Investigation of long-term hazards of chemical weapon agents in the environment of Sardasht area, Iran

The present study aimed to investigate the persistence and existence of chemical warfare agents (CWAs) and related dissipation products in the environment of Sardasht area, Iran. Three types of environmental samples including water, soil, and native local plant materials were collected and analyzed. Gas chromatography-mass spectrometry in the electron impact ionization mode has been developed for the separation, screening, identification, and qualification of chemicals after the sample preparation methods. The initial results revealed that no trace of related compounds or CWAs was detected in the soil and water samples. However, trace amounts of some degradation products of blistering agents like mustard gas (HD) and lewisite were found in a tree wood from a house subjected to chemical attack as well as in barley samples (a mixture of leaves and root) collected from an agricultural field in the area indicating chronic low exposure to the environment and people. In order to validate the applied extraction procedures, ethylene glycol was spiked to some of the samples including groundwater, surface soil, grape, and alfalfa plants. All the recoveries were in the range of 83.6–107.4% with the relative standard deviations varying from 4.9% to 12.4% (n = 3) successfully.


Introduction
In several cases, the laws of war have been broken by both parties leading the war to be directed toward civilians, women, and children or the use of mass destruction weapons against the military and civilians including chemical weapons, despite the use of which have been banned for decades under the Geneva Convention almost 100 years ago. After the war, the Chemical Convention has been established as a global convention with a control regime under the Organisation for the Prohibition of Chemical Weapons (OPCW) (Sydney 1972). Chemical war agents (CWAs) are strong chemicals that are used as mass destruction weapons. The chemical agents are used to kill, seriously injure, or disable people due to their physiological effects. The toxicity of these agents is because of their ability to interact with enzymes, proteins, and nucleic acids in living organs. The chemical agents first cause stimulation in cells and then affect the cells as a cellular toxin in living tissues (Singha et al. 2016). Compounds such as sarin, soman, tabun, mustard gas, and lewisite are among the chemical warfare agents, which are also known as blistering, blood, nerve, asphyxiation, and riot control agents (Hellström and Ödalen 2013;Hanaoka 2005). Bis(2-chloroethyl) sulfide (sulfur mustard, mustard gas, also known as HD) is a blistering agent widely used during WWI, WWII, and also the Iran-Iraq conflict in the 1980s by the Iraqi army. Mustards can be divided into sulfur mustard and nitrogen mustard. As the injuries caused by mustard agents resemble those of burning and blisters, they are categorized as "blistering agents." Nonetheless, it is more suitable for those to be known as "blister and tissue-destroying agents" as they cause severe eye and respiratory system injury and internal organ failure. Also, because of the ability of mustard to covalently bind with a variety of biological molecules, the resultant biological damage could have considerably delayed consequences, and the first symptoms can occur between 2 and 24 hours after the exposure (Hellström and Ödalen 2013). However, the mentioned issues are not the only destructive effects of sulfur mustard (Nilsson et al. 1992).
From the environmental point of view, most of the CWAs including mustard (HD) hydrolyze after solvation and produce degradable products such as thiodiglycol (TDG), 1,4thioxane, 1,4-dithiane, and 1,3-dithiolane. These compounds in aqueous media may act as a reliable proof of the original existence of HD. Moreover, some previous reports have proven the persistence of these substances for up to 60 years (Tang and Keong Loke 2012). Therefore, these materials and their decomposition products are expected to exist in the environment and be transferred from soil and water to plants and then to food chains and finally to animals and humans. Thus, sensitive and precise determination methods for environmental and biomedical studies are required to verify the presence of sulfur mustard and its decomposition, oxidation, and hydrolysis products (Tang and Keong Loke 2012;Deng and Evans 1997). However, despite the importance of this subject, the studies on the impact of chemical bombs onto the environment are very limited which can be due to the laboratory prohibition for working with CWAs or limited areas in the world in which these weapons have been used.
In this context, in 2005, Aldeen and Dlawer studied the long-term hazards of CWAs in the soil of Halabja in northern Iraq. Ten soil samples from contaminated areas and six samples from not contaminated areas were studied in this research. The presence of sulfur mustard or any other volatile agents in the samples was evaluated by gas chromatography-mass spectrometry (GC-MS) (Dlawer and Aldeen 2005).
In 2010, Torrea et al. examined the environmental hazards of chemical weapons discharged into the South Adriatic Sea and demonstrated the effect of chemical gases on fish in the Mediterranean Sea (Torrea et al. 2010).

