Fresenius J Anal Chem (2001) 369 : 176–183

© Springer-Verlag 2001

O R I G I N A L PA P E R

Alireza Ghassempour · Mohammad Reza Arshadi · Feridoun Salek Asghari

Determination of the pesticide naptalam and its degradation products by positive and negative ion mass spectrometry

Received: 23 May 2000 / Revised: 21 August 2000 / Accepted: 29 August 2000

Abstract N-1-Naphthylphtalamic acid (naptalam) and its degradation products, 1-naphthylamine and N-(1-naphthyl) phthalimide were simultaneously determined in river water by two independent mass spectrometric (MS) methods. These were negative ion MS (NIMS) and programmable temperature vaporizer gas chromatography mass spectrometry (PTV-GC MS) with electron impact ionization (positive ions). Prior to the NIMS analysis, the samples were preconcentrated by solid phase extraction (SPE) of C18 membrane discs. The PTV-GC MS studies were performed without any preconcentration procedure. Selected ion monitoring (SIM) and internal standardization with naphthalene were applied in both methods. The limits of determination (LOD) of NIMS studies were 230, 270 and 260 ng L–1 for naptalam, 1-naphthylamine and N(1-naphthyl) phthalimide, respectively, with relative standard deviation (RSD) < 1% (n = 5) and of PTV-GC MS 17, 11 and 15 ng L–1 (RSD < 0.7%, n = 5). The LOD, linearity, RSD and time required for these methods are far better than for HPLC analyses.

Introduction N-1-Naphthylphthalamic acid, commonly known as naptalam, and its derivatives are useful in growing a number of food crops. These compounds constitute an important group of plant regulators, which inhibit seed germination and are recommended for pre-emergence control of weeds [1, 2].

There have been a number of investigations on the determination of naptalam and its degradation products, 1naphthylamine and N-(1-naphthyl) phthalimide, in various media such as river, drinking water, urine, etc. [3, 4]. These studies have been carried out by various spectrometric [5, 6] and chromatographic methods [7–11]. The electron impact ionization mass spectrometry (positive ions) of naptalam along with some other herbicides has been studied by Cairns et al. [12], but no attempts have been made to undertake a quantitative investigation and/or natural degradation processes by mass spectrometric methods. In a previous paper, the degradation process and simultaneous determination of naptalam and its degradation products by FT-IR were reported by Kargosha et al. [13]. In the present work, two MS methods, NIMS and PTV-GC MS, are investigated for the simultaneous determination of naptalam and its degradation products at low concentrations in river water. NIMS is often used for complex compounds to complement the information obtained from positive ions [14–16]. The PTV injection is usually used for large volumes and unstable compounds at higher temperature [17, 18]. This paper shows that naptalam, 1-naphthylamine and N-(1naphthyl) phthalimide can be easily identified and determined by NIMS and PTV-GC MS in river water.

Experimental Reagents The HPLC grade methanol, acetic acid, naphthalene, ether, phthalic anhydride and 1-naphthylamine were supplied by Merck.

A. Ghassempour () Department of Chemistry, Faculty of Science, Guilan University, P. O. Box 41335–1914, Rasht, Iran M. R. Arshadi Department of Chemistry, Sharif University of Technology, Tehran, P. O. Box 11365–9515, Iran F. S. Asghari Quality Control Laboratory, Zakaria Pharm. Co., Tabriz, 51575–5181, Iran

Procedure Naptalam was prepared by dissolving 143 mg of 1-naphthylamine in 15 ml ether, adding 136 mg of phthalic anhydride. This solution was refluxed for 10 min. The mixture was filtered and recrystallized from ethanol, mp 185 °C [2]. When ether is substituted by pyridine, in the preparation of naptalam procedure, N-(1-naphthyl) phthalimide is produced.

