179.84

-78.73

143.88

7251.88

-6.22

7.91 -0.2412 -0.0564

485.39

2.76%

bdmpzpy bdcpzpy

5.19 4.66

233.64 -78.06 267.46 -135.11

141.62 175.81

6981.98 -9.33 6327.24 -13.55

10.05 -0.2306 -0.0513 2.99 -0.2717 -0.0976

470.74 457.20

2.27% 2.72%

Table 9: The correlation of the molecular properties with the properties of *N atom

FC SD PSO DSO J 15 *N-NMR shift Bond length Bond stretch ∇2 (ρ) ε VG q(A) µintra(A)

G -0.398 -0.173 0.324 0.942 -0.389

ß esu (1x1030) HOMO -0.681 -0.888 0.219 0.162 0.320 0.665 0.896 0.253 -0.740 -0.870

LUMO -0.633 -0.705 0.920 0.233 -0.478

Band Gap KJ/Mol 0.573 -0.450 0.211 -0.776 0.867 -0.502 -0.229 -0.828 0.546 -0.267

0.338 0.478 -0.724 0.540 -0.072 -0.627 -0.367

0.623 0.597 -0.702 0.666 -0.553 -0.829 -0.309

0.061 0.831 -0.547 0.820 -0.928 -0.800 -0.544

-0.562 0.817 -0.840 0.794 -0.246 -0.611 -0.951

-0.760 0.639 -0.579 0.575 -0.171 -0.309 -0.898

-0.052 -0.650 0.790 -0.704 0.228 0.749 0.515

0.086

0.347

0.873

0.699

0.687

-0.336

0.386

0.454

0.710

0.903

0.808

-0.560

0.192 -0.220

0.041 -0.117

0.343 0.484

0.909 0.698

0.962 0.919

-0.332 0.029

0.050

0.057

-0.433

-0.814

-0.965

0.135

0.630

0.203

0.118

0.601

0.388

-0.608

0.265

0.128

-0.410

-0.644

-0.873

-0.066

0.915

0.673

0.136

0.431

0.025

-0.832

0.084

-0.076

0.235

0.874

0.976

-0.238

-0.761

-0.651

-0.540

-0.890

-0.602

0.856

0.064

-0.096

0.223

0.863

0.976

-0.217

-0.586

-0.395

-0.503

-0.896

-0.726

0.674

µbond(A) µ(A) Vol(A),0.001 χIntra_Iso(A)

χbond_Iso(A) χIso(A) χzz(A) σ Iso(A,A) σ Iso(A',A) σ Iso(A) Anisotropy

Highlight 1. The synthesis of pyrazole derivatives four were carried out 2. The strong correlations of the experimental spectroscopic properties with the theoretical 3. The experimental λmax are HOMO or HOMO-1 to LUMO or LUMO+1 excitation using TDDFT. 4. The properties *N atoms correlate with the molecular properties of the ligands.

Graphical abstract