Determination of acid dissociation constants of some substituted salicylideneanilines by spectroscopy. Application of the Hammett relation

1 Introduction

A large number of Schiff bases have emerged as an important class of antioxidant compounds; o-N-salicylideneanilines have received much attention in a wide variety of fields due to their different applications owing to their characteristic properties such as biological activities, pharmacological and catalytic properties (Thorat et al., 2012; Montielli et al., 1988; Dash et al., 1984; Casaszar et al., 1985; Köseoglu et al., 1995). The synthesis of these compounds has been well developed since 1960s, and many improved methods have been reported (Cohen and Schmidt, 1962; Hadjoudis et al., 1977; Higelin and Sixl, 1983; Kownacki et al., 1993; Johmoto et al., 2009). In addition, it is also important for us to understand the molecular and chemical properties of such Schiff bases. For instance, these are of crucial importance in understanding the distribution, receptor binding and mechanism of action of certain organic preparations (Yoke-Leng et al., 2013; Kaufman et al., 1975; Fleuren et al., 1979; Shamsipur et al., 1993; Almasifar et al., 1997), as well as in many analytical procedures, such as acid–base titration, solvent extraction and complex formation (Jiménez-Lozano et al., 2002; Popovic et al., 2000; Falcomer et al., 2006; Gasowska, 2005; Evagelou et al., 2003).

Acid dissociation constant (pK_a value) is an important parameter to estimate the extent of ionization of molecules at different pH values. It is worth mentioning that there have been a number of studies regarding this parameter (Yurdakoç and Özcan, 1993; Akay et al., 2002; Kiliç et al., 1998; Gündüz et al., 1993; Ögretir et al., 2009), in order to produce the effect of several electron donating and withdrawing substituents; and also the effect of solvent on the acidity of a number of Schiff bases (Köseoglu et al., 1995; Asuero et al., 1986; Yurdakoç and Öscan, 1990).

The present work therefore reports on the determination of the acid dissociation constants of salicylideneaniline (Schiff base taken as the standard), salicylidene-2-methylaniline, salicylidene-3-methylaniline, salicylidene-4-methylaniline, salicylidene-2-chloroaniline, salicylidene-3-chloroaniline, salicylidene-4-chloroaniline, salicylidene-2-nitroaniline, salicylidene-3-nitroaniline and salicylidene-4-nitroaniline. Various methodologies have been used for the experimental determination of acid dissociation constants (Poole et al., 2004; Lachenwitzer et al., 2002; Rouhani et al., 1995; Moghimi et al., 2003; Albert and Serjeant, 1984). Among these techniques potentiometry and spectrophotometry are the preferred methods due to their simplicity and ease of application. For this work the UV–Vis spectroscopic method has been employed because of its high sensitivity.

Moreover, the purpose of this research was to use the Hammett equation to predict the effect of substituent on the acidity of salicylideneaniline, which has been a relationship between the effect of meta and para substituents on either rates of reactions or equilibrium constants.

2 Materials and methods

All Schiff bases were prepared by condensing salicylaldehyde with aniline and substituted anilines in the 2-, 3- and 4-position by methyl, chloro and nitro groups in ethanol on a water bath for 2–5 h (Issa et al., 2005; Saw, 1967). The Schiff bases were purified and filtered off by recrystallization from ethanol. Salicylaldehyde, aniline and substituted anilines were purchased from Merck Millipore and were used as received. Stock solutions of our compounds were prepared in double distilled water.

The concentration of stock solutions of perchloric acid (BIOCHEM) and sodium hydroxide (BIOCHEM) was 0.1 M and was prepared as described elsewhere (Gündüz et al., 1993). The ionic strength was maintained constant by means of sodium perchlorate (NaClO₄) as 0.1 M. All test solutions were freshly prepared before taking the measurement.

3 Experimental

Titration was performed at a constant temperature and in a nitrogen atmosphere with sodium hydroxide, in concentration of 0.02 M, in 50 ml solutions containing 0.1 M NaClO₄ with: (i) 2.10⁻⁴ M HClO₄ (for cell calibration) plus (ii) 2.10⁻⁴ M HClO₄ + 2.10⁻⁴ M Schiff bases.

