Many faces of arsenic

For centuries, arsenic has gained a negative reputation as “the poison of kings” or “king of poisons” because it is colorless, odorless, tasteless and difficult to detect even after death (Antman 2000; Frith 2013; Akhtar et al. 2017). The first medical reports on the effects of arsenicals were written by Hippocrates (460–357 BC), Aristotle (384–322 BC) and Pliny the Elder (23–79 AD). One of the earliest documented cases of arsenic poisoning was that of Britannicus who was poisoned by the psychopathic murderer, emperor Nero, to secure his Roman throne in 55 AD (Nriagu 2002). Later, Rodrigo Borgia better known as Pope Alexander VI (1492–1503) and his son, Caesar Borgia, became truly legendary in the history of the world when they assassinated a number of wealthy cardinals and accrued in this way a great fortune (Frith 2013). The next infamous poisoner was Catherine Deshayes (1640–1680), who was sentenced to death by a judicial commission (established by Louis XIV) for the murder of more than 2000 infant victims (Klimczak 2016). Famous victims of arsenic poisoning are King George III of Great Britain and the American explorer Charles Francis Hall. Even Napoleon was rumored to have died of arsenic poisoning on St. Helena Island, although the debate is contentious and continues to this day (Antman 2001; Klimczak 2016).

Arsenic was the favorite poison in the 19th century France and Great Britain, becoming part of the social and political life. Until 1851, there were no legal restrictions on the sale of arsenic (Bartrip 1992; Frith 2013). Nearly everyone could buy and sell any poison including arsenic, which was also readily available and very cheap. The official statistics in the United Kingdom (1839–1849) indicate that 239 people were tried for murder or attempted murder by poison (Bartrip 1992). Fortunately, the popularity of arsenic poisoning rapidly decreased in 1830, since the development of a chemical test for detecting arsenic in human tissues by the British chemist James Marsh (Magdalan 2007).

Arsenic has been used in medicine as a therapeutic agent. Ancient Indians, Greeks, Romans, Arabs and Chinese employed arsenic in their medicinal treatments for more than 2400 years (Akhtar et al. 2017). Hippocrates used arsenic sulfides, realgar and orpiment to treat ulcers, while Dioscorides used orpiment as a depilatory agent. In Latin culture, the first pharmacological experiment was initiated in the 15th century by William Withering, who became a strong proponent of arsenic-based therapies. In 1786, Fowler’s solution (As₂O₃ in potassium bicarbonate solution; 1% wt/vol.) was discovered and later used for treating various diseases, including malaria, syphilis, asthma, chorea, eczema and psoriasis (Antman 2001). Since the 1880s, pharmacological texts have described the use of arsenical pastes and arsenous acid, the former for treating skin and breast cancers, and the latter for the treatment of hypertension, bleeding gastric ulcers, heartburn, chronic rheumatism and promyelocytic leukemia (APL) (Antman 2001; Akhtar et al. 2017). In the 1990s, scientists from China presented the results of clinical studies on the use of trioxide arsenic (ATO) in the treatment of acute promyelocytic leukemia and myelodysplastic syndromes (Akhtar et al. 2017). As a result, ATO has become a more effective treatment of APL than other, newer medications and has less severe adverse reactions and greater safety.

Research aimed at locating tumors using arsenic-74, a positron emitter, has recently been reported. Arsenic trioxide has demonstrated efficacy and safety in patients with first and subsequent relapses or refractory promyelocytic leukemia (APL), regardless of the disease-free interval. The research results showed the complete remission in 87% of patients, and molecular remission in 83% (Douer et al. 2003). Clinical trials are currently underway, using arsenic-based medicines for the treatment of hematopoietic and lymphocytic proliferative diseases, not only as monotherapy, but also in combination with other compounds, including retinoic acid, ascorbic acid, and GM-CSF – a growth factor (Izdebska et al. 2008; Wu et al. 2016).

