Fresenius J Anal Chem (1999) 363 : 265–269
© Springer-Verlag 1999
O R I G I N A L PA P E R
A. Ruiz-Medina · M. L. Fernández-de Córdova · A. Molina-Díaz
Flow injection-solid phase spectrofluorimetric determination of pyridoxine in presence of group B-vitamins
Received: 4 June 1998 / Revised: 16 July 1998 / Accepted: 6 August 1998
Abstract A flow-through optosensor has been prepared for the sensitive and selective determination of pyridoxine (vitamin B6) in aqueous solutions. The sensor was developed in conjunction with a monochannel flow-injection analysis system with fluorimetric detection using Sephadex SP-C25 resin as an active sorbent substrate. This method of determination is carried out without any derivatization. The wavelengths of excitation and emission were 295 and 385 nm, respectively. When a HCl (10–3 mol L–1) / NaCl (3 × 10–2 mol L–1) solution is used as carrier solution, the sensor responds linearly in the measuring range of 5–200, 10–400 and 50–1800 ng mL–1 with detection limits of 0.33, 0.67, and 5.70 ng mL–1 for 2000, 1000 and 200 µL of sample volume, respectively. The relative standard deviation for ten independent determinations is less than 0.75% for 0.2 and 1.0 mL of sample volumes used, and 1.31% for 2.0 mL of sample volume used. The method was satisfactorily applied to the determination of vitamin B6 in pharmaceutical preparations.
Introduction Fluorimetry, one of the most widely used techniques due to its high sensitivity, has often been used as an officially recommended technique for determining the purity of drugs or their contents in pharmaceutical samples [1]. The FIA (flow injection analysis) assembly required for developing analytical methods based on the measurement of native fluorescence is very simple. It has been used successfully for the resolution of analytes such as 9aminoacridine [2], ergotamine tartrate [3], sulphanilamide, sulphaguanidine and sulphamethazine [4] and
A. Ruiz-Medina · M. L. Fernández-de Córdova · A. Molina-Díaz (쾷) Department of Physical and Analytical Chemistry, Faculty of Experimental Sciences, University of Jaén, E-23071 Jaén, Spain e-mail:
[email protected]
oxytetracycline, tetracycline and chlortetracycline [5, 5a]. Nevertheless, substances emitting very little or no fluorescence must be transformed into a fluorescent compound by means of reactions as the formation of fluorescent chelates, formation of ion-pairs, redox processes or enzyme catalysis. Many strategies and reactions have been used in FIA for fluorimetric analysis of drugs. So, ascorbic acid can be readily oxidized by mercury (II) chloride [6], cupric ion is used as oxidant for the determination of cysteine and cystine [7] and pyridoxal and pyridoxal 5-phosphate are determined by oxidation with cyanide [8]. Recently, ion-exchange resins (on which the analyte is preconcentrated) combined with non-destructive spectrophotometric or spectrofluorimetric detectors (batch or FIA methodology) have been used [9–12]. The use of a solid support placed in the flow-cell of the fluorescence detector monitoring the inherent signal of the analyte sorbed on the support (namely a flow-through fluorimetric sensor) allows an important increase both in the selectivity and the sensitivity of the determination [13]. This has been used for the determination of analytes such as riboflavin, by direct measurement of the native fluorescence [14], or B6 vitamins by previous derivatization [15, 16]. Vitamin B6 is widely distributed in animal and plant tissues. Some spectrophotometric methods have been described for the determination of pyridoxine [17, 18] or the simultaneous determination of multiple vitamins including B6 [19–22]. Only one method for the fluorimetric determination of this vitamin is described [23]. The aim of the present work was to develop a flow-through sensor for the determination of pyridoxine. We found that pyridoxine was well retained on a Sephadex SP-C25 resin showing a much higher fluorescence signal than in homogeneous aqueous solution. The retained pyridoxine could also be quickly and completely eluted from the resin by the carrier itself. This appears to be the first attempt to determine pyridoxine with a flow-cell type chemical sensor with simultaneous retention, determination and elution.
