Toxicity Assessment of Phosphonium Based Ionic Liquids Towards Female Guppy Fish

In this study, two phosphonium based ionic liquids (butyl triphenyl phosphonium chloride and hexyl triphenyl phosphonium bromide) have been synthesized using quarternisation process. The toxicities of these Ionic Liquids (ILs) are unknown and may be harmful to humans and the environment. Therefore, the toxicity assessment of these ILs was carried out according to the Organization for Economic Cooperation and Development (OECD) 203 guideline using female guppy fish (Poecilia reticulata). The median Lethal Concentrations (LC50) have been estimated for butyl triphenyl phosphonium chloride and for hexyl triphenyl phosphonium bromide to be 73.35 and 61.36 mg L −1 respectively. Both LC50 obtained can be identified as slightly toxic ILs based on Acute Toxicity Rating Scale by Fish and Wildlife Service (FWS). The findings from this study can be used for better design of phosphonium-based ILs with consideration of their aquatic toxicities.


INTRODUCTION
The world of Ionic Liquids (ILs) started by Walden (1914) who synthesized ethylammonium nitrate [EtNH 3 ] [NO 3 ]. Later, after several years, ILs began to gain more of the attention of scientists when first room temperature IL with a 1-alkyl-3-methylimidazolium cation was reported (Wilkes et al., 1982). ILs are composed of ions only and they are fluid below 100°C due to asymmetry of one of the ions. They are thermally stable with liquid range up to 300°C compared to 100°C for water and show very low vapour pressure, which makes them of great interest due to these properties. ILs typically consist of bulky organic cations, such as imidazolium, pyridinium, ammonium, phosphonium, paired with various anions, such as hexaflurophosphate (PF 6 -), bromide (Br -) Freire et al., 2006;Flieger, 2011;Peric et al., 2012). They are currently considered as a remarkable class of solvents due to their large liquidous range and non-volatile behaviour and can be used for a broad range of industrial applications. Research on this area is rapidly increasing due to their exciting applications and wide usages. The reason for the increasing interest is clearly due to the realization that these materials have a greater utility as solvents for reactions and material processing, media for extraction and working fluids for mechanical applications. Volatile Organic Compounds (VOCs) were previously used as solvents (such as, methanol, ethanol, benzene, toluene, ether) in industrial applications causing a major concern in the current chemical processing industry. Although intensive information and data regarding ILs' physical and thermodynamic properties have been reported and continuously published, only limited data with regard to the toxicity and ecotoxicity were reported (Bernot et al., 2005a;2005b;Latala et al., 2005;Pretti et al., 2006;Madria et al., 2013). The toxic effects of many solvents combined with serious environmental issues, such as Science Publications AJES atmospheric emissions and contamination of aqueous effluents are making them harmful to use. Thus, many researchers have emphasized on the improvement of green engineering which represents research aimed at finding environmentally benign alternatives to harmful chemicals. Among the neoteric solvents applicable in green technologies, ILs have garnered increasing attention over the others due to their numerous applications in recent years. In general, the toxicity of chemicals can be measured using (a) mammalian acute toxicity test (b) bacteria acute toxicity test (c) fish acute toxicity test and (d) biological dissociation test (Ardali, 1990;Dube and Hosetti, 2010). The