Chemicals were purchased from Merck. Glycine, sodium tartrate, L-tryptophan, copper(II) nitrate trihydrated, sodium nitrate, potassium hydrogen phthalate and standard solutions of sodium hydroxide (titrasol), 2,2’- bipyridyl, 1,10-phenanthroline, nitric acid, EDTA and buffer solutions of pH 4.0, 7.0 and 9.0 were from Merck. All the starting materials were pro analysis and used without further purification. Water was purified by Mili-Q water purification system, deionized and distillated.

Reagents: Carbonate-free sodium hydroxide 0.03 M was preparated and standardized against sodium hydrogen phthalate and a standard solution of nitric acid 0.5 mM. Copper (II) nitrate solution (0.03 M) was prepared by dissolving the above substance in water and was standardized with standard solution of EDTA 0.1 M (triplex).

All pH titrations was performed using a Metrohm 794 basic automatic titrator (Titrino), coupled with a Hero thermostating bath at 25˚C (±0.1˚C) and a Metrohm combined glass electrode (Ag/AgCl). The pH meter was calibrated with Merck standard buffer solutions (4.0, 7.0 and 9.0).

For the determination of acid dissociation constants of the ligand L an aqueous solution (0.3 mM) of the protonated ligand was titrated with 0.03 M NaOH at 25˚C under nitrogen atmosphere and ionic strength of 0.1 M, NaNO₃. For the determination of binary (one ligand and Cu²⁺) and ternary systems (Cu²⁺, one of the other L ligand (Har) and L), the ratios used were 1:1:1, Cu(II) : L: Har, 0.3 mM. This solution was titrated with 0.03 M NaOH under the same conditions mentioned above. Each titration was repeated seven times in order to check the reproducibility of the data.

The acid dissociation constants, and for H₂(L) were calculated by an algebraic method. The equilibria involved in the formation of 1:1 complex of L and a M (Cu²⁺, Cu(Bpy)²⁺, and Cu(Phen)²⁺) may be expressed as equations (3) & (4).

If we abbreviate for simplicity Cu²⁺, Cu(Bpy)²⁺, and Cu(Phen)²⁺ with M²⁺, one may write the following two equilibria (3) & (4):

The experimental data of the potentiometric pH titrations may be completely by considering the above mentioned equilibria (1) through (4), if the evaluation is not carried into the pH range where hyrdoxo complex formation occurs.

The stability of ternary complexes may be evaluated by the following equilibrium:

where M is the metal ion, H is the proton, A and B are the ligands. The global stability constants for the ternary complexes may be represented as following:

It is possible to define the stability constants for ternary complexes in relation to their binary ones [9], represented by the equilibrium (6) & (7).

Differences between the stability constants of the ternary and binary complexes show the tendency of the formation of ternary species [10]. This could be expected by Equation (8):

The difference between the constant refined from experimental data and those calculated statistically using Equation (8) indicates the possibility of ligand-ligand interaction.

The model of species for these ternary systems that was used in superquad program includes all the species of table 1 as well as the hydrolysis of Cu²⁺ [11,12]. The stability constants of the binary complexes were refined separately using the titration data of this system in a 1:1 and 1:2 ligand: Cu²⁺ ratio in the same conditions of temperature and ionic strength. They were fixed and, consequently, only ternary species were refined in ternary model of the species. The results are summarized in Table 1. The order of the resulted stability constants are Cu²⁺ < Cu(Bpy)²⁺ < Cu(Phen)²⁺. Figure 2 shows schematic structures of the species with interactions according to equilibrium (4) & (7) for Cu(Phen)(Trp) and Cu(Phen)(Ttr). The results of the acidity constants show good agreement with reported values [13]. The reported stability constant of Cu(L) complex is similar to our results (table 1). The difference between stability constants according equation (8) show that mixed ligand complexes [14-17] formed by a divalent 3d ion, a heteroaromatic N base and an O donor ligand possess increased stability. Now one can calculate the free energy DG, used DlogK received from equation 8 (table 1). We receive for Dlog 5.44 kJ/mol and for Dlog 7.34 kJ/mol, which are considerable high. This means that interaction between Cu(Har)²⁺ and L^2- is relative strong and the observed increased stability indicate strong complex bilding of ternary systems.