Arsenic is obtained as a by-product of the smelting of copper, lead, cobalt, and gold ores (Brininstool 2017; Li et al. 2018). In 2016, the world’s leading producers exported the following approximate amounts of arsenic compounds (in metric tons): 25 000 to China, 7000 to Morocco, 1500 to Russia, 1000 to Belgium, 50 to Bolivia and 45 to Japan. This means that the estimated global production of As was 36 500 metric tons (t) (Brininstool 2017). Currently, China and Morocco are the world’s leading producers of arsenic trioxide, accounting for 87% of the estimated world production. In China, in addition to reclaiming arsenic as a by-product of nonferrous smelting, arsenic is recovered as a by-product of gold mining from orpiment (As₂S₃) and realgar (AsS), which are more common ore minerals of arsenic. In 2010, the price of arsenic reached $3.20 per kg (Walker 2010). The price is expected to rise in the future. This means that arsenic production is still a profitable business.

Since the 1970s, the demand for arsenic has been growing, mainly in response to its increased use in the grids of lead-acid batteries. In 1974, the price peaked at nearly $2.00 per pound (Brooks 2013). However, its major production (about 70% of the global arsenic production) was intended for the preparation of arsenical pesticides. As a result, organic arsenical pesticides such as: monosodium methane arsenate (MSMA) – HAsO₃CH₃Na, disodium methane arsenate (DSMA) – Na₂AsO₃CH₃, dimethylarsinic acid (cacodylic acid) – (CH₃)₂AsO₂H, arsenic acid – H₃AsO₄ have been released into the marine environment and some of them are transported, modified or even absorbed on sediments/organic matter (Panagiotaras & Nikolopoulos 2015). These organic compounds are resistant to environmental degradation through chemical, biological and photolytic processes. Moreover, due to their persistence and ability to accumulate in living organisms, they constitute a dangerous group of chemical waste (ATSDR 2007).

Currently, the most recognized arsenical pesticide is chromated copper arsenate (CCA). This compound was widely used for many years as a wood preservative. However, exposure to arsenic leached from CCA-treated wood caused the serious health problems. In 2003, CCA was replaced in industrial applications by new alternative wood preservatives, including alkaline copper quaternary, ammoniacal copper quaternary, ammoniacal copper zinc arsenate, copper azole and copper citrate (Brininstool 2017). Presently, the production of some arsenic agents is prohibited or strictly limited; the range of restrictions depends on the requirements of a given country and the usage of arsenic compounds for domestic consumption.

Arsenic warfare agents (CWA) belong mostly to the arseno-organic group and are characterized by high affinity for the sulfhydryl groups and high toxicity (Firth 2013). The CWA irritate the mucous membranes of the eyes, nose and throat. They cause tearing, coughing, sneezing, pain in lungs and breathing difficulties.

The following popular warfare agents based on arsenic compounds exist: Clark I (diphenylchloroarsine), Clark II (diphenylcyanoarsine), Adamiste (diphenylaminechlorarsine), Lewisite (Dichlor (2-chlorvinyl)-arsane) and Dick (ethyldichloroarsine) (Radke et al. 2014; Bełdowski et al. 2016a). The first modern arsenic-based chemical agent, i.e. Dick, was invented in 1881 by LaCoste. In 1918, two organic arsenical compounds, Lewisite and Adamsite, vesicant and respiratory irritant agents, were developed by the U.S. Army (Firth 2013). Clark I and Clark II were developed in German laboratories in 1917.

Although arsenic-based agents were produced on a large scale, generally they have never been used much in combat. The Chemical Weapons Convention (CWC) on the Prohibition of the Development, Production, Stockpiling and Use of Chemical Weapons and on their Destruction was established on 29 April 1997. The Convention has been ratified by 192 countries (Arms Control Association 2018). The global campaign of chemical weapons destruction is still in process.

At present, arsenic-based munitions dumped on the seabed is a serious problem. Shortly before the end of War World II (WWII), the Nazis started the process of dumping weapons on the seafloor, mostly in the southern part of the Baltic Sea. Based on the agreement signed in Potsdam on 17 July 1945, the Allies continued the dumping of stockpiles collected during WWII. Consequently, dangerous conventional as well as unconventional weapons were dumped on the seafloor at 300 dumpsites all over the world, including the Atlantic, the Pacific and the Indian Ocean, the east and west Canadian and U.S. coasts, the Gulf of Mexico, the coasts of Australia, New Zealand, India, the Philippines, Japan, Great Britain and Ireland, as well as the Caribbean Sea, the Black Sea, the Red Sea, the Baltic, the Mediterranean and the North Sea (Radke et al. 2014; Bełdowski et al. 2018). The issue of corroded chemical munitions on the seabed has attracted public attention and appears not to be completely resolved.

