Charge Transfer Spectrophotometric Determination of Metronidazole in Pharmaceutical Formulations by Normal and Reverse Flow Injection Analysis Coupled with Solid-Phase Reactor Containing Immobilized FePO4

Two rapid, simpleand sensitive flow injection methods were developed for the estimation of metronidazole (MRZ) in pharmaceutical formulations. The proposed methods were based on charge transfer reaction between metol (N-methyl-paminophenol sulfate) as a π-acceptorand reduced MRZ as an n-donor to produce a blue colored chargetransfer complex. Method A depends on the reaction of reduced MRZ with metol (MT) in the presence of NaIO4 using two lines manifold to form blue colored product exhibiting absorption maxima at 700 nm.While method B depends on charge transfer reaction of reduced MRZ with MT in presence of a solid phase reactorcontainingfixedFePO4 on cellulose acetateusing reverse flow injection manifold to form a blue colored productwhich was measured spectrophotometrically at690 nm.Various experimental parameters for both methods were studied. Beer's law was obeyed in the ranges of2.5-200 and 2.5-150 μg mL -1 ,with r 2 of 0.9995 and 0.9972;whilethe detection limit values were2.53 and 2.12μg mL -1 for methods A and B, respectively. Both of the suggested methods were successfully applied for the estimation of MRZ in commercial formulations. The results of the developed methods were compared with those obtained by the British pharmacopeia method, showinghigh accuracy and precision.

The present work describes a sensitive and simple two flow injection spectrophotometric methods for the determination of MRZ in its pure form along withpharmaceuticals formulations. Method Adepends on the chargetransfer complexation of MRZ molecule with MTto form the colored product in the presence of NaIO 4 using two lines flow injection manifold,whilethe resulting colored was measured at 700 nm.Method B employs one line reverse flow injection manifold coupled with solid phase reactor containing immobilized FePO 4 and the response was measured at 690 nm.

Reduction of nitro group and preparation of the standard solution
Pure MRZ powder (100 mg)was dissolved in 20 mL of methanol. 10 mL of 5 N Hydrochloric acid (37 % w/w, BDH, England) was added to the methanolic solution of MRZ and 0.5 g of zinc powder (BDH, England) was addedat room temperature. The solution was filtered using a Whatman filter paper (No 41) after standing for 20 minutes to remove the insoluble matter,then the volume was made up to 100 mL with methanol.

Procedure of pharmaceutical forms (Tablets)
An equivalent to 100 mg of MRZ was weighed, powderedto ten tablets of MRZ,and then dissolved in 20 mL of methanol. The resulting filtrate was treated as described above for theestimation of MRZ.

Metol (MT) reagent solution (0.1)
Metol reagent solution was prepared by mixing0.861 g of MTreagent (MT, Merck, Chemicals Ltd., Germany, M.Wt.344.38 g mol -1 ) in distilled water and the solution was made up to 50 ml and stored in a dark flask. Sodium periodate solution (0.1 M) 2.1391 g ofNaIO 4 (BHD, England,M. wt. 213.91 g mol-1, purity99%) was dissolved in 5 mL distilled water and the volume was completed to 100 mL in a volumetric flask.

Preparation of solid -phase reactor containing immobilized FePO 4 (F-SPR)
In light of previous studies [20], a new method for the preparation of immobilized FePO 4 was successfully used to prepare a solid phase reactor which could be used in many oxidation methods, such as spectrophotometric flow injection analyses. The immobilizingsteps for the preparation of F-SPR were carried out by dissolving 0.5 g of cellulose acetate (CA) in 5 mL of acetone and 0.5 mL of dimethylformamide with continuous stirring.Then, 1.5 g of FePO 4 was added aftermanual homogenization by stirringuntil homogenous mixture viscosity was increased. A few minutes later,distilled waterwas usedfor washing and rigid oxidizing material was formed. Different sizes(0.15 -1.18 M) wereobtained by crushingof the driedfixed FePO 4 into the desired particle size which was selected by sieving on sieves with known mesh sizes (Retsch GmbH & Co.KG, Germany). Finally, the FePO 4 particles werepackedinto different lengths of glass tubes (2 mm i.d.)for the preparation of F-SPR.To hold the packed particles in place, small sponge pieces were inserted at the ends of the tubes.

