Analyst, January 1996, Vol. 121 (89-92) 89 Sensitive Determination of Nitrite Using its Catalytic Effect on the Bromate Oxidation of Prochlorperazine Ashraf A. Mohamed", Mohamed F. El-shahat", Tsutomu Fukasawab and Masaaki Iwatsukib a Department of Chemistry, Faculty of Science, Ain Shams University, Abbassia, Cairo, Egypt b Department of Applied Chemistry and Biotechnology, Faculty of Engineering, Yamanashi University, Takeda-4, Kofu, 400 Japan A highly sensitive and selective catalytic method is described for the determination of trace amounts of nitrite based on its effect on the oxidation of prochlorperazine (PCP) with bromate. The reaction rate was monitored spectrophotometrically by following the formation of the red oxidation product of PCP at 525 nm, within 1 min of mixing the reagents.The optimum reaction conditions were 40 pmol l-1 PCP, 0.20 moll-' H3P04 and 10 mmol l-1 bromate at 25 OC. A 1 mmol l-l EDTA concentration and 2 pmol 1-1 Hgrl were used as effective masking agents for vanadium and iodide ions, respectively. By using the recommended procedure, the calibration graph was linear up to 70 ng ml-l of nitrite; the detection limit was 0.8 ng ml-1. The method was applied to the determination of nitrite in rain and polluted river waters. Keywords: Kinetic method; nitrite determination; prochlorperazine-bromate redox reaction; natural water Introduction Nitrite is a characteristic pollutant the determination of which at ng ml-1 levels is important in environmental studies. Ion- chromatographic, 1-3 potentiometri~~.~ and spectropho- tometric6-8 methods have been reported for nitrite determina- tion with detection limits of 5-20, 200-400 and 10-1000 ng ml-1, respectively.Of the sensitive spectrophotometric methods, the modified Griess reaction was adopted as a standard method for nitrite determination in natural waters.8 However, such a diazo-coupling reaction is generally time consuming and requires a careful control of acidity for each step of the process.c-8 A number of more sensitive, kinetic methods have been reported for nitrite determination.9-l7 However, some1 1,13,14,17 are characterized by non-linear calibration graphs, whereas others1 1-13,16,17 lack the high sensitivity expected from kinetic techniques. Prochlorperazine (PCP) is a well known anti-emetic and antipsychotic phenothiazine derivative, which is susceptible to oxidation.It is first oxidized to a red radical cation (PCP') which may be further oxidized to a colourless sulfoxide derivative (PCPS).l* Therefore, PCP has been used as a redox indicator in titrimetric analysis.l9 However, it was only recently that PCP was introduced into kinetic analysis and applied to the determination of trace amounts of vanadium20 and iodide21 ions using their catalytic effects on the PCP-bromate and PCP-H202 reactions, respectively. The aim of this work was to study the catalytic oxidation of PCP with bromate in the presence of nitrite and to use the results obtained to develop a catalytic method for nitrite determination in natural waters. Experimental Reagents and Apparatus Analytical-reagent grade chemicals and distilled, de-ionized water were used throughout.A stock standard nitrite solution, 1000 pg ml-1, was prepared by dissolving 150 mg of pre-dried sodium nitrite in water, in the presence of one pellet of NaOH. The resulting solution was made up to the mark in a 100 ml calibrated flask and a few drops of chloroform were added as a stabilizer, to prevent bacterial growth. This solution was kept at 4°C and was used within 2 weeks of preparation.