About the study area
The city of Sardasht with an area of 144.454 km 2 is located in northwestern Iran and near the border of Iraq (Fig. 1). The altitude of this city is 1840 meters above sea level, and it has a population of about 50,000 people. The city has about 96 km of border with Iraq and is located in the south and southwest of the small Zab River Basin. Sardasht City has a mountainous climate and a Mediterranean rainfall regime (Khezri 2010). During the years of the Iraq and Iran conflict, the city of Sardasht was considered as a point of contact between the two parties, culminating in the chemical bombing, which is still taking its toll on the people after so many years. In July 1987, Saddam Hussein's military had attacked the city of Sardasht with seven chemical bombs, four of which hit the city center, and the other three landed in a nearby valley inside Rashaharameh village.
It is the first city in which the weapons of mass destruction and chemical weapons have been used against its civilians after World War II. According to the letter of the Ministry of Foreign Affairs to the United Nations, at least 4600 peoples were killed or severely injured in that bombing. However, some resources were estimated this number to be about 7000 people (Ahmadi et al. 2010).
Herein, gas chromatography-mass spectrometry (GC-MS) as one of the most popular and powerful instrumental methods was used for qualitative analysis of chemical warfare residuals through the comparison of obtained mass spectral data with authoritative known and analogous compounds available in the library (Hanaoka 2005;D'Agostino and Hancock 2003).
The aim of this work was to investigate the presence/ absence or stability of chemical weapon compounds and their decomposition by-products in soil (from representative attacked points), water (surface and groundwater), and the major consumer plants in the Sardasht region of Iran. Considering that stability of these constituents in the environment could be a possible threat to the health of residents, thereby, providing solutions for cleanup of contaminated areas is of great importance.
To the best of our knowledge, there has been no research onto the existence of chemical weapon residuals in the Sardasht area of Iran so far.

Sampling
In order to have a plenary investigation into the existence of chemical warfare agents and their related by-products, three types of environmental samples including two water and eight soil samples and eight local plant materials were selected and collected during a day from the study area. All samples were collected in July 2019 from eight sites (three sites inside the Sardasht City, four sites inside the Rashaharameh village, and one site between city and village as a control sample) that were exposed directly to the chemical attack bombardment. Figure 2 (a and b) shows Google Earth images of the study area that has been subjected to chemical attack, and Figure 3 shows a map of sampling stations in Sardasht City and Rashaharameh village, respectively. Two types of water including groundwater in Sardasht City (W1) and surface water from Rashaharameh village (W2) were collected in 250-ml amber glass bottles from the point near to the site of the attack. Care was taken to ensure that vials were filled full with no headspace and were quickly sealed with Teflon®-lined screw caps.
Sodium azide solution (10% W/V) was added in situ as a preservative to stop all enzymatic and degradation processes (Ahmadnezhad et al. 2021). Soil samples were collected with a grab or core sampler and then air-dried in the laboratory condition at room temperature for 48h before extraction (see Table 1). All the soil samples were homogenized with a manual pounder and sieved by a 2-mm sieve to remove any large particles (Rasoulifard et al. 2015). Plant samples including the mixture of leaves, roots, and pedicles were selected from some endemic local plants located near or at the center point of subjected chemical attack sites. Then, they were air-dried for 7 days through consecutive aerations.
To have homogenized fine plant powders leading to having effective extraction content, they were grinded completely with an electrical mortar and sieved through a 600-μm sieve, respectively. Some of the characteristics of selected plant samples are presented in Table 2. All the studied samples (water, soil, and plant) were stored at 4°C and extracted within 14 days after collection.