266 Fig. 1 a Schematic diagram of the FIA system: C/E: carrier/ eluent; P: peristaltic pump; IV: injection valve; S: sample; F: spectrofluorimetric detector; FC: flow-through cell; C: computer; D: printer; W: waste; b Fluorescence flow-through cell filled with support
a
Experimental section Apparatus Fluorescence emission measurements were obtained with a Perkin Elmer LS-50 spectrofluorimeter equipped with a xenon discharge lamp (20 kW), Monk-Gillieson monochromators, a Quantic Rhodamine 101 counter to correct the excitation spectra and a Gated photomultiplier. The luminescence spectrometer was interfaced with a Mitac MPC 3000F-386 microcomputer supplied with FL Data Manager Software for spectral acquisition. Instrument excitation and emission slits were set at 2.5 and 15 nm, respectively, and the scan rate of the monochromators was maintained at 240 nm min–1, throughout the study. A four-channel Gilson Minipuls-3 peristaltic pump with rate selector was used to generate the flow stream. A Rheodyne Model 5041 injection valve with variable sample loops and PTFE tubing of 0.8 mm i.d. were also used. Figure 1 a illustrates the simple optosensing FIA manifold used. The spectrofluorimeter was furnished with a Hellma 176-QS quartz flow-through cell with a light-path length of 1.5 mm, packed with the appropriate resin. The complete light-path was filled with a suspension of Sephadex SP-C25 resin in water with the aid of a syringe. The cell was blocked at the outlet with some glass wool to prevent displacement of the Sephadex particles by the carrier; the inlet of the flow-through cell was kept free (Fig. 1 b). Reagents All the reagents used were of analytical-reagent grade and solutions were prepared in bidistilled water. Pyridoxine stock standard solution (4.86 × 10–4 mol L–1) was prepared by dissolving the required amount of pyridoxine hydrochloride (FLUKA) in water. The solution was stable for two weeks in a refrigerator at about 5 °C. Working solutions were prepared daily by appropriate dilution of the stock solution with bidistilled water. The carrier solution used in the FIA experiments was HCl (10–3 mol L–1) / NaCl (3 × 10–2 mol L–1) solution. Sephadex SP-C25 resin (Aldrich) was used in the H+ form as the solid support without any pretreatment.
b
After reaching the maximum signal, pyridoxine was desorbed from the flow-through cell by the carrier itself, therefore regenerating the active ion-exchanger gel microzone, and allowing the luminescence value to return to the baseline. Treatment of sample Pharmaceuticals containing vitamin B6 were directly dissolved in bidistilled water, filtered through a 0.45 µm Millipore filter membrane, and diluted to volume with bidistilled water and used without any pretreatment.
Results and discussion Spectral characteristics Figure 2 shows the fluorescence spectra of pyridoxine, both in aqueous solution and on the Sephadex SP-C25 resin. When the spectra in aqueous solution and on the resin are compared a 5 nm bathochromic shift in its excitation wavelength is found, and a 5 nm hypsochromic shift in its emission wavelength. This can be attributed to the modification of the surrounding environment of pyridoxine in the solid phase with respect to the solution. The peak wavelength in the excitation spectrum of the analyte on the resin is 295 nm and the emission maximum is located
Procedure The sample solution (200, 1000 or 2000 µL) containing 50–1800, 10–400, 5–200 ng mL–1 of pyridoxine hydrochloride was inserted into the carrier stream (HCl / NaCl, total concentration 0.031 mol L–1) at a flow-rate of 1.55 mL min–1. The analyte was transported through the flow cell where it was absorbed on the cationic Sephadex SP-C25 resin. The fluorescence emission intensity was measured continuously at 385 nm (15 nm slit-width) using an excitation wavelength of 295 nm (2.5 nm slit-width). Three different calibration lines were constructed in the concentration ranges and sample volumes indicated above.