177 Standard solutions of naptalam, 1-naphthylamine and N-(1-naphthyl) phthalimide were prepared in 25 different concentrations from 1 ng L–1 to 2 mg L–1 in river water. For high concentration samples of N-(1-naphthyl) phthalimide, slightly heating will aid the dissolution. Negative ion mass spectra were recorded by VG instrument model Trio 1000. The ion source was operated at 70 eV and the temperatures of ion source and probe were 150 and 120 °C, respectively, under a pressure of 10–2 Pa with a constant helium flow of 0.1 mL min–1. For NIMS analyses, the preconcentration was accomplished by passing the river water (1 L) through a C18 membrane disc with a diameter of 47 mm (Varian Co., No. 1214–5004), followed by passing a flow of dry nitrogen for 2 min. The disc was washed with 5 × 2 mL methanol and the eluate was collected in a 10 mL graduated flask and finally evaporated to 1 mL volume. After extraction has been carried out, the disc was washed again with 10 mL of water and 10 mL of methanol, respectively. In order to remove any residual of inorganic and organic compounds it was dried by passing nitrogen through the disc. The quantitative NIMS analysis was performed by the SIM method with naptalam (m/z 148), 1-naphthylamine (m/z 142), N(1-naphthyl) phthalimide (m/z 273) and naphthalene (m/z 128), respectively. Sample volumes of 250, 500, 750, 1000, 1250 and 1500 mL of water spiked with naptalam to maintain constant the total pesticide amount (1 µg each), adjusted to pH 5–5.2 (pH of river water), were used for calculation of the disc recovery and breakthrough volume. Alternately, the volume of the aqueous solutions can be kept constant (1 L) and the concentrations of naptalam, 1-naphthylamine and N-(1-naphthyl) phthalimide are varied, 4, 2, 1.33, 1, 0.8, 0.66 µg L–1, adjusted to pH 5–5.2. These samples were also used for breakthrough test. Electron impact ionization analyses were performed by HP GC MS model 5992A [18]. Capillary column was BP5 (30 m × 0.32 mm i.d., 0.33 µm film thickness). Aliquots of river samples (100 mL), dissolving different concentrations of the compounds mentioned above, were used for analyses by PTV-GC MS. Determination by PTV-GC MS was carried out under the following conditions: injection temperature, 110 °C for 3 s at split mode and raised to 210 °C with 50°C min–1 at splitless mode; oven temperature 175 °C; carrier gas, helium at 1 mL min–1; injection, 100 µL. The SIM method for electron impact ionization and internal standardization with naphthalene was performed at m/z 143, 143, 273 and 128 for naptalam, 1-naphthylamine, N-(1-naphthyl) phthalimide and naphthalene (internal standard), respectively. The liquid chromatography system was equipped with a Waters pump model 616. The UV-spectra of HPLC detector were recorded at wavelengths of λ = 200–600 nm using a diode array detector (DAD). Aliquots of river water (100 mL), spiked with different concentrations of naptalam, 1-naphthylamine and N-(1-naphthyl) phthalimide and naphthalene passed through a SPE cartridge. The samples were then eluted with 10 mL of acetonitrile. These samples were measured by an isocratic system of HPLC using a Waters C18 column and a C18 pre-column. The mobile phase was 0.1% acetic acid in acetonitrile. The flow rate was 2 mL min–1. The diode array detector was adjusted at 220 nm for quantitative works. The LOD is calculated based on the ratio of three times the standard deviations (Sd) to the slope of the calibration curve. The linearity of these methods is calculated based on the limit of linear response to the limit of the quantitative measurements (where the calibration curves depart from linearity). In this work, the standard solutions mentioned above were used for the determination of the LODs. For analyses of these compounds in river water samples which usually contain microbial activities and particular organic and inorganic substances, the samples must be filtered by GH polypro filter (WAT 200537). For the reliable peak integration of these samples, peak heights of approximately 25 times the noise level are required resulting in determination limits that are approximately 10 times higher than the LOD.

