Introduction

Quinolones or fluoroquinolones constitute a large class of synthetic antimicrobial agents that are highly effective in the treatment of many types of infectious diseases, particularly those caused by bacteria. They are widely used to treat human and veterinary diseases (Currie et al., 1998; Ihrke et al., 1999; Chen et al., 1999). Due to the fluorine atom at position C-6 and the piperazine or methyl piperazine at position C-7, these antibiotics exhibit a broad spectrum of activity against Gram-positive and Gram-negative bacteria. Over time, bacteria become resistant to medicines that are used to combat them; because of this, the medical world is always in search of new and improved ways to battle these disease-causing bacteria. New antibiotics are continually being developed and quinolones are at the forefront of this research (Wolfson and Hooper, 1989; Carlucci, 1998).

Several analytical methods for quantitative determination of quinolones in pharmaceutical formulations are described in the literature, including capillary electrophoresis (Flurer, 1997; Sun and Chen, 1997; Bhowal and Das, 1991) and UV spectrophotometry (Venugopal and Saha, 2005). Spectrophotometric analysis of gatifloxacin was done using the latter method, but different buffers were used; however, in our case no buffer was used, so the method has an advantage over the previous one. In a method reported by Fratini and Schapoval (1996), the Lambert-Beer law was obeyed in the concentration range of 20–100 μg mL−1, and also nitric was used, but in the newly developed method the linearity range is much broader and, also, the limit of quantification is much less than in this previous work, titrimetry (British Pharmacopoeia, 1999; Belal et al., 1999), and high-performance liquid chromatography (HPLC) (Samanidou et al., 2003; Joshi, 2002; Sanzgiri et al., 1994). Some authors prepared derivatives of different quinolones and compared their properties against those of quinolone (Shaharyar et al., 2007; Gopalakrishnan et al., 2007; Jayashree et al., 2009); and the in vitro availability of atorvastatin, in the presence of ciprofloxacin, gatifloxacin, and ofloxacin has been reported (Arayne et al., 2009).

The objective of this research was to develop and validate rapid, economical, and sensitive methods for quantitative determination of six quinolones—ciprofloxacin (Fig. 1), gatifloxacin (Fig. 2), norfloxacin (Fig. 3), levofloxacin/ofloxacin (Fig. 4), and pefloxacin (Fig. 5)—in bulk and tablet formulations using different salts of iron. The major advantage of the proposed methods is that these six flourquinolones can be determined on a single system with minor modifications in detection wavelength. The proposed methods were applied successfully to determination of the six quinolones in both reference material and dosage forms with high values of accuracy and precision. No interference was observed in the assay from common excipients at levels found in pharmaceutical formulations. These methods were validated by the statistical data. In addition, the association constant, stochiometric ratio of reactants, and standard free energy changes (ΔG°) were determined.

Fig. 1
figure 1

Ciprofloxacin

Full size image
Fig. 2
figure 2

Gatifloxacin

Full size image
Fig. 3
figure 3

Norfloxacin

Full size image
Fig. 4
figure 4

Levofloxacin/ofloxacin

Full size image
Fig. 5
figure 5

Pefloxacin

Full size image

Experimental

Instrumentation

A double-beam UV–vis spectrophotometer (Shimadzu model 1601) equipped with 10-mm quartz cells was used to make absorbance measurements and Shimadzu UVPC version 3.9 software was used to control the instrument, data acquisition, and data analysis. Spectra of quinolone–iron complexes were recorded over the wavelength range 360–800 nm.

Chemicals

All chemicals used were of analytical grade; demineralized double-distilled water was used throughout the study. Ciprofloxacin, gatifloxacin, norfloxacin, levofloxacin, ofloxacin, and pefloxacin were kind gifts from local pharmaceutical companies. Pharmaceutical formulations were purchased from the market and ferric chloride, ferric nitrate, and iron ammonium citrate were from Merck, Germany.

Preparation of standard solutions

Reference stock solutions of ciprofloxacin, gatifloxacin, norfloxacin, levofloxacin, and pefloxacin were prepared in water at a concentration of 500 μg mL−1, whereas ofloxacin at the same concentration was prepared in methanol. These stock solutions were diluted to obtain the desired concentration ranges for different quinolones (5–200 μg mL−1 for ciprofloxacin, 10–250 μg mL−1 for gatifloxacin, 6–300 μg mL−1 for norfloxacin, 10–200 μg mL−1 for levofloxacin, ofloxacin, and pefloxacin).