Fig. 2 Fluorescence spectra (excitation and emission spectra) of pyridoxine: (1, 1′) equilibrated on Sephadex SP-C25 resin; (2, 2′) in aqueous solution. Pyridoxine hydrochloride 1.8 ng mL–1 (sample volume: 200 µL)
267
at 385 nm. At these wavelengths the emission signal due to the resin is low. The absorption on Sephadex SP-C25 resin resulted in a fluorescence signal about 10 times greater than in aqueous solution in the same flow cell. This is the result of the accumulation of the cationic pyridoxine on the resin. Influence of the pH of the carrier solution and of the sample The effect of pH on the fluorescence intensity was studied in the carrier and in the sample. Various solutions with an adequate concentration of HCl or NaOH (for the 1–10 pH region) were used as carrier. For pH values higher than 2, the fixation of the analyte on the resin was so quick that the resin area with the retained analyte could not completely be irradiated. Therefore it was necessary to use an electrolyte (NaCl 0.02 mol L–1) which transported the analyte to the totally irradiated area. The maximum fluorescence signal was obtained between HCl concentrations of 10–3 and 10–4 mol L–1. A concentration of HCl 10–3 mol L–1 was chosen. At this HCl concentration value, several electrolytes at various concentrations were tested. The best results were obtained with a concentration 3 × 10–2 mol L–1 NaCl. Therefore, a solution HCl (10–3 mol L–1) / NaCl (3 × 10–2 mol L–1) was used as carrier solution. The sample pH value did not influence the analytical signal when its value was maintained in the range 2 to 10 and, hence, there is no need to adjust the sample pH. Influence of flow rate The relative fluorescence intensity of pyridoxine fixed on the resin was studied at different flow rates using the same concentration of the analyte (Fig. 3). A decrease in flowrate resulted, as expected, with increasing residence time and peak heights, which indicates that the absorption of the analyte was not instantaneous. A flow-rate of 1.55 mL min–1 was selected as a compromise between the sensitivity and the throughput.
Fig. 4 Calibration lines obtained from different sample volumes injected. (1) 2000, (2) 1000 and (3) 200 µL. Inset: Sample volume effect in increasing order (pyridoxine hydrochloride 0.2 ng mL–1 and flow rate 1.55 mL min–1): 200, 300, 600, 800, 1000, 1300, 1600, and 2000 µL
Effect of sample volume One of the main advantages of the sensor is the potential increase in the sensitivity by increasing the sample volume taken for analysis due to the concentration of a higher amount of analyte on the active microzone. Sample volumes from 200 to 2000 µL using the same concentration of pyridoxine (200 ng mL–1) were studied and it was found that the increase of the size of the sample loop resulted in a linear increase of luminescence emission (Fig. 4). The use of higher injection volumes increases the sensitivity but also reduces the sampling frequency, so the sensor was calibrated for three different injection volumes: 200, 1000 and 2000 µL. Figures of merit Table 1 contains the figures of merit of the method proposed for the three calibration volumes, and Fig. 4 shows the three calibration lines. The data were fitted by standard least-squares treatment and the calibration equations obtained are shown. Table 1 Figures of merit Parameter
Fig. 3 Effect of flow rate: (1) on fluorescence intensity and (2) on residence time. Pyridoxine hydrochloride 0.75 ng mL–1 (sample volume 200 µL and flow rate 1.55 mL min–1)
Volume of sample loop (µL) 200
1000
2000
Linear dynamic range (ng mL–1)
50–1800
10–400
5–200
Calibration graph Intercept Slope (mL ng–1) Correlation coeffiencet Detection limit (ng mL–1) RSD (%) (n = 10) Sampling frequency (h–1)
16.40 0.547 0.999 5.70 0.45 40
0.00 2.284 0.999 0.67 0.70 32
–0.74 4.729 0.999 0.33 1.31 26
268
Other relevant analytical figures of merit are also included in Table 1. The reproducibility was established for ten independent analyses of solutions containing 1000, 200 and 50 ng mL–1 of analyte for 200, 1000 and 2000 µL, respectively. The detection limit was calculated, by using the 3 σ-criterion [24], and values of 5.70, 0.67 and 0.33 ng mL–1 were obtained for the three different calibrations. A sampling frequency of about 40 h–1 can be achieved by using this methodology provided a 200 µL sample volume is used. This is a relatively high value for this type of sensor. Interference study In order to determine the potential effect of foreign species, a study was carried out with a pyridoxine conTable 2 Interference study (determination of 1 µg mL–1 of pyridoxin) Foreign species
