In order to improve the water situation in rural areas of Bangladesh, a research group at the University of Dhaka has been developing low cost domestic methods to remove pathogens from surface water through pasteurisation using free solar energy. Pasteurisation is a process in which water is heated to approximately 60 °C and maintained for about 30 minutes to destroy pathogens. In these methods, the water is also exposed to UV-light from the sunshine, which causes destruction of diarrhoeal pathogens at temperatures somewhat lower than required in normal pasteurisation. However, despite many advantages these devices need to be installed for each time of use. Recently, a semi-permanent device has been developed which is expected to be more user friendly. The objective of this Master thesis has been to study and optimize the low cost semi-permanent device that can deliver safe drinking water to people in rural areas. Two test devices were constructed to determine the most effective treatment e.g. temperature, time, solar radiation, user-friendliness and cost. To replicate the results from the solar heating tests a model, based on the solar radiation and convective heat loss from the device, was used. The model was also able to determine the time duration at a certain solar radiation level to estimate when the water is safe to drink. The results revealed that the performance of the device depends on thickness of the insulation and thickness of the air gap. This is because the most important factors to achieve safe drinking water are solar radiation and time. The modelling indicated that the measured water temperature corresponds well with the calculated water temperature and also showed that the lowest required solar radiation is 390 W/m2 to reach drinking water criteria, at an air temperature of 25 °C. A study of microbiology showed that the semi-permanent low cost device could purify surface water to a safe level.