Investigating the potential of ceramic pot filters to treat stored rainwater

A lack of access to safe centralised water services in many parts of the world means that alternative water supplies, such as rainwater harvesting (RWH), are increasingly being explored and developed.

However, some studies have shown that RWH may pose a health risk because of its potential to carry microbial pathogens through wet deposition (bonding of chemicals in the air before hitting the roof), transit via the catchment area (usually a rooftop) and drainage gutters and pipes, and the residence time in the storage tank itself.

To improve the microbiological quality of stored drinking water, ceramic pot filters (CPFs) may be a robust point-of-use technological solution. Pores in the ceramic act as a physical barrier to pathogens and have been demonstrated to be effective at removing >99% of protozoa and 90-99% of bacteria. CPFs may also work through the additional mechanisms of adsorption and biofilm metabolism. Through a combination of mechanisms, CPFs are associated with a 60-70% reduction in diarrheal disease incidence reported by users in some studies. However, the relationship between different materials used to make the filter, pore size distribution, flow rate and bacterial removal is still insufficiently understood.

Supported by the UWE Global Water Security Programme, Joshua Oldfield, a BSc Geography student at the University of the West of England, has investigated the hydraulic properties of CPFs associated with removing bacteria (indicated by assessing total coliform, or TC) in stored rainwater in Kisoro, Uganda.

Kisoro is a district in south west Uganda where 92% of the population are living in rural, mountainous areas and 66.3% of households have no access to piped water. In 2017, the Ministry of Water and Environment reported that only 47% of the people in Kisoro district had access to safe water. Given the average rainfall of 1250-1500 mm in the area, RWH is a promising low-cost technology to store rainwater for use during the dry season, yet only 1.3% of Kisoro’s 305,000 residents currently uses RWH.

For his research, Joshua worked with water engineer Alan Cook and Dr Tavs Jorgensen, Associate Professor and AHRC Leadership Fellow at the Centre for Fine Print Research, to manufacture prototypes of three types of ceramic filter and evaluated their performance:

  • Filter A – 100% clay and no combustible material, fired at 750°C.
  • Filter B – 95% clay and 5% sawdust, fired at 650°C.
  • Filter C – 95% clay and 5% sawdust, fired at 600°C.

Samples of pond water with a TC of over 500 CFU/100ml were used to imitate water with microbiological contamination similar to Kisoro’s water quality. The World Health Organisation guideline for TC in drinking water is 0 CFU/100ml. The idea was to test how effective our lab-produced CPFs were at lowering TC towards the WHO standard, whilst maintaining an adequate flow rate (of 1 to 5 litres per hour).

In addition to testing for total coliform in the filtered sample, Joshua also conducted hydraulic and visual investigations using a Scanned Electronic Microscope (SEM) to examine porosity size and flow rate, which affects the achieved water disinfection effect.

The research showed that Filter B had the highest removal rate of bacteria (66.25% at a flow rate = 28.5 ml/hour) compared with Filter C (50.625%; flow rate = 26 ml/hour) that was made of the same material. Joshua hypothesises that the higher firing temperature gave Filter B more carbon within the clay through more completion pyrolysis of burnout material, making it more successful. Without any combustible burnout material, Filter A was not porous and no water was filtered. The differences in microporosity between Filter A and Filter B are apparent in the SEM images.

In conclusion, controlling the porosity of the filter seems to be key to successful CPFs, balancing off flow rate against treatment efficacy. The 66.25% treatment efficiency, being well below the >90% removal rates found in previous studies, suggests that more research on the formulation of material, especially the type and proportional volume of burnout material, production process and firing temperature, is needed. Furthermore, the lower than expected removal efficiency may indicate the need to combine CPFs with other household water treatments, such as solar disinfection and boiling. When designing CPFs, consideration will also have to be given to the feasibility of producing good quality CPFs locally, with materials available in and around Kisoro district and other locations seeking decentralised, off-grid solutions.