* A stock solution of 100 mmol 1-1 PCP was prepared by dissolving prochlorperazine maleate (Sigma St. Louis, MO, USA) in 20 ml of warm 1.0 moll-' sulfuric acid and making up to the mark in a 100 ml calibrated flask; the solution was stored in the dark at 4°C and was replaced every month. Working solutions were prepared daily by diluting the respective stock solutions.The following working standard solutions were used: a mixed reagent solution of 1.0 mmol 1-1 PCP and 5.0 mol 1-1 H3P04, 250 mmol 1-l sodium bromate and 1.0 pg ml-1 nitrite. The following masking solutions were also used: 1 mmol 1-1 Hg" and 50 mmol l-1 EDTA. Kinetic measurements were made on a Shimadzu (Kyoto, Japan) UV- 160A spectrophotometer equipped with 10 mm matched cells. The cell compartment of the spectrophotometer was thermostatically controlled by circulating water from a thermostated water-bath with a temperature stability of kO.1 "C. Eppendorf (Westbury, NY, USA) micro-pipettes (10-100 and 200-1000 pl) were used to deliver accurate volumes. Recommended Procedure The working solutions, sample solutions, water and 20 ml stoppered glass test-tubes were kept at 25 "C in the thermostated water-bath. A portion ( S 4.40 ml) of the sample was placed in one of the test-tubes and diluted to 4.40 ml with water after which 0.10 ml of each of the Hg" and EDTA masking solutions was added.After adding 0.20 ml of the mixed reagent solution, 0.20 ml of the bromate solution was added rapidly and the mixture was shaken well. A portion of the reacting solution was immediately transferred into the cell of the spectrophotometer in order to record the absorbance (A) versus time ( t ) graph at 525 nm against water. The rate (AAlAt) was calculated from the slope of the initial linear part of the A-t graph, within 1 min after addition of bromate. The nitrite content of the unknown sample was determined from a calibration graph, similarly prepared with the working standard nitrite solution.90 Analyst, January 1996, Vol.121 Results and Discussion Eflects of Reaction Variables The red radical cation, PCP', shows a sharp absorption band at 525 nm. Preliminary experiments showed that the position of this band does not change with varying acidity or reagent concentrations. Hereafter, k, and k, denote the rates of the catalysed and uncatalysed reactions, respectively, whereas (k, - k,) denotes the sensitivity and is expressed in ml min-1 ng- 1 . Fig. 1 shows that the reaction rates, k, and k,, linearly increased with PCP concentration in the range 40-120 pmol l-l; however, the sensitivity (k, - k,) remained almost constant.A 40 ymol 1-1 PCP concentration was adopted in the recommended procedure in order to provide a low reagent blank. The reaction rates, k, and k,, and the sensitivity largely increased with bromate concentration, as shown in Fig. 2. However, 10 mmol 1-1 bromate was chosen for the recom- mended procedure because it gave a moderate sensitivity and a low blank value. Phosphoric acid was chosen as the reaction medium in order to provide a suitable selectivity for the method. Fig. 3 shows that the reaction rates, k, and k,, and the sensitivity linearly increased with phosphoric acid concentration in the range 0.15-0.40 mol 1-I. Hence, 0.20 mol 1-1 phosphoric acid was adopted in the recommended procedure because of its low blank value. A reversed order of addition in which bromate was added before the mixed reagent solution gave lower rate values, probably because of partial oxidation of nitrite by bromate.The rate of the catalysed reaction, k,, gradually decreased with an increase in the standing time before addition of bromate, i.e., with an increase in the opportunity of reaction between nitrite and PCP. After 1,3 and 5 min standing time, k, decreased by 2, 10 and 26%, respectively. This might be attributed to the involvement of nitrite in a side reaction, probably formation of a catalytically less active nitrosated species of the reagent? Therefore, it is necessary to add bromate rapidly after adding the mixed reagent solution to the sample. Fig. 4 shows that the reaction rates, k, and k,, linearly increased with temperature; a temperature of 25 "C was adopted in the recommended procedure because of its convenience for the operation and its low blank value.Effects of Foreign Ions Iodide and vanadium are known to catalyse the oxidation of PCP,20,*1 and their interfering effects should be eliminated. On the other hand, preliminary experiments showed that