Trace determination of sulphur mustard and related compounds in water by headspace-trap gas chromatography–mass spectrometry

Abstract

A method for trace determination of sulphur mustard (HD) and some of its cyclic decomposition compounds in water samples has been developed using headspace-trap in combination with gas chromatography–mass spectrometry (GC–MS). Factorial design was used for optimisation of the method. The trap technology allows enrichment and focusing of the analytes on an adsorbent, hence the technique offers better sensitivity compared to conventional static headspace. A detection limit of 1 ng/ml was achieved for HD, while the cyclic sulphur compounds 1,4-thioxane, 1,3-dithiolane and 1,4-dithiane could be detected at a level of 0.1 ng/ml. The method was validated for the stable cyclic compounds in the concentration range from the limit of quantification (LOQ: 0.2–0.4 ng/ml) to hundred times LOQ. The within and between assay precisions at hundred times LOQ were 1–2% and 7–8% relative standard deviation, respectively. This technique requires almost no sample handling, and the total time for sampling and analysis was less than 1 h. The method was successfully employed for muddy river water and sea water samples.

Introduction

One of the most employed chemical warfare agents (CWA) in history is the skin damaging agent bis(2-chloroethyl) sulphide (sulphur mustard, with military designation HD). HD was frequently used in World War I, and more recently in the Iran–Iraq war and during the campaign against the Iraqi Kurdish population in 1987–88 [1]. Today, one of the concerns related to HD is the large amount of sea dumped or abandoned weapons from World War II (WWII). In the Baltic Sea and along the coast of Japan, fishermen have snared mustard agents from old artillery with their nets, and been injured from contact with the agents [2]. In China, large amounts of abandoned CWA, including HD, were left behind during Japanese retreat in the closing stages of WWII. It has been estimated that abandoned CWA in China have caused two thousand casualties or fatalities since the end of the war [3]. After the Chemical Weapons Convention (CWC) entered into force in 1997, the production, storage and use of CWA have been prohibited [4]. In light of the CWC and the environmental concerns of abandoned CWA, the need of sensitive analytical methods for determination of HD and related compounds has increased.

HD has low aquatic solubility (1 g/l) and freezes at 14 °C [5]. The hydrolysis of HD in larger lumps is slowed down or completely prevented by formation of oligomeric and polymeric layers of the degradation products [6]. Hence, HD in sea dumped munitions can stay intact at the sea bed for a long time after the artillery shell is corroded. When dissolved in water, HD hydrolyses to a set of sulphides, disulphides, sulphoxides, sulphones, and thiols [7]. In addition, munition grade HD often contains impurities that survive in the environment longer than the agent itself. Thus, determination of several common degradation products and impurities in aqueous samples may act as a reliable proof of the original existence of HD. This study includes determination of HD, two of the most common degradation products 1,4-thioxane and 1,4-dithiane [7], [8], [9], and 1,3-dithiolane. The latter has been found in water and soil samples near an old destruction site of HD [10], [11].

Gas chromatography–mass spectrometry (GC–MS) has been extensively used for the identification of HD and related compounds in environmental samples [12], [13], [14], [15]. For determination of CWA in aqueous samples, a recommended protocol from sample treatment to final instrumental analysis is available [16]. Liquid–liquid extraction (LLE) or solid phase extraction (SPE) is recommended as sample preparation techniques. It has been shown that HD could be determined in aqueous samples with the LLE procedure [11], and with the SPE procedure [17] followed by GC–MS analysis. For the determination of non-volatile degradation products of HD, a derivatisation step must be included prior to the GC–MS analysis [18], [19]. Recently, several microextraction techniques have been applied for HD determination in water, like the single drop microextraction (SDME) [20] and hollow fibre-mediated liquid-phase microextraction (HF-LPME) [21], [22] followed by GC–MS analysis. The solid phase microextraction (SPME) technique, combined with GC and flame ionisation detection (FID) has also been used [23]. For the water-soluble non-volatile degradation products, liquid chromatography (LC) and MS with atmospheric pressure chemical ionisation (APCI) [24], [25] or electrospray ionisation (ESI) [26] has gained increasing utility. Microcolumn LC with MS or flame photometric detection (FPD) with large volume injection and peak compression has also been applied for the water-soluble degradation products [27], [28].

The headspace (HS) extraction and sample introduction technique has not been used extensively for determination of HD and related compounds. However, Wils et al. have employed dynamic HS followed by GC–MS to determine the hydrolysis product bis(2-hydroxyethyl) sulphide (TDG) in water and urine, where TDG was converted to HD prior to the determination [29], [30]. In addition, HS has been used in combination with SPME for determination of HD and related compounds in soil, where water was added to the soil to form a slurry prior to analysis [10], [31]. In the present study, the headspace-trap (HS-trap) technique in combination with GC–MS is applied for the first time for determination of CWA. The HS-trap technique patented by Tippler and Mazza [32] is an enhanced static headspace system which was commercialised in 2004. The technique has shown a great potential for determination of various volatile organic compounds in water [33], [34], [35].