Simple titration spectra (absorbance (A) vs. wavelength) resulting from spectrophotometric titration of Schiff bases at different pH values are given in Figs. 1–3.

Close

Figure 1 UV absorption spectra of salicylidene-4-chloroaniline (2.10–4 M) in 60% ethanol–water at different PH values.

Export to PPT

Close

Figure 2 UV absorption spectra of salicylidene-3-methylaniline (2.10⁻⁴ M) in 60% dioxane–water at different pH values.

Export to PPT

Close

Figure 3 UV absorption spectra of salicylidene-2-nitroaniline (2.10⁻⁴ M) in 60% ethanol–water at different pH values.

Export to PPT

The spectra of all the Schiff bases studied except the spectrum of salicylidene-2-nitroaniline that follows two isosbestic points, show that the deprotonation of nitrogen atom and naphtholic hydrogen atom is completely dissociable in separate steps (Figs. 1 and 2). But for the nitro group in the ortho position the absorption spectrum does not present significant changes throughout the titration, which could be due to the presence of the keto form (Fig. 3).

pK_a values of substances were calculated from this method and pH measurement data according to the procedure defined in the literature (Asuero et al., 1986; Yurdakoç and Öscan, 1990) based on the following equations (Akay et al., 2002; Polster and Lanchmann, 1989):

4 Results

The o-hydroxy Schiff bases, where a highly stable hydrogen (H)-bridged, quasi six-membered ring exists due to intramolecular H-bonding between the H-atom of the hydroxyl group and the N-atom of the imine linkage (Polster and Lanchmann, 1989; Blagus et al., 2010; Krygowski et al., 2008; Filarowski, 2005; Dominiak et al., 2003). The advantage of these systems is that they show two tautomer forms, enol-imine (OH) and keto-enamine (NH) forms. These tautomers are awaited to be in equilibrium with each other, but it is the (OH) that is commonly found (Scheme 1).

Close

Scheme 1 Structure of the Schiff bases in which R: H, CH₃, Cl₂ and NO₂ in three different positions.

Export to PPT

Considering the structures of the Schiff bases, we can deduce that they have two protonation sites; the first one is the imino-nitrogen atom and the second one is OH group of the phenol ring. The log K₁ and log K₂ values are related to the dissociation acid of imine nitrogen and phenolate respectively, as follows:

5 Discussion

The acid dissociation constants given in Table 1 are considered in more detail in order to gain more information about the specific effect of the substituent and the effect of solvent on the basicity of the Schiff bases.

We note that the values of pK₁ for protonated Schiff bases bearing the substituent R from para to ortho, follow an order of increasing acidity.

The interpretation of classification could be obtained taking into account the electronic effect of the substituent on the one hand and the spatial configuration of our legends on the other hand. Our compounds tend to adopt a configuration in space of two aromatic rings are hardly coplanar. This geometry results in two types of resonance in the molecule, the first taking place between phenol and the double bond of the imine, the second relating to the lone pair of the nitrogen and the phenyl part of the aniline.

The order of acidity of salicylidèneaniline derivatives compared to the reference molecule SA (salicylideneaniline) is as follows:

The methyl group with an inductive effect donor leads to pK₁ whose values are relative to the salicylideneaniline following descending order of acidity: $$2-\text{CH}3<\text{SA}<3-\text{CH}3<4-\text{CH}3$$ We note that the salicylidene-2-methylaniline has a pK₁ lower than in the meta and para positions.if R is a nitro group the order of the acidity is as follows: $$3-\text{NO}2<4-\text{NO}2<\text{SA}$$ Similarly, we also note that the salicylidene-3-nitroaniline has a pK₁ lower than in the para position.

As in the case of the methyl group and nitro, order of acidity for the chloro substituent group is as follows: $$2-\text{Cl}<3-\text{Cl}<4-\text{Cl}<\text{SA}$$ For this substituent it is observed that the acidity is high when the substituent is in the ortho position.In addition to the inductive effect of donor or attractor for the three substituents, another effect can be added to better interpret these results, which is the character of the electron-withdrawing or electron mesomeric substituent.