Results and discussion In the aqueous environment, naptalam may degrade to 1-naphthylamine, a well known carcinogenic agent, and N-(1-naphthyl) phthalimide [3, 4]. These two degradation pathways are represented in Fig. 1. When electron impact ionization is used for determination, many types of fragment ions of naptalam, 1-naphthylamine and N-(1-naphthyl) phthalimide would be formed in the ion source. Figure 2 a–c represents the electron impact ionization fragments of naptalam, 1-naphthylamine and N-(1-naphthyl) phthalimide, respectively. These spectra reveal high degree of similarity. Thus, electron impact ionization mass spectrometry alone cannot be utilized for the simultaneous determination of naptalam, 1-naphthylamine and N-(1-naphthyl) phthalimide cations in aqueous media. Negative ion mass spectrometry Naptalam, 1-naphthylamine and N-(1-naphthyl) phthalimide can easily be detected by NIMS. Negative ions are usually formed by secondary electron capture, dependent on ion source pressure. These low-energy electrons produced a simple spectrum [14]. Figure 3 a–c represents typical negative ion spectra of naptalam, 1-naphthylamine and N-(1-naphthyl) phthalimide, respectively. These compounds show molecular ions (or M-1) in their negative ion spectra. The helium flow rate of 0.1 mL min–1 through the ion source produces a constant ion source pressure. Small variation of the helium flow rate did not result in a significant change of the negative ion spectra of these compounds. Thus, it can be assumed that a minor variation of the ion source pressure in the region of this study (10–2 to 10–3 Pa) had no effect on negative ion spectra of these compounds. As it can be seen from these figures, each compound shows characteristic peaks that can be used for the simultaneous determination of them. Naptalam is a thermal unstable compound; therefore, temperatures of the ion source and probe are critical parameters for naptalam analyses and should be optimized. The results show no decomposition at 150 and 120 °C of the ion source and probe temperature, respectively.

Fig. 1 Pathways of the naptalam degradation

178

a

b Fig. 2 Positive ion mass spectra of (a) naptalam, (b) 1-naphthylamine and (c) N-(1-naphthyl) phthalimide

All three compounds register either the molecular or (M-1) peaks that can be used for unique identification and quantitative calculations. In the NIMS of the original compound, naptalam, the base peak at m/z 148, in all likelihood,

corresponds to the phthalic anhydride anion. The peak at m/z 273 can be attributed to the loss of a water molecule and the formation of N-(1-naphthyl) phthalimide anion from the molecular anion. Therefore, individual detection of these three compounds is almost easily possible in all natural media. The SIM method with naphthalene (MW = m/z 128), as an internal standard, can be employed for

179

c Fig. 2 c

a Fig. 3 Negative ion mass spectra of the (a) naptalam, (b) 1-naphthylamine and (c) N-(1-naphthyl) phthalimide

180

b

c Fig. 3 b, c

quantitative NIMS works. The total abundance of the m/z 273 peak in the mixture belongs to naptalam and to N-(1naphthyl) phthalimide. Thus, the abundance of the m/z 273 peak used for the determination of N-(1-naphthyl) ph-

thalimide must be taken account of. The ratio of the m/z 291 to the m/z 273 peaks of naptalam, related to the internal standard peak, at constant temperature, is fixed. Therefore, the abundance of the m/z 273 peak of naptalam must be deduced from the total intensity of the m/z 273 peak for the determination of N-(1-naphthyl) phthalimide.

181 Fig. 4 HPLC chromatograms of 1-naphthylamine, N-(1-naphthyl) phthalimide, naptalam and naphthalene in different concentrations

The LOD was further improved by means of the SPE method, employing C18 disc, prior to NIMS analysis. The membrane extraction discs have been used as an alternative to the SPE cartridge. Higher flow rates can be applied with SPE discs and the extraction time is much shorter than that using cartridges. They do not get plugged, channeling does not occur, and they exhibited lower blanks than C18 cartridges [19–21]. Preconcentration on C18 membrane disc improves the LOD of naptalam, 1-naphthylamine and N-(1-naphthyl) phthalimide by about 100 times with recoveries ranging from 94.5 to 100.3%. If the solvent (methanol, 10 mL) is vaporized to the minimum volume (1 mL), the LOD was further improved by another factor of about 10 (1000 overall). The LOD for naptalam, 1-naphthylamine and N-(1-naphthyl) phthalimide, after preconcentration, were 230, 270 and 260 ng L–1, respectively. These values are better than LOD of the HPLC results, which are around 2400 ng L–1. Figure 4 shows an HPLC chromatogram of the separation of naptalam, 1naphthylamine, N-(1-naphthyl) phthalimide and naphthalene (internal standard) in five different concentrations. The linearity of the HPLC method is low (2.4 to 4800 µg mL–1) and the SPE cartridge requires preconcentration of naptalam and its degradation products for the determination of the low concentrations (about 168 ng mL–1) in river water [3]. The NIMS linearity of naptalam is about 7000 in comparison with HPLC (about 2000). Although NIMS can simultaneously identify and determine compounds at short time and high precision (RSD < 1%, n = 5), it is not sophisticated enough to control the threshold value for drinking and ground water as defined for the European Community (100 ng pesticide L–1 water, same as the HPLC method [17]).