Preparation from pharmaceutical formulations

Twenty tablets of each formulation were weighed and powdered. A powdered tablet equivalent to 50 mg of active substance was transferred to a 100-ml volumetric flask and diluted up to the mark with the same solvent as mentioned for the standard preparation. These solutions were stirred on a magnetic stirrer for 60 min, filtered, and further diluted to obtain the desired concentration ranges. All solutions were stored at 4°C. One percent solutions of ferric chloride, ferric nitrate, and iron ammonium citrate were prepared in double-distilled water.

Quinolone complexes with ferric chloride

To prepare 10–200 μg mL−1 ciprofloxacin, gatifloxacin, ofloxacin, and pefloxacin, 10–160 μg mL−1 levofloxacin, and 6–300 μg mL−1 norfloxacin, different aliquot portions of reference standard solutions of each drug were transferred into separate series of 25-ml volumetric flasks. In each flask, 3 ml of 1% ferric chloride solution was successively added. The volume was made up to the mark with water and set aside at room temperature for 10 min. The absorbance of quinolone–iron complexes was measured against a reagent blank at 375, 473, 442, 375, 375, and 434 nm for ciprofloxacin, gatifloxacin, norfloxacin, levofloxacin, ofloxacin, and pefloxacin, respectively. The calibration graph was prepared by plotting absorbance versus concentration of drugs (Table 1).

Table 1 Linear regression functions and their statistical parameters
Full size table

Quinolone complexes with ferric nitrate and iron ammonium citrate

Assays were completed for the formation of complexes of quinolones with ferric nitrate (using 3 ml of a 1% solution) and iron ammonium citrate (using 1 ml of a 1% solution) as mentioned for ferric chloride.

Results and discussion

Iron(III) salts have been shown to be ideal for the derivatization of carboxylic groups (Franch et al., 2004; Arayne et al., 2008), which makes them a suitable reagent for detection and quantification of quinolones via their carboxylic group. Fe3+ holds three molecules of quinolone, resulting in the complex having a brown color which absorbs radiation in the visible range. In this article, the development and validation of sensitive and precise spectrophotometric methods for determination of six quinolones, using this derivatization technique, have been described.

Optimization of derivatization conditions

The optimum reaction conditions for quantitative determination of all quinolones were established via a number of preliminary experiments. The concentration of iron(III) was optimized by using 1–5 ml of a 1% iron solution. Steady and maximum color development of the complex was achieved with a volume of 3 ml of ferric chloride and ferric nitrate, but in the case of iron ammonium citrate the best results were observed when 1 ml of salt solution was used. Hence 3 ml of 1% ferric chloride and ferric nitrate solutions and 1 ml of 1% iron ammonium citrate solution were used as the optimal concentrations for validation of the method.

Calibration curves

Calibration curves were prepared by linear least squares regression analysis plotting of the absorbance of quinolone–iron complexes versus the concentration of quinolone (6–300 μg mL−1) (Table 1).

Reaction of quinolone with ferric chloride

An intense brown color developed in the visible region, showing minor bands at 360 and 440 nm for all quinolones except gatifloxacin, which exhibited absorption at 445 nm (Fig. 6) when the solutions of quinolones were mixed with ferric chloride solution individually. These bands were attributed to the formation of a quinolone–iron complex, in which three quinolone rings bind to iron(III).

Fig. 6
figure 6

UV spectra of quinolone/FeCl3 complexes: (a) ciprofloxacin; (b) gatifloxacin; (c) levofloxacin; (d) norfloxacin; (e) ofloxacin; (f) pefloxacin

Full size image

Reaction of quinolone with ferric nitrate and iron ammonium citrate

An intense brown color developed in the visible region at 360 and 440 nm (Figs. 7 and 8) when solutions of different quinolones were mixed with ferric nitrate and iron ammonium citrate. These bands have been attributed to the formation of a quinolone–iron complex, which is formed by three quinolone rings and iron.