Water samples
In this research, liquid-liquid extraction (LLE) was recommended as a simple and cost-effective sample preparation technique for all the applied samples (Hanaoka 2005). For water specimens, the pH was checked and neutralized (pH~7) with diluted hydrochloric acid and sodium hydroxide solutions before extraction. Then, 10 ml of each sample was extracted with 2-ml dichloromethane as extraction solvent two times alternatively and shaked at 600 rpm for about 30 min. Then, the solvent was collected and dried up to 0.5 ml under a stream of nitrogen. The necessary dehydration with anhydrous sodium sulfate was done at the end of extraction before injection.

Soil samples
First of all, about 2 g of each soil sample was weighted and coded according to the original sample boxes. Since the nitrogen-/sulfur-containing compounds, phosphonothiolates, and corresponding alkylated or protonated salts so-called V agents are sometimes difficult to be efficiently extracted from soil specimens because of adsorbing to silicon hydroxide network (Hanaoka 2005), impregnation was done with 5 ml of 0.5 M potassium hydroxide/methanol for 3h as the first step of solvent extraction system. A sonication system with an ultrasonic disrupter of the horn type (300 watts power) was used to increase the transfer of the desired organic compounds to the solvent. After collecting the above solution (potassium hydroxide/methanol) in screw cap tubes, 5 ml of dichloromethane was added to the same samples and sonicated for about 30 min. Further extraction was repeated 2 more times with dichloromethane to have good recoveries. The organic phase was then collected through centrifuging and added to the previous solution.
All the tubes were centrifuged at 3500 rpm for about 5 min. Then, the total organic phase was transferred to another ultra clean tube and filtered through a 0.45-μm Minisart SRP 15 polytetrafluoroethylene membrane, dried into 0.5 ml under the nitrogen atmosphere, and used for injection.

Plant samples
Two grams of each dried and powdered samples were soaked with sort mixed of 10-ml ethanol and 10-ml distilled water and stored at room temperature for 48h. The extracts were then filtered through Whatman filter paper (No. 41) to remove any sediments and fine particles. A simple liquid-liquid extraction (LLE) method was used to extract semipolar and polar byproduct compounds (Gherman et al. 2000). The extraction procedure was as the following: An aliquot of 5.0-ml extraction solvent (ethyl acetate to nhexane to dichloromethane, 5:1:1, v:v:v) was mixed and agitated with the as-made hydro-alcoholic plant extract for about 30 min. Since the two phases could be observed, the mixture was centrifuged at 4000 rpm for 5 min. Finally, the organic phase including desired components was separated within a pipet pastor and dried under the nitrogen atmosphere into 1.0 ml and used for chemical derivatization for subsequent analysis with GC-MS. Figure 4 shows some of the sampling and cleanup procedure schematic images.

Chemical derivatization
Since CWAs are known as very reactive chemicals, which can react with the stationary phase of the chromatographic column in their underivatized forms, and also for having volatile compounds from nonvolatile degradation products of HD, a derivatization step was carried out after the extraction procedures of all types of samples (water, soil, and plant) and prior to GC-MS analysis (Matsuda et al. 1998). Herein, trimethylsilylation processing was applied as a most commonly used derivatization method that could sufficiently replace the active hydroxyl functional groups bonded to some degradation, remaining, and reaction products of mustard gas. Briefly, the extraction procedure was carried out as follows: 200 μl of TMCS and 80 μl anhydrous pyridine added to the extracts and sonicated in 50°C for about 60 min. The above solution was dried under a stream of nitrogen, and finally, the residual solution was dissolved in 100-μl dichloromethane before injection to GC chromatograph. With the simple mentioned method, all the polar hydroxyl, carboxylic acid, thiol, and amine functional groups could be successfully replaced with a TMS [-Si(CH 3 ) 3 ] group.