Indeed, the values of pK₁ Schiff bases substituted in the ortho, meta and para-nitro in both systems dioxane–water and ethanol–water are quite similar and are significantly lower than those substituted by chloro and methyl groups. This becomes obvious if one considers only the character of the nitro group mesomeric electron which strongly attracts the electronic charge at the nitrogen which generates high acidity that facilitates the deprotonation.

We note that, during the transition from ortho to para of the chloro pK₁ values increased significantly compared with nitro. This is probably due to the inductive effect attractor that strongly attracts the electron density at the nitrogen when the substituent is in the ortho position, that is to say, a high acidity.

The methyl group with an inductive effect donor leads to a value of pK₁ lower when the substituent is in the ortho position. If we consider the hypothesis that greater electron density at the nitrogen, that is to say, the deprotonation is difficult (low acidity), this disagrees with the result. Another effect that can add to interpret these results is probably the steric effect of the methyl group.

As to the variation of the pK₂ regarding the values of these compounds, we have not observed any order between these values and the type or position of the substituents. This lack of regularity can probably be attributed to the fact that the substituent was far from the OH group (Köseoglu et al., 1995; Gündüz et al., 1993).

This study is also concerned with the effect of solvent on the acid dissociation constants. This effect has been studied by several authors (Kwan-Kit et al., 1974; Gündüz et al., 1986).

We note that the variation of the acid dissociation constants with the nature of solvent follows the same trend for other anilinesalicylidenes differently substituted. We can say that the effect of solvent on the protonation constants remains the same for all the hydroxy Schiff bases studied.

It is observed that the values of pK_a follow the following order: $$\text{p}{K}_1(\text{dioxane})<\text{p}{K}_1(\text{ethanol})$$ $$\text{p}{K}_2(\text{ethanol})<\text{p}{K}_2(\text{dioxane})$$ A difference between the values of pK₁ equal to 0.41 is observed when going from ethanol to dioxane. The difference between the values of pK₁ is due to the difference in polarity of the two solvents.

For values of pK₂ a difference of 0.38 is obtained. This can be explained, in addition to the difference in polarity of the two solvents, by establishing hydrogen bonds between the solvent and the phenolic OH group.

By using the data listed in Table1 Hammett relations, (σ ∼ ΔpK₁) of substituted salicylideneaniline in 60% ethanol–water and 60% dioxane–water mixtures are given in Figs. 4 and 5 respectively.

Close

Figure 4 Plot of the ΔpK_a of substituted salicylideneaniline against the Hammett substituted constants (σ) in ethanol solvent.

Export to PPT

Close

Figure 5 Plot of the ΔpK_a of substituted salicylideneaniline against the Hammett substituted constants (σ) in dioxane solvent.

Export to PPT

As shown in Figs. 4 and 5, excellent linear correlations govern the influence of the substituent R on the acid dissociation constant of the azomethine nitrogen.

For both solvents, the values of ρ are negative and lie between −1.86 for water–dioxane and −2.02 for the system water–ethanol which means that the reaction is sensitive to the effect of the substituent electron.

We note that the value of ρ in the system water–dioxane is greater than that determined in the system water–ethanol. This difference is due to the solvent polarity. This implies that as the solvent polarity decreases, the role of the substituent increases and ρ values increase, and this is consistent with Tokura’s results (Tokura et al., 1969).

6 Conclusion

In summary, we have discussed and revised to understanding of the acid–base behavior of N-salicylideneaniline (SA) and some substituted salicylideneaniline in ethanol and dioxane–water binary mixtures. The determined pK_a values are appropriate for predicting the effect of different substituents in different positions.

Also, the important data extracted from this exploration are the constant of reaction (ρ) which can be used for other substituent in order to determine approximately their acid dissociation constants, without making calculations in the same conditions. The Hammett equation represents the effect of substituent on the reactivity of azomethine nitrogen of Schiff bases in ethanol and dioxane solvents.