It has to be further emphasized that the basicity of the carboxylate group in aqueous solution is very low and consequently this also applies for the coordinating properties of this group.

Comparison of the stability constants for the Cu(Bpy)(L) and Cu(Phen)(L) complexes in table 1 with the corresponding values for Cu(L) indicates an increased stability of the mixed-ligand species. As it is well known for a number of Cu(Har)(L) complexes that an increased complex stability is connected with the formation of intramolecular stack between the aromatic ring systems of 2,2’-Bipyridyl and 1,10-phenanthroline and the heteroaromatic ring of L (opened form « closed

^*The given errors are three times the standard error of the mean value or the sum of the propabable systematic errors, ^aaccording equation (4), ^baccording to equation (8).

form) [10]. The difference between the determined stability constants indicate the experimentally ligand-ligand stack interaction in the Cu(Har)(L) complexes.

As we can see from the experimentally results from Table 1, there is no increased stability constants in case of Cu(Har)(Gly), this means that there is no indication of intramolecular stack interactions. For this reason we can use the stability constants of Cu(Har)(Gly) as opened form in our next calculations.

By employing equation (8) the following definition is possible (equation (9)):

It is evident that the coordination sphere of Cu²⁺ ions on both sides of this equilibrium are identical, consequently the value for DDlogK is a true reflection of the extent of the intramolecular hydrophobic or stacking interaction in Cu(Har)(L) complexes. The corresponding results are listed in the fourth column of table 2.

Now we can define the intramolecular and thus dimensionless equilibrium constant K_I is than given by equation (10) for opened and closed form:

The observed increased complex stability is linked to K_I by equation (11):

Knowledge of K_I allows calculation of percentage of the macrochelated form according to equation (12) [10]:

The results of the calculations of above mentioned equations are summarized in table 2.

Comparison of the percentage of the macrochelated form according to equation (12) in table 2 shows the high stacking tendency of Trp based on heteroaromatic structure of indole moiety [5].

The distinguishing structural characteristic of tryptophan is that it contains an indole functional group. It is an essential amino acid as demonstrated by its growth effects on rats. Now it is interesting to investigate the complex building of ternary systems with Trp. The comparison of stability constants of these ternary complexes show that Cu(Har)(Gly) exists in open form but Cu(Har)(Trp) is found near 100% in closed form (see last column in table 2). The differences between the stability constants are based on stacked form of Cu(Har)(Trp). The last provides increased stability. The results described in this study show that Trp is a very versatile ligand. Due to the dominating conformation in aqueous solution, hardly any macrochelates are formed in Cu(Har) (Trp) complexes. The energy differences between closed and open form in Cu(Har)(Trp) are significant. One can calculate the free energy DG for Cu(Har)(Trp). So we receive respectively values for Cu(Bpy)(Trp) and Cu(Phen)(Trp) 11.66 kJ/mol and 12.62 kJ/mol. The according structure of ternary Cu(Phen)(Trp) is shown in figure 2. Due to the fact that the resulting data is very interesting, that affects the ternary complexes of Trp in biological systems as active. This might be used, for example in the case of cell separation.

The stability constant of the binary complex was refined separately in the same conditions of temperature and ionic strength. It was in good agreement with reported value [9-14]. Figure 2 shows schematic structures of the species with interactions according to equilibrium (4) & (7) for Cu(Phen)(Trp).

^*The given errors are three times the standard error of the mean value or the sum of the propabable systematic errors. ^afrom table 1, ^baccording equation (8), ^caccording equation (9), ^daccording equation (11), ^eaccording equation (12).

As we can use from the results in Table 1, the corresponding value for Dlog K_{Cu(Phen)(Ttr)} = 0.27 ± 0.13 is relatively high compared with Dlog K_{Cu(Phen)(Gly)} = –0.94 ± 0.11. The reason for this increased value of Dlog K_{Cu(Phen)(Ttr)} is the involved functional site, such as two carboxyl groups of Ttr in Cu(Phen)(Ttr) complex, compared with one carboxyland one amine groups of Gly in Cu(Phen)(Gly). It is important for the coordination capability of Cu(Phen)(L) complexes, whether the entering ligands Oor N-groups offer. The structure of the ternary complex Cu(Phen)(Ttr) has been shown in figure 2.