Polyhedron Vol. 8, No. 5, pp. 647451, 1989 Printed in Great Britain

0277-5387189 $3.00 + .Dc Q 1989 Pergamon Press plc

CUPRIC ION BINDING BY DIHYDROXYBENZOIC ACIDS

TAMAS KISS

Department of Inorganic and Analytical Chemistry, Kossuth University, 4010 Debrecen, Hungary

and

HENRYK KOZLOWSKI

Institute of Chemistry, University of Wroclaw, Joliot-Curie 14, 50-383, Wroclaw, Poland

and

GIOVANNI MICERA” and LILIANA STRINNA ERRE

Dipartimento di Chimica, Universita di Sassari, Via Vienna 2,071OO Sassari, Italy

(Received 18 August 1988 ; accepted 29 September 1988)

Abstract-A potentiometric and spectroscopic study has been carried out on the copper(I1) complexes formed by 2,x-dihydroxybenzoic acids (X = 4, 5 or 6) in aqueous solution. It has been found that, in the low pH range, the ligands coordinate through the carboxylate group. At above pH 5 the major species are chelated complexes in which one or two ligands bind the metal ion through the carboxylate and adjacent phenolate groups. The results are compared to earlier literature data available for copper(I1) dihydroxybenzoate and salicylate complexes.

The modelling of the interactions of metal ions with soil organic matter, as well as the study on the bioavailability to plants of metal ions, have started much interest in the elucidation of the behaviour of phenolic ligands. ’ Dihydroxybenzoic acids (DHB) and their analogues were found to be interesting low molecular weight models with very efficient binding ability, especially the ligands having two aromatic hydroxyls in ortho positions.2-6 In the copper(II)- 3,4-dihydroxybenzoic acid system, rather com- plicated coordination equilibria were observed. 6 In dilute aqueous solutions the bidentate catechol part of this ligand predominates as the binding site for the metal ion, although the carboxyl group may also participate in the metal ion binding. This leads to the formation of the polynuclear species.

Earlier solid-state studies have shown that at low pH, the carboxylic group could be the main donor in the formed complexes with several other di- phenolic acids such as 2,4- 2,5- and 2,6-dihydroxy-

*Author to whom correspondence should he addressed.

benzoic acids,7-‘0 though the phenolate oxygen binding was also clearly indicated.2g3*6 In this communication, we report the potentiometric and spectroscopic findings obtained for the systems containing copper(I1) and 2,4-, 2,5-, and 2,6- dihydroxybenzoic acids in aqueous solutions.

EXPERIMENTAL

Potentiometric studies

Stability constants for complexes containing pro- ton and copper(I1) were determined by pH-metric titration of 25 cm3 samples. The concentration of the ligand in the samples was 0.004 or 0.002 M. Metal-to-ligand molar ratios of 1 : 1, 1 : 2, 1 : 4, 1 : 6.6 and 1: 10 were used in the titrations and the

ianic strength was adjusted to 0.2 M with KCl. The titrations were performed in the pH range from 2.5 to 11, or until precipitate, depending on the molar ratio used, occurred. The pH was measured with a Radiometer pHM 64 instrument with G2040B glass

647

648 T. KISS et al.

and K4040 calomel electrodes. Since the ligands tend to undergo oxidation, all measurements were performed in a TTA 80 titration unit under an argon atmosphere. The electrode system was cali- brated by the method of Irving et al. l1 so the pH- meter readings could be converted into hydrogen ion concentrations. In all cases, the temperature was 25°C. The calculations of the stability constants were made with PSEQUAD computer program. ’ 2

Spectroscopic studies

Absorption spectra were recorded on a Beckman Acta M IV spectrophotometer and EPR spectra were measured on a Varian E 9 spectrometer (X band) at 110 K. The measurements were performed under argon for the concentrations of samples simi- lar to those used in the potentiometric studies.

RESULTS AND DISCUSSION

Proton complexes

All ligands studied have two measurable pro- tonation constants (see Table 1). The acidity of the phenolic groups in position two is very weak, thus their dissociation was studied at high ligand con- centration (0.08 M) up to high pH (N 13.4). It was found that the ligand titration curves run together with the titration curve for the strong acid of the same concentration. Accordingly, the phenolic groups do not undergo dissociation in the measur- able pH range ; this is distinctly different from that reported earlier. 2 The high dissociation constants of these hydroxyl groups were proved also by spectrophotometric titration. The UV spectrum of the ligands remained unchanged over the pH range 11-13.2, showing the lack of deprotonation of phenolic hydroxyls. The very weak acidity of these OH groups in the ortho position to carboxylate, can be explained by the strong intramolecular hydrogen bond between COO- and phenolic groups.

