Naturlig förekomst av arsenik och avskiljning av arsenik från grundvatten : test av olika filtertekniker avsedda för enskilda brunnar

Abstract: During the last few years the presence of high arsenic (As) concentrations in ground water has been of major concern both internationally and in Sweden. Much evidence has been reported about toxic effects caused by arsenic. The carcinogenic effects and the possibility to measure the toxic impacts at low concentrations made the World Health Organisation (WHO) to reduce the guideline value from 50 to 10 μg l-1 in 1993. In Sweden the corresponding reduction was implemented in 2003. Several surveys in Sweden have shown that high arsenic concentrations may occur in ground water. The National Board of Health and Welfare is responsible for guidance and supervision of health protection. This responsibility covers privately owned wells and drinking water facilities that produce less than 10 m3 drinking water per day or support less than 50 people. In 2006 the National Board published a report in which they stated that the existing knowledge was limited concerning As removal from ground water. The objective of this master's thesis is to test some of the current commercially available filter techniques that are used to remove As from ground water. In addition, the national geographic distribution of arsenic and factors governing that distribution are discussed. Experiments regarding As removal by different filter techniques were carried out in a small water purification plant in the village of Lästringe, located 40 km NE of the city of Nyköping. Only ground water is used in the purification plant. In the experiments, As(III) was added to the water as NaAsO2. Seven different filters of varying size were tested, three of which were based on adsorption to titanium oxide or iron hydroxide, while another two were based on a strong base ion exchanger (SBA) combined with iron or aluminium (hydr)oxide. Finally, two filters were based on reversed osmosis. The filter experiment was set up in two rounds of 30 days each, and no firm conclusions can therefore be made about the endurance of the filters. The incoming water contained 26-40,5 μg As l-1 during the first round and 125-142 μg As l-1during the second round. After the first round the five most efficient filters were selected for the second round. Magnetic valves and timers were used to adjust the daily volume of water put through the filters in order to simulate the water consumption pattern of a household during 24 hours. A chemical analysis of existing As species revealed that almost all As(III) had been oxidized to As(V). The oxidation could have been catalyzed by manganese oxides together with bacteria. Consequently the filters were supplied with water containing both As(III) and As(V). The filters containing TiO2 and Fe(OH)3 granulates and the Fe(OH)3 impregnated anion exchanger had the highest As removal efficiency (³ 97,3 %), resulting in As concentrations well below the guideline value of 10 ug As l-1. The rather large reverse osmosis filter showed a smaller percentage of As removal and the As concentration in the water that had passed the filter was mostly higher than the guideline value whenever the incoming water had an As concentration ³ 125 μg l-1. However, the reverse osmosis filter removed many other substances to a large extent. The lower extent of As removal can probably be explained by a high fraction of As (III) occurring as arsenite. Before choosing the filter technique, it is important to analyse the groundwater with respect to existing As species, total concentration of As and other elements forming oxyanions, dissolved organic carbon (DOC) and pH. Silica (H2SiO40) can compete with arsenic for adsorption sites in filters based on metal (hydr)oxides. Sulphate (SO42-) can be a competitor through anion exchange in filters based on SBA.