Gas chromatography-mass spectrometry
All the chromatographic analyses were carried out using an Agilent 7890B GC/5977A MS (USA) instrument, equipped with an HP-5MS (5% phenyl methyl siloxane) capillary column (60 m × 0.25 mm Id × 0.25 μm film thickness). Helium (99.999%) was used as the carrier gas with a flow rate of 1.1 ml/min. The injector temperature was set at 290°C. The oven temperature programs for all the target samples were set as following: The initial temperature of 50°C was held for 2 min, then increased to 180°C with a ramp of 15°C/min (held 5 min), then increased to the final temperature of 295°C with a ramp of 10°C/min, and held at 5 min. Samples were injected in the split mode with ratio of 1:10, and the total run time was 36 min. The GC-MS transfer line was held at 300°C, MS ion source temperature was set at 230°C, and MS single quadruple temperature was set at 150°C. The ion source was operated in the electron ionization mode (EI; 70 eV). The mass spectrometer was tuned by perfluorotributylamine (PFTBA) as a calibration substance once a week according to the instrumental manual. Full-scan mass spectra (45-600 m/z) were recorded for all types of species identifications.

Screening, identification, and processing of data
Data acquisition and interpretation on mass spectrum of GC-MS was conducted using the database of both National Institute Standards and Technology (NIST) and Wiley Registry of Mass Spectral Data, 6th Edition (Wiley Interscience, New York) with more than 140,000 patterns. One of the newest versions of Agilent commercial software package (MassHunter) was used for efficient interpretation and acquisition data too. In addition to identify related byproduct compounds in the studied environmental specimen, Retention Index (RI) values were used by comparing the obtained RI value with that of an authentic compound (Hanaoka 2005).

GC-MS analysis
Water Figure 5a and 5b shows the chromatograms of groundwater and surface water from the places subjected to the chemical attack, respectively. As can be seen, no organic-related compound was found fortunately in both water samples. This could be attributed to a phenomenon called self-purification and attenuation of water that progressively eliminates the additional organic contamination loaded due to any reason and leads the aquatic ecosystems to recover almost its original balance with the surrounding environment over the years (Vaezihir et al. 2020;Oliva González et al. 2014).

Soil
Eight soil samples were prioritized according to Table 2. The screening data from GC-MS analysis revealed that although some samples had trace amounts of organic compounds like insecticides and pesticides, no signs of chemical weapon agents were found in any of cases. This could be attributed to environmental soil bioremediation by heterotrophic Schematic images from steps of sample preparation procedure of water, soil, and plant species. A)sampling from river water, B) sampling from trunk of walnut tree, C) the collapsed building in the chemical attack that was sampled D)drying of the sampled soils E) preparing of the samples before extraction F) sampling from the extract of the soil samples microorganisms such as aerobic and anaerobic bacteria via biological processes. Therefore, decomposition and biodegradation of the impacted media could be occurred successfully (Certini et al. 2013). Since in this study soil samples were collected from topsoil (0-30 cm), leaching the organic contaminants by passing over the years deeper into the soil might be another reason to find no chemical warfare agents in the soil. Figure 6a and 6b shows chromatograms of samples S9 and S5 as the representatives of soil samples, respectively.