Tolerated interferent/ analyte (w/w)ratio
Glucose, saccharose, saccharin, L-carnitine, L-lysine, nicotinamide, Ca pantothenate, methionine, glutamic acid, citric acid, biotine Caffeine, ascorbic acid, diclofenac Na Salicylamide Vitamin B12, vitamin B2 Vitamin B1
> 100 100 25 10 100
centration of 1 µg mL–1. A foreign species is considered not to interfere if it produces an error not exceeding 3% in the determination of the analyte. The influence of 18 different interferent species on the determination of pyridoxine was studied. The results are summarized in Table 2. It can be seen that the procedure proposed shows a high level of tolerance to other species frequently found along with pyridoxine and to the other vitamins of the B group (B12, B2, B1). We found that in conventional spectrofluorimetry in homogeneous phase solution the tolerance levels (w/w ratio) to vitamins B12, B2, B1 and salicylamide are 2, 0.01, 1 and 1, respectively. If these values are compared to the amount tolerated in the proposed method, they are 5, 103, 1000 and 25 times higher than those in the homogeneous solution method; therefore, the proposed method is highly selective. This effect is mainly due to the retention of the pyridoxine cation by the resin and its consequent separation from the matrix.
Application of the method The proposed sensor was applied to the determination of pyridoxine in pharmaceutical preparations using a 200 µL loop for sample injection. The results obtained are summarized in Table 3. As can be seen, in all the cases the data were in good agreement with the labelled amounts.
Table 3 Determination of pyridoxine hydrochloride in pharmaceutical preparations Sample
Other species
Vitagama Fluor (Amirrall, Ltd.) Astenolit (Elmu, Ltd.) Neurodavur (Belmac, Ltd.) Antomiopic (Cibavision, Ltd.) Trimetabol (J. Uriach, Ltd.) Vitalter (Alter, Ltd.)
Vitamins (A, B1, B2, C, D3, E, H, PP), sodium fluoride, sodium pantothenate, Fe2+, benzoic acid sulphimide Vitamins (B1, B12), inositol, carnitine, saccharin, saccharose, potassium and magnesium aspartate, acetylglutamine, ethanol Vitamins (B1, B12), lidocaine, dexamethasone 21-phosphate disodium salt Vitamins (A, E), L-citrulline, N-acetyl-L-aspartic acid
Gota Cerebrina (Derly, Ltd.) Actilevol Orex (Wassermann, Ltd.) Antineurina (T-Meiji F., Ltd.) Agudil (Sigma-Tau, Ltd.) Dolo-Nervobión (Merck, Lab.) a Mean
Pyridoxine labelled (mg)
Recovery mean ± RSDa (%) (mg)
1
101.8 ± 0.6
50
100.2 ± 0.7
50
101.8 ± 0.8
25
99.8 ± 0.7
Vitamins (B1, B12), lysine, carnitine, saccharose, saccharin, D-sorbitol 70% Vitamins (A, B1, B2, B12, C, D3, E, PP), sodium fluoride, sodium pantothenate, magnesium oxide, calcium carbonate, Fe2+, folic acid, carbohydrates Vitamins (B12, A, D3, PP), pantothenic acid, saccharin
30
101.6 ± 0.4
Vitamins (B1, B12, C), saccharose, saccharin, ciproheptadine, carnosine, hematoporphyrin Vitamins (B1, B12), lidocaine
50
L-asparagine, o-phospho-DL-serine, L-glutamine, lactose Vitamins (B1, B12), saccharose, diclofenac
of three determinations
1.1
101.8 ± 0.7
1.85
100.2 ± 0.8 101 ± 1
125
101.7 ± 0.6
10
101.4 ± 0.8
200
102 ± 1
269
Conclusions The proposed sensor shows the feasibility for an automatic continuous analysis based on the integrated separation and spectrofluorimetric detection of pyridoxine without any derivatizing reaction. The results obtained by applying this method to pharmaceuticals lead to the conclusion that this sensor allows a simple, fast, inexpensive and precise determination of pyridoxine in pharmaceutical preparations. Moreover, the proposed method has the following features: it uses a very simple single-channel manifold with a conventional spectrofluorimetric detector, the elution being carried out by the carrier itself requires no reagent, requires no sample pre-treatment, provides excellent selectivity, allows a high sampling frequency and provides low determination limits and a good linear determination range. Acknowledgement The authors are grateful to the Ministerio de Educación y Cultura, Subdirección General de Proyectos de Investigación Científica y Técnica (Project No. PB97-0849) for financial support.