The water system is enriched with arsenic from anthropogenic (e.g. runoff contamination) as well as natural sources (e.g. geological contribution). The transport and partitioning of arsenic in water depend on its chemical form as well as interactions with other materials present. All soluble forms move with water and can be transported over long distances via rivers. However, arsenic can also be adsorbed from water onto sediments and soils, particularly clay, iron oxides, aluminum hydroxides, manganese compounds and organic material (ATSDR 2007).

The form in which this element exists in the marine ecosystem depends on biotic and abiotic processes, where pH values and redox potential play an essential role. Arsenic in the water system (i.e. freshwater, groundwater and seawater) is mainly represented by two inorganic species, i.e. arsenite – As(III) and arsenate – As(V). Inorganic As(V) is a predominant form under oxic conditions and dominates over arsenite As(III). As a result, the ratio of As(V) to As(III) ranges from 10 to 100 µg l⁻¹. At pH values of 7.5–8.2, arsenate is an anion and occurs as HAsO₄⁻² and H₂AsO₄⁻, while arsenite is present as a neutral species (H₃AsO₃) (ATSDR 2007; Flora 2015). Under reducing conditions at pH below 9.2, the uncharged arsenite species H₃AsO₃⁰ (Smedley & Kinniburgh 2002) predominates, but when pH exceeds the value of 12, the form of HAsO₃^2– occurs in water. As reported by Li et al. (2018), the percentage contribution of As(III) reduced from As(V) can reach > 80% of total As(III) in most hypoxic zones. This applies in particular to seawater from the seabed of the Baltic Sea, the Black Sea and the Changjiang Estuary.

The oxidation of As(III) by O₂ is a long and slow process. The organic species (i.e. monomethylarsonic acid MMA and dimethylarsinic acid DMA) naturally occur in seawater, however, they are presented at lower concentration levels (Panagiotaras & Nikolopoulos 2015). In seasonally anoxic estuarine water, variations in the relative proportions of As(III) and As(V) can be large. In marine and estuarine waters, organic forms are usually less abundant, but are often detected and depend on temperature and variations in aquatic biota (Flora 2015).

As(V) predominates also in aerobic sediment, while As(III) dominates in anaerobic conditions. Arsenic(III) partitions into the aqueous phase in anoxic environments. Unlike As(III), As(V) usually remains bound to minerals, such as ferrihydrite and alumina, which reduces its mobility and bioavailability (ATSDR 2007). Clay minerals (kaolinite, montmorillonite and illite) have greater affinities for As(V) than As(III) (Panagiotaras & Nikolopoulos 2015).

At the maximum range of adsorption (from 5.5 to 7.5), a higher adsorption occurs for As(V) than As(III), because the positively charged particle edges have more propensity for adsorbing HAsO⁰ (pK= 2.20) ⁻ ₃⁰ ₄₁ and H₂AsO₄(pK₂=6.97) than H₃AsO₃(pK₁= 9.22) (Lin & Puls 2000). However, illite shows slightly greater adsorption of As(III) compared to kaolinite. At pH <9, goethite hydroxide and iron hydroxide have higher adsorptive capacity for arsenite compared to arsenate. Moreover, carbonate species (e.g. bicarbonates) present in the aqueous environment may affect some sorption processes and lead to a decrease in arsenic adsorption onto clay minerals (Lenoble et al. 2002). Furthermore, arsenic in sediments can be released back into the water through chemical and biological interconversions of arsenic species. In this way, arsenic in the marine environment becomes again available for living organisms. Therefore, it should be considered as a potential threat.

Due to the fact that arsenic occurs in over 200 minerals, arsenic-rich minerals can be one of the main natural sources of this element in the marine environment. Other natural sources responsible for the release of arsenic into the marine environment include: volcanic eruption, rock erosion and forest fires (Smedley & Kinniburgh 2002; Flora 2015; Li et al. 2018).

Air is one of the main media for chemical and biological reactions, transport and circulation through soil to the marine environment. The concentration of As measured in remote or rural areas ranges from 0.02 to 4 ng m⁻³, whereas in urban areas varies from 3 to 200 ng m⁻³ (Chappell et al. 2001; Panagiotaras & Nikolopoulos 2015). Higher concentrations (> 1000 ng m⁻³) occur in the vicinity of industrial sources (WHO 2011).