Hg" and EDTA did not interfere with nitrite determination for up to 2 and 500 pg ml-l, respectively. Hence, the addition of Hg" and EDTA was adopted in the recommended procedure, in order to suppress the expected interferences of iodide and vanadium, respectively. The effects of possible interferents, which commonly accompany nitrite in natural waters, were investigated in the determination of 30 ng ml-1 of nitrite, following the recom- mended procedure.These ions were tolerated to a great extent without any pre-treatment. A foreign ion was considered to interfere seriously when it gave a determination error of more than 5%. The maximum tolerable concentrations of foreign ions are given in Table 1 and are much higher than the reported concentration levels of common pollutants in natural waters,* showing the high selectivity of the method. Calibration Graph A linear calibration graph was obtained, for up to 70 ng ml-1 of nitrite, using the optimum conditions of the recommended procedure. The linear regression equation for the calibration 0.04 I .- E 9 .g 0.02 8 2 . c) 5 0 40 80 120 [ PCP]/pmot I-' Fig. 1 Effect of PCP concentration on reaction rates. Except for the abscissa variable, reaction conditions are as given in the recommended procedure; k,, reagent blank; k,, 40 ng ml-1 nitrite.0 0.2 0.4 [H,POJ/mol I-' Fig. 3 Effect of H3P04 concentration on reaction rates. Conditions and symbols are as in Fig. 1. 0.04 S .- E . H 0 5 10 15 20 [Brornate]/mrnol I-' Effect of bromate concentration on reaction rates. Conditions and Fig. 2 symbols are as in Fig. 1. 0 20 30 40 TemperaturePC Fig. 4 Effect of temperature on reaction rates. Conditions and symbols are as in Fig. 1.Analyst, January 1996, Vol. 121 91 graph is: Rate (min-1) = 4.5 X "02-1, where "02-1 is the nitrite concentration expressed in ng ml-1. The correlation coefficient ( r ) is 0.9998 and the detection limit is 0.8 ng ml-1 of nitrite, which was calculated as three times the standard deviation of the blank (3s criterion).+ 6.0 X Table 1 Tolerance limits of foreign ions in the determination of 30 ng ml-1 of nitrite following the recommended procedure Tolerance limit/ pg ml-' Foreign ion > 100 Acetate, tartrate, citrate, oxalate, EDTA, Na+, K+, N&+, NO3-, F-*, C1-, C104-, S042-, Mg", Ca", Zn" 30 AP', Cd", Ni", Pb" 5 Cu", Feu', Fell*, S032- 1 0.1 * Species that produced negative interference. CoIi*, Cr"', Mn", MeV', W', ZrIV, Hg", I-, 103-, SCN-*, S2032-*, S2-*, Vv Ag', Br- Table 2 Comparison of dynamic ranges (detection limits) and selectivities of methods for nitrite determination Dynamic range (detection limit)/ Reaction system* ng ml-' Remarks? Ref. Sulfanilamide + NED 10-1000 R1 8 I- + SCN- - R2 10 11 PCPH + Br03- 40-920 - Thionine + BrO- 0.3-55 R3 13 Bindeschler's Green + Br2 WOO(50) F b 15 Chlorpromazine + Hz02 0-1500(3) R5 17 Prochlorperazine + Br03- 0-70(0.8) - This work * NED, N-( 1 -naphthyl)ethylenediamine; PCPH, pyridine-Zcarbal- dehyde-2-pyridylh ydrazone.t R1, coloured species that alter the colour system should be absent, along with Fe"', Pb", Vv and Cu" ions; R2, measurements of induction period for fading of the red colour of [Fe(SCN)]*+ with different calibration graphs depending on the initial concentrations and medium acidity; R3, a fixed- time method where measurements are made at 30 "C after 5 min of mixing with poor precision resulting from a very high blank value; %, poor selectivity; R5, non-linear calibration graph. ~~~~ ~ Table 3 Determination of nitrite in natural waters Sample* Nitrite/ng ml- * Amount Found+ s, Recovery pH takerdm1 Added in sample (%) (%) River water- 8.01 3.20 - 170.0 f 0.4 0.2 1.60 - 172.4 f 0.6 0.4 Average = 171.2 f 0.7 1.10 20.0 190.4 f 0.5 0.3 96 Rain water- 4.10 4.00 - 19.0 f 0 .2 1.1 3 .OO - 19.4 f 0.2 1.0 Average = 19.2 f 0.3 2.00 35.0 54.0 f 0.2 0.4 99 * Collected at Kofu city, Japan; river water was taken from the Aikawa river on April 6, 1995; rain water was collected on the roof of Yamanashi University on April 9, 1995. + Mean f standard deviation (n = 5). In order to examine the accuracy and precision of the method, standard solutions of 10, 40 and 70 ng ml-1 of nitrite were analysed using the recommended procedure. Five replicate determinations of each concentration gave relative standard deviations (sJ of 2.1,0.9 and 1.4%, respectively, and the values of the Student's t-test were G 1.3 (the tabulated t-value for the 95% confidence level and n = 5 is 2.7Q.22 The above results show that the t-test could not detect any systematic error in the method and thus indicates its reliability.Table 2 compares the proposed method, the standard method utilizing the modified Griess reaction and other well established kinetic methods. Determination of Nitrite in Natural Waters A freshly collected water sample was filtered through a 0.45 pm Millipore membrane filter (MF-HA type), kept at 4 "C, to retard bacterial growths and analysed within 6 h of collection. Since the concentrations of common pollutants in natural waters are generally far below the tolerance levels shown in Table 1, the method was applied directly to the determination of nitrite in natural waters. Table 3 shows the analytical results for rain and polluted river waters, obtained following the recommended procedure.The reliability of the method to analyse real samples was checked by recovery experiments, which gave quantitative results (9699%) with convenient reproducibility (s, = 0.2-1.1 %). One determination takes about 2 min compared with about 20 min taken by the standard method utilizing the modified Griess reaction. The rapidity, simplicity, high sensitivity and selectivity are significant advantages of the proposed method. This work was accomplished, in part, at Yamanashi University, where A.A.M. was on a scholarship from the Ministry of Education, Egypt.This is gratefully acknowledged. References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Shotyk, W., J . Chromatogr., 1993, 640, 309. Ohta, K., and Tanaka, K., Bunseki Kagaku, 1994, 43, 471. Mou, S. F., Wang, T. H., and Sun, Q., J . Chromatogr., 1993, 640, 161. Li, J.-Z., Wu, X.-C., Yuan, R., Lin, H.-G., and Yu, R. Q., Analyst, 1994,119, 1363. Schaller, U., Bakker, E., Spichiger, U. E., and Pretsch, E., Talanra, 1994,41, 1001. Boltz, D. F., and Howell, J. A., Colorimetric Determination of Non- metals, Wiley, New York, 1978. Fox, J. B., Crit. Rev. Anal. Chem., 1985, 15, 283. APHA-AWA-WEF, Standard Methods for the Examination of Water and Wastewater; ed. Greenberg, A. E., Clesceri, L. S., and Eaton, A. D., American Public Health Association, Washington, DC, 18th edn., 1992.Perez-Bendito, D., and Silva, M., Kinetic Methods in Analytical Chemistry, Ellis Horwood, Chichester, 1988. Utsumi, S., Okutani, T., Sakuragawa, A., and Kenmatsu, A., Bull. Chem. SOC. Jpn., 1978, 51, 3496. Kupparis, M. A., Walczok, K. M., and Malmstadt, H. V., Analyst, 1983,107, 1309. Sakuragawa, A., and Okutani, T., Bunseki Kagaku, 1985,35,43. Montes, R., and Laserna, J. J., Talanta, 1987, 34, 1021. Jiang, M., Jiang, F., Duan, J., Tang, X., and Zhao, Z., Anal. Chim. Acta, 1990, 234, 403. Okutani, T., Sakuragawa, A., Kamikura, S., Shimura, M., and Azuchi, S . , Anal. Sci., 1991, 7, 793. Ensafi, A. A., and Saminifar, M., Talanta, 1993, 40, 1375. Liang, B., Iwatsuki, M., and Fukasawa, T., Analyst, 1994, 119, 21 13. Advances in Heterocyclic Chemistry, ed. Katritzky, A. R., and Bulton, A. J., Academic Press, New York, 1968, vol. 9, pp. 321-460.92 Analyst, January 1996, Vol. 121 19 Phenothiazines and 1,4-Benzothiazines: Chemical and Biomedical Aspects, ed. Gupta, R. R., Elsevier, New York, 1988, pp. 861-898. 20 Mohamed, A. A., El-shahat, M. F., Fukasawa, T., and Iwatsuki, M., Anal. Chim. Acta, submitted for publication. 21 Mohamed, A. A., El-shahat, M. F., Fukasawa, T., and Iwatsuki, M., Anal. Chim. Acta, submitted for publication. 22 Miller, J. C., and Miller, J. N., Statistics for Analytical Chemistry, Ellis Honvood PTR, Prentice-Hall, New York, 3rd edn., 1993. Paper 5l04081 G Received June 23, I995 Accepted September 9,1995