In a previous work, the effect of sample volume on recovery of naptalam with two solvents was studied by Kargosha et al. [13] and the maximum suitable volume of the sample was determined. In this work, the narrow volume range of a single solvent (methanol) is used, near the suitable volume of previous work (1000 mL), for recovery of low concentrations of these compounds. The recoveries of naptalam, 1-naphthylamine and N-(1-naphthyl) phthalimide, after eluting different volumes of samples with methanol, are shown in Table 1. The maximum suitable recovery volumes of these samples (low concentration for MS) were 750 mL. Larger volumes of low concentration samples cause losses of naptalam (breakthrough volume). The recoveries for naptalam, 1-naphthylamine and N-(1-naphthyl) phthalimide were found to be 97.4 ± 2.9, 96.0 ± 2.3 and 97.5 ± 2.8%, respectively. When sample volume is kept constant (1 L) and the concentrations of naptalam, 1-naphthylamine and N-(1-naph-

Table 1 The effect of sample volume on recovery of naptalam, 1naphthylamine and N-(1-naphthyl) phthalimide Sample volume/mL

250 500 750 1000 1250 1500

1000 ng L–1 Naptalam

1-Naphthylamine

N-(1-Naphthyl) phthalimide

95 ± 1 98 ± 2 99 ± 1 98 ± 1 97 ± 3 83 ± 3

94 ± 3 94 ± 2 98 ± 2 96 ± 2 90 ± 3 83 ± 1

89 ± 2 95 ± 1 97 ± 1 96 ± 2 89 ± 2 85 ± 3

182 Fig. 5 Thermal decomposition of naptalam in the GC split/splitless injector

Fig. 6 TIC chromatogram of PTV-GC MS analysis of naphthalene, 1-naphthylamine, N-(1-naphthyl) phthalimide

thyl) phthalimide are varied, the maximum suitable recovery concentration is 1.3 µg L–1. Programmable temperature vaporizer-GC MS A close similarity between electron impact ionization (positive ion) mass spectra of naptalam and its degradation products is observed. Gas chromatography is an ideal separation technique, but naptalam is an unstable compound at higher temperature. The GC inlet system temperature was the most critical part for this determination, as it has been previously stated [22]. When naptalam is injected into the GC split/splitless injector, appearance of two peaks in the total ion current (TIC) curve is observed, Fig. 5. The corresponding mass spectra of this chromatogram revealed the thermal decomposition of naptalam into phthalic anhydride and 1-naphthylamine; the thermal decomposition of the weak amide bond is probably due to the high temperature of the GC inlet system. In the PTV injector, thermal decomposition is minimized. It is due to the smaller residence time at higher temperature compared with the conventional GC split/splitless injection, Fig. 6. Consequently, all GC MS experiments were carried out at a PTV-GC inlet system.