Arsenic concentrations in uncontaminated soil are generally in the range of 0.1 to 40 μg g⁻¹. However, higher levels, i.e. 100–2500 μg g⁻¹, were observed in the vicinity of copper smelters and in soil contaminated with pesticides (WHO 2000; Litynska et al. 2017).

Ocean water contains arsenic at a concentration level up to 2 μg l⁻¹, while open seawater – from 1 to 3 μg l⁻¹ (Panagiotaras & Nikolopoulos 2015). However, in the case of a semi-closed area such as the Baltic Sea, the arsenic concentrations may depend on the distance from a shore, the location of point sources of pollution, atmospheric distribution and specific activities at sea (e.g. shipping, dredging, dumping). For example, Truus et al. (2007) reported total arsenic concentrations between < 0.1 and 1.75 μg l⁻¹ in seawater samples from the highly industrial area of Tallin (Estonia). Values between 0.49 and 1.10 μg l⁻¹ of As(V) + As(III) were found in samples collected from the region between Kiel Bight and the northern Gotland Sea, where hypoxia in deep water contributed to the increase in As(III) concentrations. Furthermore, the near-bottom seawater from the chemical weapon dumpsite in the Bornholm Basin contained about 1.04 μg l⁻¹ of As(V) + As(III) (Li et al. 2018). Another report from the Bornholm chemical weapon dumpsite showed that the total concentration of As was between 0.55 (off the dumping site) and 0.76 (the central part of the dumping site) (μg l⁻¹) (Khalikov & Savin 2011).

More variable conditions occur in marine estuaries, where the level of As is assessed at less than 4 μg l⁻¹. This is due to varying river inputs, changes in salinity and redox gradients or some anthropogenic discharges from terrestrial sources (Smedley & Kinniburgh 2002; Flora 2015). However, the concentration of arsenic in the estuarine water can be even higher (from 12.7 to 17.0 mg l⁻¹), when it is strongly affected by soil enriched with acid sulfate (AS). This applies in particular to water samples from the estuary of the Vörå River (Nystrand et al. 2016). The lower concentration of As (between < 2 and 4.9 μg l⁻¹) was found in water samples from the estuary of the Tinto River (southern part of Spain), though this region is highly polluted with acid lixiviates from old sulfide mines (Hierro et al. 2014). In the Mahanadi estuary (India), As (8.0 μg l⁻¹) comes from both natural and anthropogenic sources. In addition, arsenic was partially removed from water in the mixing zone (Mandal et al. 2016). In water of the Taehwa River estuary (South Korea), As was found mainly near urban and industrial areas and it occurred primarily in the As(V) form (94 μg l⁻¹) (Hong et al. 2016) (Table 2).

Panagiotaras & Nikolopoulos (2015) reported that the concentration below 10 μg l⁻¹ is mainly addressed towards natural waters of rivers and lakes. However, several studies addressing natural and anthropogenic sources have shown higher levels of arsenic (μg l⁻¹) in freshwater, e.g. up to 20.0 (Po River, Italy), 13.31–41.53 (Karnaphuli River, Bangladesh), 26.70–68.30 (Carmo River, Brazil), 35–157 (Manchar Lake, Pakistan), 58–413 (Chasicó Lake, Argentina) (Arain et al. 2009; Chetia et al. 2011; Varejão et al. 2011; Marchina et al. 2015; Ali et al. 2016; Puntoriero et al. 2014) (Table 3). These levels often exceed the tolerance limits for drinkable water defined by the European Council Directive (Council Directive 98/83/EC 1998) and by the World Health Organization as well as the Environmental Protection Agency (10 μg l⁻¹) (U.S. EPA 2002; WHO 2011). Considering that symptoms of arsenic poisoning may occur at the ppb (µg l⁻¹) level (Panagiotaras & Nikolopoulos 2015), the presented concentrations may be dangerous for terrestrial and aquatic organisms. Fortunately, there are still some places where arsenic is at the background level (< 1.0 μg l⁻¹, e.g. Red and Anllóns rivers ) or close to the limit established by WHO (<10 μg l⁻¹, e.g. Caohai and Waihai lakes) (McArthur et al. 2012; Wei & Zhang 2012; Pietro et al. 2016).