Plants
Since most of the chemical warfare residuals are classified into polar and semipolar decomposition products, a mixture of polar and nonpolar extraction solvents were used as described in "Extraction procedure." According to the GC-MS profiles of plant extracts (P1 to P10), widely interesting organic compounds were found in all the samples. Fatty acids as methylsilylated derivatized form, phenolic compounds, some of the antibacterial constituents, a group of antioxidants, and also some metabolites were a series of compounds identified in plant samples. In Fig. 7a and 7b, we can see GC-MS chromatograms of samples P5 (grape) and P1 (walnut) as the representative plant samples. As can be seen some from the characteristics of the identified compounds in Table 3, arsine oxophenyl (C 6 H 5 AsO) in retention time (RT) of 15.60 min and with a content of 0.55% was detected in sample P8, as a stable component. Some of the most frequent m/z ions were found as follows: 133, 119, 96, and 116. It could be originated from hydrolysis of phenylarsonic acid (C6H5AsO(OH) 2 ) which again can originate from lewisite as one of the utilized organoarsenic chemical warfare agents (Hatlelid et al. 2010;Sanderson 2011) (see Fig. 7c and d).
The presence of nonbonded pairing electrons onto the arsenic and oxygen atoms along with unstable phenylic Π electron pairs could lead to a high affinity between this volunteer composition and most of biochemical metabolites via electrostatic and covalent bonding. Since mustard gas (HD) was one of the major chemical weapons used in the study area, most emphasis was placed on its primary or multistage degradation products like thiodiglycol (TDG), 1-oxa-4,5dithiepane, and 1,2,5-trithiepane (Chmielińska et al. 2019). Herein, diglycolic acid, isobutyl octadecyl ester as its derivitized form in retention time of 18.11 min with content of 0.65% was found in sample P10 (see Fig. 7e and f). Some of the important m/z ions with the most abundance were found as follows: 191, 71.1, 57.1, and 85.1. It could be one of the probable toxic oxidized degradation metabolites of mustard agents (HD) like thioglycol acid methyl ester (TGM) and ethylene glycol monomethyl ester via releasing of thiol molecules. These two plant samples were of different sampling stations but from the center of chemical attack. So, of special importance is the matrix of the environmental case studies. Bioavailability and the ability of residual organic/inorganic compounds to be absorbed and taken up from contaminated soil into plant tissues so-called bioaccumulation especially by crop plants such as barley and wheat could lead to confirm the results found in P8 and P10 samples. This phenomenon depends on many environmental factors in terms of the effects of organic target on plant growth, shoot biomass, root development, or other physiological functions thereby helping environmental monitoring and risk assessment of organic residuals in the contaminated study area, respectively (Ait Ali et al. 2004).
Because that plant-soil system, has a unique ecotoxicological environment, the clear distinguish of whether soil or plant contamination is still ambiguous and needs more information or database too.

Validation test
In order to evaluate the validation of the sample preparation methods described above (see the "Extraction procedure" section), a recovery test was carried out onto different random studied samples (W1, S3, P7, P2). Ethylene glycol (EG) was utilized for this purpose owing to its similarity to thiodiglycol (TDG) and corresponding related hydrolysis products of mustard gas. It was spiked at two concentration levels of 1.0 mgL -1 and 0.5 mgL -1 to each selected samples before the main extraction procedure and performed in triplicate. The average recoveries ranged between 83.6 and 107.4% depending on the matrix complexity of the spices (the best recovery obtained from groundwater sample). Also, the RSD% (relative standard deviation) varied admissibly from 4.9% to 12.30% (n=3). The obtained results confirmed the suitability of the applied sample preparation methods to various environmental samples having a good accuracy and precision (see Table 4).

Conclusion
In this research, a perspective investigation onto the environment of Sardasht, Iran, was done successfully to evaluate the persistence/absence of chemical warfare agents. Water, soil, and plants were the target studied samples. Gas chromatography-mass spectrometry was used as a highly regarded instrumental analysis method due to its good resolution, sensitivity, and selectivity. Comparison of the mass spectral data obtained from specimens with those of the authentic compounds could give us an accurate impression to identify and qualify the chemical warfare agents. The results indicated that no trace of related chemical warfare compounds was The recovery percent of the spiked componds is 83 to 107 and standard devieation value is between 8 to 11 ecxept to a sample of groundwater. Both of this results showed that the analysis procedure is reliable found in the soil and water samples. However, a trace amount of arsine oxophenyl (0.55%), as one of the oxidation products of organoarsenic chemical weapon agents, was found in one plant sample (mixture of the barley organ from Rashaharameh village). Also, diglycolic acid, isobutyl octadecyl ester (0.65%) as one of the toxic degradation products of chemical warfare agents was identified in another plant sample (wood from a house attacked in Sardasht City). Due to the possible genotoxicity of compounds described in this paper, further toxicity data and evaluation of those findings need to be generated in the following studies. In fact, this paper provides a comprehensive environmental sampling of CWAs in the target study area and assessment of environmental risks near the sites subjected to chemical attack to date in the open scientific literature. Hence, this contribution could give us a long window of opportunity to study the other environmental contaminated areas. The objective of the next stage of the work is to address potential public health risks from potential exposure to chronic low exposures to chemical warfare-related compounds via the environment. Of course, it should be noted that the opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Army or the Department of Defense respectfully.