Copper(I1) complexes

The calculated copper(I1) complex formation constants, together with the earlier literature data, are given in Table 2. These three ligands behave as the hydroxy-salicylic acids (derived equilibrium constants for salicylic acid calculated from litera- ture dataI are also included in Table 2). The data for the 2,4- and 2,5-DHB are in fairly good agree- ment with those presented earlier.’ The fitting of the experimental pH-metric data by the PSEQUAD program can be slightly improved by the intro- duction of the species in which the metal ion binds the DHB molecule only via the carboxylate oxygen, i.e. when the formation of the CuAH or Cu(AH), species is assumed. The species distribution vs pH, see Figs l-3, indicates that the total concentration of each of these species is rather low (10% or less), except in the case of the copper(IIk2,6_DHB system. In the latter case the concentrations of the CuAH and Cu(AH), complexes approach 25 and 15%, respectively. It has to be mentioned, however, that the pH-metric determinability of these species is poor, since metal ion coordination and proton dissociation take place in the same pH range, hence the copper(II)-carboxylate coordination has only minor pH effects. The EPR spectra prove un- ambiguously, the formation of these minor species at a pH below 5 in all the cases studied (Table 3). The formation of the complexes with monodentate carboxylate coordination was also proposed earlier for the cupric complexes with salicylic acid. ’ 4

The major species formed at above pH 5 is the CuA chelated complex in which the metal ion is bound to the DHB ligand through COO and the phenolate oxygen (O-) being in the ortho position to the carboxylate (2-OH). A further increase of pH leads to the formation of the other chelated species, the CuA, complex, in which two ligands are bound in the mode described above. The equilibrium con- stants characteristic to the part-processes of com- plex formation (see Table 2) clearly indicate the salicylate-type coordination in the complexes CuA and CuA,. Thus, their stoichiometric compositions are more precisely Cu(HAH_ J and Cu(HAH_ J2.

Table 1. Proton dissociation constants for 2,x-DHB (x = 4, 5 or 6)

Ligand pK (COOH) pK (x-OH) pK (2-OH) Reference

2,4-DHB 3.11 f0.02 8.68 +0.02 > 14.0 This work 3.12 8.62 13.03 2

2,5-DHB 2.73 _+ 0.02 10.05 f 0.02 > 14.0 This work 2,6-DHB 1.OkO.l 13.10*0.14 > 14.0 This work

0.91 12.57 13.0 2

Cupric ion binding by dihydroxybenzoic acids 649

Table 2. Stability constants for copper(I1) complexes with 2,x-DHB (x = 4, 5 or 6)

Species 2+DHB 2,5-DHB 2,6-DHB SaP

Log fi values CuAH

Cu(AH) z CuA

C& Ct&H_, CuA,H_,

Log K values

PK (Ctw PK (CuA,H- 1) Cu+HA+CuA+H+

CuA+HA F? CuA,+H’

log [K (CuA)IK (C&)1

10.31 f0.07 11.52f0.05 21.46+0.09 23.97 + 0.05

5.97 fO.O1 7.18kO.01 9.8OkO.04 11.65kO.03 0.98 +0.07 1.26f0.07

-8.41 f0.07 - 9.67 f 0.09

8.82 9.39

-2.72 - 3.076 -4.88 -5.OOb

2.16 1.93b

10.39 10.93

-2.87

- 5.58 -

2.71 -

15.20+0.08 - 30.45 +0.09 - 10.19+0.01 10.6 18.47kO.03 18.5 6.45f0.17 -

12.02 - - -

-2.91 -2.8 - 2.97’ -4.82 -5.5 -4.776 -

1.91 2.7 1 .79b -

n Sal = salicylic acid (ref. 13). b Ref. 2.

In the more basic solutions, the deprotonation of probably due to the strong electron-releasing effect the unbound hydroxyl at positions four, five or six of the coordinated phenolate. This value for the takes place. The stepwise deprotonation constants copper(2,6-DHB system was too high to be (Table 2) obtained for the 2,4- and 2,SDHB evaluated reliably. The differences between the two complexes are higher than the pK values of these stepwise pK’s obtained for the CuAz complexes is groups obtained for the respective free ligands, equal to 0.6 log units, corresponding to the stat-

Table 3. Spectroscopic data for copper(I1) complexes of 2,x-DHB (x = 4, 5 or 6)

Species ESR Absorption Donor set A,, (10e4 cm-‘) 1 (nm) [s]

2,4-DHB CuAH CllA

2,5-DHB CuAH CuA

CuA, 2.299 172

2,6-DHB CuAH cu

2.371 155 2.327 168

2.304 176

2.371 155 2.327 166

2.378 147 2.333 163

2.299 172

735” [40] 38@ [130] 650” [102]

720“ [63] 42@ [2071 620“ [MO] 425” [352]

740” [48] 41ob [184] 645” [l lo]

coo- {COO-, o-}

2 x {COO-, o-}

coo- {Coo-, o-}

2 x {COO-, o-}

coo- (COO-, o-}

2 x {Coo-, o-}

“d-d transition. ’ Phenolate oxygen-to-copper(I1) charge-transfer transition.

