Treatment of surface water (PAUL Water Treatment, University of Kassel) (2026– )

The Sundarbans are located in the southern regions of Bangladesh, in the estuaries of the great rivers Ganges, Brahmaputra and Meghna. In this area, the groundwater is often brackish, meaning it is undrinkable due to its natural salt content. This effect is exacerbated by the increasing number of shrimp farms in the area, as the salty water seeps into the ground and contaminates other aquifers. Even the deep groundwater is becoming saline due to salt intrusion from the sea. This effect is set to worsen in the future as a result of rising sea levels.

In these areas, rainwater is increasingly being collected. To tide people over during the dry season, water is drunk from ponds. This water is often heavily contaminated with microbes. The filters developed by the University of Kassel could provide a solution. The PAUL (Portable Aqua Unit for Lifesaving) weighs just 20 kg and purifies contaminated water through ultrafiltration (nano-membrane filtration) without electricity or chemicals. We have purchased one of these filters and are installing it in Shatkira in the Sundarbans. It can supply enough water daily for up to 400 people (theoretically up to 4,000 litres). In addition, this year we will be carrying out further optimisations of water use with our project partner BARCIK, like the construction of larger cisterns for communal use.

Sand and manganese filters as replacements for SIDKO filters (2025–)

The Bangladeshi government is increasingly turning to deep wells. SIDKO filters are still being purchased in isolated cases. Both approaches have significant drawbacks. We are therefore advising the Department for Public Health Engineering (DPHE) in collaboration with the University of Heidelberg and presenting our Rooftank filter. Sand of varying grain sizes is layered into existing roof tanks and separated by layers of fleece/foam rubber. In these sand layers, the iron dissolved in the water is oxidised and precipitates as iron sludge. We use layers of manganese oxide to remove arsenic. All materials are available locally at a low cost. Thanks to the newly developed backwash system, this filter can be backwashed very easily. A very cost-effective and robust technology as an alternative to deep wells. The DPHE is enthusiastic about the development and has asked us to install 200 such filters in state schools with particularly poor water quality. This will now be implemented in 2026 with funding from the Foundation for Development Cooperation (SEZ) using funds from the State of Baden-Württemberg.

Following the completion of the funding project, local technicians at both sites, Ullapara and Nabigonj, will be supported in marketing these filters themselves through newly established start-ups.

Development of innovative and sustainable methods for arsenic removal in collaboration with various universities (2016–2025)

• Since 2024: Installation of 20 backwashable sand and manganese filters to remove iron and arsenic from groundwater
• 2021: Trial of a flow-through coagulation cell to remove arsenic from groundwater at the school in Sherpur
• 2019: Geochemical studies on electrocoagulation as a sustainable approach to drinking water treatment in Bangladesh (Master’s thesis at Heidelberg University)
• 2018: AsFreeH2O: Suitability and social acceptance of water purification technologies under the influence of future sea-level rises (HeiKa cooperation project between Heidelberg University and the Karlsruhe Institute of Technology)
• 2016: Arsenic removal using new, recyclable filter media (Master’s thesis at the Universities of Heidelberg and Darmstadt)

Installation of four SIDKO filters for the removal of arsenic from drinking water (2014–2019)

To fight the arsenic problem in Bangladesh, AGAPE installed four SIDKO filter systems, which reliably remove arsenic in a two-stage process using sand and granulated iron hydroxide. These filter systems were open to the public.

The filter systems are only practical in central locations or for a small catchment area, as large quantities of drinking water are not transported over long distances. A further limitation is the filtration rate. On average, only 1,000 litres of water can be filtered per day, meaning that these systems are not big enough to supply large numbers of people.

SIDKO filters are made of metal. As a result, the material rusts very quickly in tropical, humid conditions, and the service life of these systems is consequently limited to just a few years.

Having weighed up the costs and benefits, AGAPE has decided to discontinue the use of these types of filters in future and has asked Heidelberg University to provide advice on innovative cleaning methods.