Results and Discussion Absorption spectra
A spectrophotometric method for the determination of MRZ has been reported [21] by charge transfer complex of MRZ with the MT to give a blue colored product in the presence of an oxidizing reagent. The reported reaction was the base for developing nFIA (method A) and rFIA (method B).
Before applying the proposed reaction by means of nFLA or rFIA,the blank and colored product absorption spectra were measuredmanually to obtain the bestparameters for the colored productformation. The test wascarried out in 10 mL flask containing 40 μg mL -1 of reduced MRZ, 0.5 mL of NaIO 4 (0.03M) and 0.5 mL (0.01 M) of MT for method A. While method B was based on the use of 30 μg mL -1 of reduced MRZ, 1 mL (0.01 M) of MT and 0.06 g of immobilized FePO 4 particles (1.18 mm). As soon as the solutions were mixed and swirled, the blue colored product was formed. The flasks were made up to the volume with distilled water and then filtered. Depending on the optimum conditions, the absorbance spectrum of the colored product versus reagent blank was recorded between 250 and 1000 nm. The  max values werefound to be 700 and 690 nm) for methods A and B, respectively(Figure-3, which will be used in all subsequent experiments. The proposed reaction was used for developing normal flow injection method (A) and a reverse flow injection method (B),coupled with one line packed F-SPR containingfixed FePO 4 . The reaction mechanism may be suggested and established by depending on the previously reported mechanismwherethe nitro compound is first reduced to the corresponding amino derivative.The reaction mechanism is dependenton the reaction of MRZ amino group ( n-donor) with the oxidized metol ( π-acceptor) to form charge transfer complex, which subsequently forms a blue colored product that was measured at 700 and 690 nm for methods A and B, respectively. On the basis of the literature survey, tentative reaction mechanisms for MRZ and MT complexes in the presence of NaIO 4 or FePO 4 are proposed and given in schemes 1 [21][22][23]. It can be seen that the charge transfer complex was formed in the ratio of 1:1 (Drug: Reagent).

Optimization of the experimental conditions
The effectof different parameters (physical and chemical)was studied for both methods (A and B). The optimizationconditions were carried out by changing one parameter and keeping all the others constant. Table-1summarizesthe preliminary conditions for both suggested methods.    6 , NaOI 4 , Cr +6 , K 2 S 2 O 8 andKIO 3 ]) were optimizedin order to select the most appropriate oxidizing agent for method A. The maximum responsewas obtainedby the use of NaOI 4 ( Figure-4-a).Therefore, itwill be used in next studies for method A.
The effect of various NaOI 4 concentrations (0.001 to 0.07 M)was examined. It was found that the response was increased with increasingNaOI 4 concentration up to 0.03M. However, any level beyond this concentration (0.03 M) led to the reductionof theresponse ( Figure-4-b). Therefore, the NaOI 4 (0.03M) was selected in the next studies for the estimation of MRZ.

Optimization of solid-phase reactor (F-SPR) conditions for method B Effects of solid-phase reactor composition
The proportion of FePO 4 fixedin CA has asignificant role in the activity of the F-SPR. Various weight ratios of fixedoxidizer in CA were used for the developing of the F-SPR materials; 0.25:1, 0.5:1, 0.5:1. 5, 0.5:2 and 1: 2 (CA: FePO 4 , w: w, g). It was found that the ratio of 0.5:1.5 g provided the reproducibility and highest response for F-SPR (Figure-5). Thus,it will be used in next studies for method B.

Effect of solid-phase particles size
Various particles sizesof fixed FePO 4 wereinvestigated (0.15 -1.18 mm). It can be seen in Figure-6 that the responseraiseswith increasing the particles size up to 1 mm; therefore,1 mm particle size was selected and used in the next studies for method B.

Effect of solid-phase reactor length
The effect of reactor length (F-SPR) on the response was optimized by changing the length of the reactor in the range of 4-12 cm. It was found that theemploymentof10 cm reactor length gave thehighest response, as presented in Figure-7. By comparing the stability of the response, the length of 10 cm was selected and used in the next studies for method B.

Effect of solid-phase particles weight (degree of packing)
The effect of particles weight of the F-SPR (0.04-0.1 g) was studied using variousweights of the fixed FePO 4 on CA. It can be seen (Figure-8) that the weight of 0.091 g gave the highest response. Therefore, 0.091 g, as an optimum degree of packing (particle weight), was selected and used in the next studies for method B. The reagent concentration was varied in the range (0.001-0.005)% in order to maximize the peak height.