aquatic toxicity assessment is the study of the effects of the chemicals and other anthropogenic and natural materials and activities on aquatic organisms at various levels of organization and ecosystems. The effects can cause both positive and negative alteration from previously existing circumstances, but aquatic toxicology focuses primarily on the deviations that are considered to be adverse in nature and on recovery processes in biota that may occur when the exposure diminishes. Adverse effects at the organismal level include both the short-term (acute) and long-term (chronic) lethality (expressed as mortality or survival) and sub-lethal effects such as changes in behaviour, growth, development, reproduction, detoxification activities and tissue structure. Adverse effects at the sub-organismal level include induction or inhibition of enzymes and/or phenotypes, as well as changes in the number, relative abundance and physiological condition of species typically found in a given community type (Rand and Wells, 1995). There are several test species that are recommended by the OECD guidelines, which are rainbow trout (Oncorhynchus mykiss), fathead minnow (Pimephales promelas), bluegill (Lepomis macrochirus), zebrafish (Danio rerio), medaka (Oryzias latipes), guppy (Poecilia reticulata) and common carp (Cyprinus carpio). A number of acute toxicity studies have been performed using a variety of fish. For instance, McKim (1977) used Chinook salmon (Oncorhynchus tshawytscha) for chronic toxicity evaluation. Baser et al. (2003) investigated the acute toxicity of permethrin (a synthetic pyrethroid pesticide) on guppy fish (Poecilia reticulata). Pretti et al. (2006) assessed the acute toxicity of several ILs' towards zebrafish (daniorerio). Sarikaya et al. (2007) used guppy fish to estimate the toxicity of Wang et al. (2010) estimated the toxicity of 1-methyl-3-octylimidazolium bromide ([C 8 mim][Br]) using goldfish (Carassius auratus). Dumitrescu et al. (2010) evaluated the acute toxicity of tetrabuthylammonium bromide using zebrafish. Li et al. (2013) employed goldfish toxicity to evaluate the toxicity of 1-methyl-3-octylimidazolium bromide ([C 8 mim] [Br]). Most researchers in recent years have carried acute toxicity tests for at least 48 h and usually 96 h. For more information regarding toxicity assessment using guppy fish (Poecilia reticulata), readers can refer to PAN (Pesticide Action Network) Pesticides Database (http://www.pesticideinfo.org) and ECOTOX database (http://cfpub.epa.gov/ecotox/). The recent searching (in 2012) with "Guppy (Poecilia reticulata) toxicity studies" gave around 4104 individual studies on this species with different chemicals, which is even greater than with goldfish, Carassius auratus (3290). Other studies can be found with zebra fish, Danio rerio (4658), bluegill, Lepomis macrochirus (9677), rainbow trout, Oncorhynchus mykiss (26662) and fathead minnow, pimephales promelas (15128). For more detailed information on the acute toxicity tests of different fish species, readers are directed to the summary tables provided by the U.S. fish and wildlife service reports (Wood, 1953;Hollis and Lennon, 1954), the handbook of acute toxicity of chemicals to fish and aquatic invertebrates (Waynon and Finley, 1980) and work of Baser et al. (2003).
In this study, the acute fish toxicity study was conducted according to the prescribed guideline OECD 203 (Organisation for Economic Cooperation and Development) (OECD, 1992) using freshwater Guppy fish (Poecillia reticulata) to estimate the toxicity of two phosphonium ILs. To our knowledge there is no published literature regarding the toxicity (LC 50 ) of these ILs.