The highest arsenic concentrations are found mainly in natural hydrothermal systems (from 900 to 3560 µg l⁻¹) (Ning 2002; Litynska et al. 2017) and in samples from the most contaminated areas of the world, including West Bengal, Bangladesh, Taiwan, Mexico, Chile and India (from 100 to more than 4700 µg l⁻¹) (Saint-Jacques et al. 2014; Litynska et al. 2017).

Examples of As concentrations in sediment samples from different parts of the world are presented in Table 2. In this case, arsenic concentrations reported for rivers, lakes and seas range from 0.004 to 342 μg g⁻¹ d.w. (Table 2). According to ATSDR (2007), the concentration below 10 µgAs g⁻¹ d.w. is considered as a standard background level, whereas higher values can be attributed to anthropogenic activities such as fertilization, application of pesticides, industrial pollution, tanning and copper smelters (Ali et al. 2016). However, the occurrence of some natural rocks and minerals may also contribute to an increase in the arsenic content in sediments. The best example is authigenic pyrite, which appears in sediments of many rivers, lakes and oceans. As reported by Litynska et al. (2017), the content of arsenic in rocks of contemporary or recent volcanic activity can be up to

20 μg g⁻¹. During diagenesis or autogenesis processes, the concentration of arsenic in sedimentary rocks may reach 60 μg As g⁻¹. According to the current research on stream sediments from Europe, the concentration of arsenic ranges from 1 to 241 μg g⁻¹. The distribution of As in sediments from Northern Europe (Norway, Finland, southern Sweden, north-west Scotland, northern Poland and the Baltic States) and from south and south-east Spain shows a low total value of As, i.e. <4.0 μg g⁻¹, while higher values (> 11.0 μg g⁻¹) occur in the Skellefte mineralized belt in northern Sweden, Portugal and western Spain, the eastern Pyrenees, a zone from the Massif Central to southern Brittany in France, most of England, Wales, Northern Ireland and northern Bohemia and point anomalies in southern Hungary, south-eastern Switzerland and south-eastern Spain. The most enriched stream sediment sample occurs in Brittany, with 241 μg g⁻¹ of As and 1.7% of S. This was associated with the major shear zone of known mineralization (high content of Sn, Li, Sb and Cs), indicating a leucogranitic hydrothermal change. A similar pattern was found for floodplain sediments collected from the same regions. However, the values were slightly different, i.e. total As was between < 1 and 390 μg g⁻¹, while low values were < 3.0 μg g⁻¹, and high values of As were >13.0 μg g⁻¹ (Salminen 2005).

Finally, it is worth mentioning the sediment affected by substances originating from chemical weapons. Recently, this problem has been again investigated by researchers, which additionally has provided new information. For example, Emelyanov (2007) studied sediment samples from the Gotland Basin (a potential region of dumped chemical weapons). Surprisingly, he found no chemical munitions there. The high content of arsenic (77 μg g⁻¹) resulted from a natural (diagenetic) process and the presence of iron sulfides.

The potential leakage of arsenical chemical weapons was also examined by Bełdowski et al. (2016a). Sediment samples from the Gulf of Gdansk and the Słupsk Furrow, as well as from Gotland and Bornholm dumpsites contained arsenic in the range of 0.3 to 23 μg As g⁻¹ (mean values of arsenic from dumping sites: 13.0 μg g⁻¹ – Gotland Deep, 17.0 – Bornholm Deep μg g⁻¹). Only some samples from the Bornholm Deep contained arsenic derived from chemical weapons. However, the elevated concentrations of mercury (Hg) (130–135 ng g⁻¹) were found in three samples from the Gotland dumpsite area. Moreover, four samples from the Gdansk Deep, which contained elevated concentrations of Hg, showed some influence on CWA. Elevated levels of zinc were found in approximately 115 sediment samples (179 in total), though only 17 samples contained detectable amounts of CWA. Degradation products of CWA are widespread in sediments from the Bornholm Deep area and some places of the Gotland Deep and the Gdansk Deep. Furthermore, the obtained results suggest that munitions containing CWAs are more scattered on the seafloor than previously suspected. Bełdowski et al. (2016a) concluded that there is still some leakage of CWA around dumpsites. Therefore, it is recommended that the environment close to chemical and conventional weapons should be further investigated.