References 1. Martínez Calatayud J (1996) Flow Injection analysis of pharmaceuticals. Taylor & Francis, Great Britain, pp 151–152 2. Martínez Calatayud J, Sánchez Sampedro A, Villar Civera P, Gómez Benito C (1990) Anal Lett 23 : 2315 3. Polasek M, Solich P, Karlicek R (1989) Ceskoslov Farm 38 : 1 4. Mahedero MC, Aaron JJ (1992) Anal Chim Acta 269 : 193
5. Haagsma N, Mengelers MJB (1989) Z Lebensm Unters Forsch 188 : 227 5 a. Martínez Calatayud J (1996) Flow Injection analysis of pharmaceuticals. Taylor & Francis, Great Britain, p 158 6. Chung HK, Ingle JD (1991) Talanta 38 : 355 7. Pérez-Ruiz T, Martínez Lozano C, Tomas V, Carpena J (1992) Analyst 117 : 1025 8. Linares P, Luque de Castro MD, Valcárcel M (1985) Anal Lett 18 : 67 9. Ortega Barrales P, Fernández de Córdova ML, Molina Díaz A (1998) Anal Chem 70 : 271 10. Ayora Cañada MJ (1998) PhD Thesis, University of Jaén 11. Capitán F, Alonso E, Avidad R, Capitán Vallvey LF, Vilchez JL (1993) Anal Chem 65 : 1336 12. Capitán Vallvey LF, Valencia MC, Mirón G (1994) Anal Chim Acta 289 : 365 13. Valcárcel M, Luque de Castro MD (1990) Analyst 115 : 699 14. Zhilong G, Zhujun Z (1997) Anal Chim Acta 339 : 161 15. Danhua Chen, Luque de Castro MD, Valcárcel M (1992) Anal Chim Acta 261 : 269 16. Chauhan MS, Dakshinamurti K (1981) Clin Chim Acta 109 : 159 17. Reddy MN, Rao SP, Sankar DG (1996) Indian Drugs 33 : 569 18. Gupta MK, Jain CL (1991) Indian Drugs 26 : 275 19. Li Z, Li M, Zhi L, Zheng G, Yu R (1991) J Anal Toxicol 15 : 260 20. Hewala II (1993) Anal Lett 26 : 2217 21. Bautista RD, Jiménez AI, Jiménez F, Arias JJ (1996) J Pharm Biomed Anal 15 : 183 22. Onur F, Dinc E (1990) Gazi Univ Eczacilik Fak Derg 7 : 77; (1991) Anal Abst 53 : 11G7 23. Gao GG, Yang GY (1992) Shenyang Yaoxueyuan Xuebao 9 : 18; (1993) Anal Abst 55 : 8G247 24. IUPAC, Nomenclature, symbols, units and their usage in Spectrochemical Analysis (1976) Pure Appl Chem 105 : 45