Chemicals
Chemicals used in this study were obtained from commercial sources and had been used without drying and further purification. All ILs' were synthesized according to established methods (Bradaric et al., 2003;Cieniecka-Roslonkiewicz et al., 2005). An equivalent of tertiary phosphine was added dropwise to the appropriate alkylating reagent under nitrogen and stirred at a temperature rising from 30 to 200°C. The reaction time normally varies from 10 to 29 h. The speed of adding the Science Publications AJES tertiary phosphine to the alkylating reagent can be adjusted until no free tertiary phosphine can be formed in the reaction mixture. After the addition was complete, the mixture was stirred for 2 to 3 h at the highest temperature reached (200°C).
For the synthesis of triphenyl-n-hexyl phosphonium halides, the general procedure started using a round bottom flask with side arm adaptor and septum, which was dried over high vacuum and purged three times with nitrogen. Then, it was charged with triphenyl phosphine (2 mL, 5.79 mmoL).
The flask was subsequently immersed into an oil bath (50°C), the appropriate haloalkane was added via syringe and the mixture was stirred at 50°C, before the product was dried at high vacuum overnight. The general structures of these ILs' are shown in Fig. 1.

Fish Acute Toxicity Test
The species used for this study is female guppy fish (Poecilia reticulata). The fish were purchased from a fish tackling shop in Perak, Malaysia and brought to the laboratory within 15 min of purchase in plastic bags containing sufficient amounts of water for sustainability. Upon arrival at the test site, the fish were placed in plastic tanks for acclimation. Prior to the toxicity test, the fish were held in the laboratory for 12 days. The fish were fed twice daily, once in the morning and once in the evening. The fish were kept under normal laboratory illumination with a daily photoperiod of 12-16 h. The temperature of the water was between 23±2°C. The dissolved oxygen and pH of the water were 5-7 ppm and 7 respectively. The fish were observed carefully every day for signs of disease, stress, physical damage and mortality. Dead and abnormal specimens were removed immediately upon observation. After the adaptation period was complete; a group of 10 healthy fish were selected randomly and placed into 6.5 L plastic tanks provided with electric air pumps. Healthy fish can be recognized by their normal swimming style. Fish swimming abnormally should not be selected as they may have a weak resistance to the chemicals used in testing. Initially the experiments were conducted using tap water. Unfortunately, the results were rejected due to an abnormal death rate of the fish. Therefore, lake water was then used to avoid the chlorine contamination of tap water. Each tank was equipped with 5 L of lake water. No food was provided for the fish during the test. The weight of each fish is approximately 0.5 g. The acute fish toxicity test was performed according to the OECD standard methods (OECD, 1992). Each IL was tested for four concentrations. Different concentrations were added to each testing tank. Behaviors of the fish were monitored closely and dead fish were removed immediately. The number of dead fish for each concentration was recorded after 24, 48, 72 and 96 h. The median Lethal Concentration (LC 50 ) that is the ILs'concentration in water which kills 50% of the test batch of fish within a continuous period of exposure of 24, 48, 72 and 96 h. were calculated. The results from the experiment were compared with acute toxicity rating scale provided by U.S. Fish and Wildlife Service (USFWS) ( Table 1).

RESULTS AND DISCUSSION
The screening tests for the two ILs were conducted with the concentration of 100 mg L −1 as limit test, which showed the range of toxicity below 100 mg L −1 and it lies between 50 to 100 mg L −1 . The observations of the fish mortality for butyl triphenyl phosphonium chloride and hexyl triphenyl phosphonium bromide are presented in Table 2 and 3.
It can be clearly seen that at higher concentrations, the number of dead fish increased. Numbers of dead fish in the time period prescribed by the test were used to determine the mortality percent and lethal concentration (LC 50 ). The cumulative percentage mortality for 96 hr exposure time is plotted against a logarithm concentration ( Fig. 2 and 3).
Data presented in Table 4 and 5 were used to determine LC 50 values employing probit analysis (Finney, 1978). The best fit dose-response linesfor butyl triphenylphosphonium chloride and hexyl triphenylphosphonium bromide were plotted in Fig. 4 and 5 respectively. Transformation of the toxicity curve to a straight line indicates that the lethal threshold concentration is estimated correctly. The fact was confirmed by Burdick (1957). Probit equations were derived from the experimental data presented in Table 4 and 5 for both ILs'; which are Y = 9.95×13.56 and Y = 3.63 × -1.49 respectively. The correlation coefficients, R 2 were obtained 94 and 99% respectively. This is confirming the accuracy of the results obtained from the experiment. Probit equations obtained from this work can be used to examine the toxicity of any ILs' from the same family without the need for experimental work. It should be noted that these two equations are only valid to estimate the LC 50 for the same aquatic organism (Guppy fish).     LC 50 values have been obtained from Fig. 2 and 3 or from Fig. 4 and 5 to be 67.5 mg L −1 for butyl triphenylphosphonium chloride and 56.3 mg L −1 for hexyl triphenylphosphonium bromide.

AJES
According to the acute toxicity rating scale provided by the U.S. Fish and Wildlife Service (USFWS), these values lie within the range of slightly toxic.

CONCLUSION
Acute toxicity method was used to evaluate the LC 50 of twophosphonium based ionic liquids. The acute fish toxicity test was performed using Guppy fish (Poecilia reticulata) according to the OECD 203 standard methods. Results obtained from this work indicated that the butyl triphenyl phosphonium chloride and hexyl triphenyl phosphonium bromide are slightly toxic. The result from this study can be beneficial to industrial people who use phosphonium based ionic liquids in their industrial processes.

ACKNOWLEDGMENT
The researchers acknowledge the Petronas Ionic Liquid Center (PILC) and Chemical Engineering Department, Universiti Teknologi PETRONAS for their support in conducting this study. Mohanad El-Harbawi extends the appreciation to the Deanship of Scientific Research at King Saud University for funding this study through research group no RGP-VPP-303.