650 T. KISS et al.

4 5 6 7 8 PH

9 10

Fig. 1. Concentration distribution curves of the complexes formed in the copper(I2,4-DHB system as a function of pH. Cc, = 0.002 M and C&d = 0.004 M.

PH

Fig. 2. Con~ntration dist~bution curves of the complexes formed in the cop~r(II~2,5-DHB system as a function of pH. Details as in Fig. I.

Fig. 3. Concentration distribution curves of the complexes formed in the copper(IIk2,6-DHB system as a function of pH. Details as in Fig. 1.

Cupric ion binding by dihydroxybenzoic acids 651

istical case of the dissociations of two protonated groups with the same acidity.

The spectroscopic data fit very well with the potentiometric results and the assignment of the respective complexes is presented in Table 3. The coordination of the phenolate oxygen is clearly sieen, d-e 10 tie charge Irantitjon between 1% donor and the cupric ion, which is observed in the 38M30 nm range. 15,16 The coordination of the phenol& hydroxyl groups is also clearly indicated by the distinct increase ia energy of the &a tratr-

sitions observed for the respective complexes (see for example, ref. 6).

CONCLUSIONS

The binding ability of the dihydroxybenzoic acids strongly depends on the position of their donors. The most effective seems to be the set of two hydroxyls which are in ortho positions (catechol like) as was Found in the case of 3,tdihydroxy- benzoic acid.6 The ligands with carboxyl and hydroxyl groups placed on the adjacent aromatic ring carbons are also very effective chelating Iigands which Form complexes with the (COO-, O-) donor set. In all cases, the involvement of the carbctxylic ffunction seems to be basic at lowpI rawq.. _but phenolic groups and their positions are of primary importance for the structures and the stability of the formed complexes.

Acknowledgements-Thanks are due to Mrs A. GBnczy ffor herparticipation in the ewerimental work. This work was supported by Consiglio Nazionale delle Ricerche (Rome), the Polish Academy of Sciences (Project 01.12) a& by t& K=X>X%XX %&iii%. & !&&ii&&m $%X&X% 46/1986).

1.

2.

3.

4.

5. 6.

7.

8.

9.

10.

II.

12.

13.

14.

15.

16.

REFERENCES

G. Hurst and N. A. Burges in Soil Biochemistry (Edited by A. D. McLaren and G. H. Peterson), Vol. I, Chapt. 5, p. 260. Marcel Dekker, New York (1967). M. Aplincourt, A. Debras-Bee, C. Gerard and R. P. Hugel, J. Chem. Res. 1986, 134. C. Gerard, R. Njomnang, J.-C. Pjetrard, J. Rimbault and R. Huge], J. Cbem. Res. 1987,294. F. Cariati, S. Deiana, L. Erre, G. Micera and P. Piu, Inorg. Chim. Acta 1982, 64, L213. P. W. Linder and A. Voye’, Polyhed*n 1987, 6, 53. K. Gerega, T. Kiss, H. Kozlowski, G. Micera, L. Strinna Erre and F. Cariati, Znorg. Chim. Acta 1987, 138,31. F. Cariati, L. Erre, G. Micera, A. Panzanelli, G. Ciani and A. Sironi, Inorg. Chim. Acta 1983,80, 57. G. Micera, L. Strinna Erre, F. Cariati, D. A. Clemente, A. Marzotto and M. Biagini Cingi, Znorg. Chim. Acta 1985,109, 173. G. Micera, L. Strinna Erre, P. Piu, F. Cariati, G. Ciani and A. Sironi, Inorg. Chim. Acta 1985, 107, 223. G. Micera, L. Strinna Erre, F. Cariati, G. Ciani and A. Sironi, Znorg. Chim. Acta 1985, 100, Ll. H. Irving, M. G. Miles and L. D. Pet&, Analytica Chim. Acta 1967, 38,475. L. Zekany and I. Nagypal, in ComputationalMethods for the Deferminafion of Sfabiiity Constanfs (Edited

s, x1.Q &W&>! I%?mnr.%%%~ L%yT XJ2*,r1,9p, A. E. Martell and R. M. Smith, Critical Stability Constants, Vol. 5. Plenum Press, New York (1982). C.-H. O’Young and S. J. Lippard, J. Am. Chem. Sot. 1980,102,4920. A. Gergely and T. Kiss, Inorg. Chim. Acta 1976, 16, 51.

H. Kozlowski, M. Bezer, L. D. Pettit, M. Bataille and B. Hecquet, J. Inorg. Biochem. 1983,18,231 and -i&%+&i&%X