To obtain the determination limits for low levels of naptalam and its degradation products, high volumes (100 µL) of river water should be injected into the PTV injector of GC. Determination of low concentrations of naptalam and its degradation products using PTV injector is based on selective elimination of water while simultaneously trapping these components in the PTV glass wool. The initial injector temperature, split time and rapid temperature programmed heating of the injector, for transfer of naptalam and its degradation products to the column, should be optimized and under the established conditions they are 110 °C, 3 s and 50 °C min–1, respectively. If the initial temperature and rate of heating are higher than these conditions, naptalam would degrade to phthalic anhydride and 1-naphthylamine. The LOD of this method using naphthalene as an internal standard were 17, 11 and 15 ng mL–1 for naptalam, 1-naphthylamine and N-(1-naphthy) phthalimide, respectively. The European Community directives dictate that the concentration of individual pesticides in drinking water should not exceed a maximum admissible concentration of 100 ng L–1 [17] and the present PTV-GC MS method is sophisticated enough to control this threshold value. The comparison of the NIMS, PTVGC MS and HPLC methods for determination of naptalam in aqueous media is shown in Table 2.

183 Table 2 Comparison of NIMS, GC-MS and HPLC methods for determination of naptalam Characteristic

NIMS

PTV-GC MS

HPLC

Mean (RSD) LOD (ng L–1)

95 (0.7) 230

81 (3.1) 2400

Linearity Time analysis Typical

7000 15 min detec. and deter.

96 (0.4) 17 (without preconcentration) 6000 25 min Not need preconcentration

2000 1h –

Sample spiked with 97 ng of naptalam in 1 L water, recovery of cartridge for HPLC 83.2%, membrane disc 99%

In contrast to HPLC analysis, NIMS and PTV-GC MS techniques show higher precision and are less time consuming. Due to overlapping when complex sample matrices are analyzed, PTV-GC MS, a chromatographic separation before MS analysis, should be preferred.

Conclusions This work shows that the analysis time, linearity and LOD of NIMS and PTV-GC MS analyses of naptalam and its degradation products are far better than the classical HPLC method. The sensitivity and convenient sample handling technique of PTV-GC MS is high. The combination of SPE using C18 membrane disc and PTV-GC MS can be very suitable for the determination of very low concentrations of naptalam and its degradation products.

References 1. Hoffman OL, Smith AE (1949) Science 109: 588 2. Tomlin C, The Pesticide Manual (tenth edition). Incorporating Agrochemical Handbook 3. Diaz TG, Acedo MI, de la Pena AM, Pena MS, Salinas F (1994) Analyst 119: 1151–1155 4. Granados A, Nassetta, M, de Rossi RH (1995) J Agric Food Chem 43: 2493–2496 5. Smith AE, Stone GM (1953) Anal Chem 25: 1397–1399 6. Diaz TG, Acedo Valenzuela MI, Salinas F (1994) Fresenius J Anal Chem 350: 692–701 7. Bellar TA, Budde W (1988) Anal Chem 60: 2076–2083 8. Butz S, Stan HJ (1995) Anal Chem 67: 620–630 9. Stan, HJ, Linkerhagner M (1996) J Chromatogr A 750: 369–390 10. Erdmann F, Rochholz G, Schuetz H (1992) Mikrochim Acta 106: 219–226 11. Worobey BL, Shields JB (1987) J Assoc Off Anal Chem 70: 1021–1024 12. Cairns T, Chiu KS, Navarro D, Siegmund E (1993) Rapid Commun Mass Spectrom 7: 971–988 13. Kargosha K, Ahmadi SH, Ghassempour A, Arshadi MR (1999) Analyst 124: 367–371 14. Rahman-Setayesh S, Arshadi MR (1993) J Sci I R Iran 4: 290– 296 15. Arshadi MR (1978) Org Mass Spectrom 13: 379–381 16. Arshadi MR (1974) J Chem Soc Faraday Trans I 70: 1569–1571 17. Mol HGJ, Janssen HGM, Cramers CA, Vresuls JJ, Brinkman UAT (1995) J Chromatogr A 703: 277–307 18. Ghassempour A, Arshadi MR, Adimi B (1998) Oriental J Chem 14: 287–291 19. Beltran J, Lopez FJ, Hernandez F (1993) Anal Chim Acta 283: 297–303 20. Balinova A (1996) J Chromatogr A 754: 125–135 21. Dupas S, Scribe P, Dubernet JF (1996) J Chromatogr A 737: 117 22. Wolfe MF, Seiber J (1992) Arch Environ Contam Toxicol 23: 137–141