Optimization of chemical and physical conditions for both methods The effect of MT concentration
In order to maximize the absorbance of charge transfer product, the reagent (MT) concentration was examined for both methods in a range of 0.005-0.025 M. It can be seen that the response washeightened as the MT concentration was increased up to 0.01 and 0.015 M for methods A and B, respectively ( Figure-9). Therefore, 0.01 and 0.015 M were selected as the best concentrations for methods A and B, respectively. The effect of total flow rate The effect of total flow rate on the response of the colored product was also examined for methods A and B in the range of 1.2 to 3.6 and 0.6 to 2 mL min -1 , respectively (Figure-10). When the flow rate was increased, the signal was heightened up to 3 and 1.9 mL min -1 for methods A and B, respectively. Therefore, the flow rates of 3 and 1.9 mL min -1 were selected as optimum flow rates for methods A and B, respectively, which will be used in the next studies.

Effect of injection sample volume
The injected volume (75 to 200 μL) into the carrier stream was evaluated since it has an important rolein the response value. It can be seen that 150 μL as an injected volume gave the best response for both methods A and B (Figure-11). Therefore, were selected this volume for the next studies.

Effect of reaction coil length
The effect of mixing coil length on the response was optimized in range of 0 (without reaction coil) to 100 cm. According to the results (Figure-12), lengths of 75 and 50 cm were chosen as optimum lengths that gave the maximum absorbance for the colored product for methods A and B, respectively,andwill be used in next studies.

Selected Optimum Conditions
The optimum values of all investigated parameters are summarized in Table-2 for using the proposed methods (A and B) for the determination of MRZ.

Sampling frequency for both methods and F-SPR life-time
Depending on the optimum parameters, the sampling frequency was evaluated by recording the time from the sample injection to the maximum absorbance (27 and 37 seconds for A and B, respectively). 110 and 74samples hr -1 were achieved as practical sampling frequency for methods A and B, respectively. To examine the efficiency of the F-SPR (method B) containing immobilized FePO 4 on the CA,the experiment was performed with injection of MT (150 μL) into the MRZ stream at a flow-rate of 1.9 mL.min -1 and then thepassagethrough F-SPR. The results indicated that 34 injections with RSD% of 4.37 could be achieved with good reproducibility (RSD ≤ 5) [24] as well as life time for F-SPR.

Calibration graph
The response of the colored product was recorded and plotted against the concentration of MRZ (Figure-13), depending on the selected parameters mentioned in Table-2. Two series of MRZ solutions were prepared in the range of 2.5-200 and 2.5-150 μg mL -1 . The detection limit was 2.53 and 2.12 μg mL -1 for methods A and B, respectively. Table-3 summarizesthe other analytical values of statistical treatments for the calibration graph [25].

Sd; standard deviation Accuracy and precision
The precision and accuracy forboth methods were evaluated by injections of pure drug solution at two various concentrations. Table-4   Under the recommended procedure, the standard addition method (Table-5) was applied by preparing a series of solutions for each sample (50 and 100 μg mL -1 ) via transferring the required volume ( 0.25 or 0,5 mL, of 1000 μg mL -1 ) of commercial dosage to five volumetric flasks (10 mL), followed by the addition of various volumes ( 0 , 0.1, 0.2, 0.3 and 0.4 mL) of thereduced MRZ (1000 μg mL -1 ). The results were mathematically treated for standard additions method and the results were summarized in Table-5.In order to examine the success and the efficiency of both methods (A and B) for the estimation of MRZ in pure and commercial tablets, the results were compared statistically with the standard method [26]. The suggested and standard methods obtained results whichwere compared statistically for theestimation of MRZ pharmaceutical formulations at 95% confidence level by means of the F-test and t-test. It was found (Table-6) that there is no significant difference between theproposed and standardmethods, while the F-test and t-test did not exceed the theoretical values.  .75 *S p = pooled standard deviation Theoretical values at 95% confidence limit, n1=3, n2 = 2. **t = 2.77, where t has degrees of freedom = (n1 + n2 -2) = 3 ***F = 19.0, where F has degrees of freedom = (n1 -1) = 2, (n2 -1) = 1

Conclusions
The present study describes the successful evaluation of immobilized FePO 4 on cellulose acetate as the oxidizing agent and MT π acceptors as an analytical reagentfor the development of normal and reverse flow injection methods for the accurate estimation of MRZin pharmaceutical dosage forms. The rFIA (method B) coupled with anSPR containing immobilized FePO 4 gives many advantages;it ishighlysensitive, simple, rapid, and does not need expensive sophisticated apparatus. The results obtained showed that the reproducibility of F-SPR (RSD % ≤ 5) as well as life timewere good, in addition to havingthe capacity for loading a desirable number of reagent injections (34injections). The proposed methods used inexpensive reagents with excellent shelf life and are available in any analytical laboratory.