Fig. 7
figure 7

UV spectra of quinolone/ferric ammonium citrate complexes: (a) ciprofloxacin; (b) gatifloxacin; (c) levofloxacin; (d) norfloxacin; (e) ofloxacin; (f) pefloxacin

Full size image
Fig. 8
figure 8

UV spectra of quinolone/FeNO3 complexes: (a) ciprofloxacin; (b) gatifloxacin; (c) levofloxacin; (d) norfloxacin; (e) ofloxacin; (f) pefloxacin

Full size image

These visible spectrophotometric methods, using aqueous solutions of iron(III) ions as reagents, have an elegant simplicity; a brownish-green complex was formed in the proportion 3:1 [quinolone:iron(III)]. The optimized methods were validated for quinolone–iron complexes in pharmaceutical formulations. Results were of adequate precision and accuracy. The absorption spectra obtained revealed that all the quinolones showed almost the same behavior, except for gatifloxacin. In reaction with ferric chloride, all quinolones show the same curve except for gatifloxacin (Fig. 6); two bands were found in all quinolones except gatifloxacin, one at 360 nm, whereas gatifloxacin showed maxiumum absorbance at 440 nm. A similar trend was observed in the case of iron ammonium citrate and ferric nitrate (Figs. 7 and 8).

Association constants and standard free energy changes

The absorbance of the complex was used to calculate the association constant using the Benesi–Hildebrand (Benesi and Hildebrand, 1949) equation:

$$ C_{\text{a}} /A = \left( { 1/\varepsilon } \right) + \left( { 1/K_{{{\text{c}} \times \varepsilon }} } \right) \times \left( { 1/C_{\text{b}} } \right) $$

where C a and C b are the concentrations of the acceptor and donor, respectively, A is the absorbance of the complex, ε is the molar absorptivity of the complex, and K c is the association constant of the complex. The calculated association constants are reported in Table 2. The low values of K c are common in these complexes due to dissociation of the complex to the radical anion.

Table 2 Association constants and standard free energy changes
Full size table

The standard free energy changes of complexation (ΔG°) were calculated from the K c values by the following equation (Martin et al., 1969):

$$ \Updelta {{G}}^{\circ} = - 2. 30 3 {\text{RT log}}K_{\text{c}} $$

where ΔG° is the free energy change of the complex (kJ mol−1), R the gas constant (1.987 cal mol−1 deg−1), T the temperature in Kelvin (273 + °C), and K c the association constant of quinolone–iron complexes (l mol−1).

Validation of methods

Linearity, limits of detection and quantification, and stability

Linearity of the assay was demonstrated by at least six concentrations over the range 6–300 μg mL−1 for six quinolones. Absorbances were plotted against concentrations and analyzed using least squares linear regression (Table 1). According to ICH recommendation the detection and quantification limits of the methods were calculated using the standard deviation of the response and the slope of calibration curve as reported in Table 1.

Precision and accuracy were assessed in conjunction with the linearity studies using three spiked samples of three concentrations of each quinolone. Measured concentrations were determined by application of the appropriate standard curve obtained on each occasion. Precision was assessed in terms of percentage RSD values. Percentage recovery values were used to express accuracy (Table 3).

Table 3 Accuracy and precision of proposed method
Full size table

Sensitivity and interference study

The percentage recovery values of quinolones confirm the high sensitivity of the proposed methods. The excellent recoveries indicate the absence of interference from frequently encountered excipients. Percentage RSD values were ≤3.89 (Table 3). Under the same experimental conditions different excipients at different concentrations were added and analyzed. Potential interference problems from the commonly used excipients and other additives such as microcrystalline cellulose, lactose, povidone, starch, and magnesium stearate were examined, and it was confirmed that the excipients did not interfere with the assay. Low percentage RSD values signify good precision of the method.

Application in pharmaceutical formulations

The proposed methods were successfully applied to the analysis of quinolones in commercial formulations. The results were in good agreement with the declared contents, and no interference was observed in the assay of all quinolones from common excipients at levels found in pharmaceutical formulations. These methods rely on the use of simple and inexpensive chemicals and techniques but have a sensitivity analogous to that obtained by sophisticated and expensive techniques such as HPLC and are validated by statistical data. The reaction conditions and application of the methods for determination of quinolones in pharmaceutical formulations have been established.

Conclusion

It is rare that ferric salts are used as chromogenic reagents for spectrophotometric determination of these quinolones. The proposed methods, which are simple and rapid, offer the advantages of sensitivity over a wide range of concentrations without the need for extraction or heating. The methods do not entail any stringent reaction conditions and have been successfully applied to the determination of quinolones in pharmaceutical formulations.