In this module we're going to look at ceramic filtration, which falls in between sand filtration and synthetic membranes in terms of pore size and physical exclusion. Ceramic filtration is hot. There is tremendous growth in sales of ceramic filters in many low income countries. Although there are also some drawbacks which we'll look into, ceramic pot filters with pots that look little bit like flower pots are popular and can be locally made and there are also smaller ceramic candle filters which can be installed in different kinds of water storage containers. Here's an animation of a ceramic pot filter. Operation is very simple. The pot is filled with water which then passes through the ceramic and into a storage container below. This design includes built-in safe storage. And filtered water is collected from a tap. Ceramic pot filters don't require any electricity or consumables. But they do require periodic cleaning, especially if raw water is turbid. Normally this is done by scrubbing with a small brush. Ceramic pots can be produced using local materials. Though it's not easy to set up a high quality production facility. The main ingredient in ceramic filters is clay, which can be sourced directly from local mines or indirectly by crushing unfired bricks. Clay shouldn't have too much sand or organic material in it, and usually local potters will know where good quality clay can be found. Clay is mixed with burn-out material to increase porosity of the finished product. Usually, organic waste material like sawdust or ground rice husks can be used with three or four times as much clay as burn-out material by weight. The proportion and size of burn-out material can dramatically effect filter flow rates and efficacy. Most ceramic filters are coated with this thin layer of silver just after firing. This helps improve the biological performance of the filter. We'll talk more about silver in a bit. A few filter factories also add metal oxides, like goethite or iron-rich soils like laterite to the mix in order to improve removal of viruses. Once the materials are mixed together, they're pressed into shape. The flower pot press seen here, is a common shape, but oblong and semi-spherical filters are also widely produced. Once pressed into pots, the pots must dry for a week or two and then be fired in a kiln. It takes a bit less than a day to fire the pots, typically at a temperature of around 800 degrees Celsius. And then around a day for cooling before the pots can be handled and checked. Once pots are fired, it's very important to test every single pot to make sure there are no cracks or other defects. Quality control begins with visual inspection and checking to see if the pot makes a clear bell like sound when it is tapped or pinged. Immersing the pot in water without letting the water come over the rim, is a simple kind of pressure test that can quickly identify cracks that might not be visible. A more thorough flow rate test can also identify cracks by using a calibrated t-device like the one in shown here, which rests on the rim of the pot. When the pots are then filled with water, it's possible to check how long it takes for the water level to drop. A typical target is around two liters per hour, though that depends on the size of the pot. Finally, challenge tests can be conducted to see if filters are able to remove bacteria like E. coli. Typically, around ten to 20% of filters will fail these quality control checks and must be destroyed. Factories have to build in this loss into their business models. Unfortunately, defective filters can't all just be ground up and used in a next production run, because the clay properties have irreversibly changed. However, some factories do include small amounts of ground up fired clay, which is called grog, in production. But viruses then, are too small to be removed through physical straining. They can however be removed through electrostatic attraction. Most viruses have a negative surface charge, and they are attracted to positively charged surfaces such as metal oxides. Laterite soil contains high levels of goethite and iron oxide with a positive charge and some manufacturers add laterite soil to the clay used in filter production, in order to improve removal of viruses. However, the effectiveness of this practice hasn't been rigorously tested. Also, if electrostatic absorption is the key mechanism in virus removal, there's possibility that absorption sites could eventually be completely filled up. Or even that absorbed viruses could be remobilized due to changes in influent water. In general, virus removal in ceramic filters is poor and more research is needed to identify mechanisms that could lead to better virus removal. There is a third mechanism for removing pathogens in ceramic filters through chemical reactions. It has been known for thousands of years that metallic silver has anti-microbial properties, and it has become very common to coat ceramic filters with silver after the firing process, either by dipping filters in a silver solution or by brushing it onto the sides. The amount and type of silver applied seems to be more important than whether silver is put on the inside or the outside of the pot. There's no need to be concerned about silver leaching from the pots and causing human health problems. There is no WHO drinking water guideline value for silver. And the amounts that have been measured leaching from pots are very low. There have been reports however, that arsenic can leach into filtered water at levels of health concern. This has nothing to do with the silver. It's thought that the arsenic is naturally present in the clay and that with time, arsenic leaching will drop to safe levels. However, it is important to check the concentration of arsenic in filtered water. The silver seems to have two effects. First, to increase the removal of bacteria, and second, to prevent biofilm formation on the ceramic surface which can reduce the flow rates. It's thought that silver doesn't have much impact on viruses. Two types of silver are used in ceramic pots: colloidal silver, also called nanosilver, which consists of very small particles as the name suggests of 100 nanometers or less in size. Quick quiz. How many microns is that? Go back to science if you don't know. The interactions on nanosilver and bacteria are complex and the subject of active research. It does seem that size is important and that nano particles may even enter into bacteria cells and cause internal damage. It also seems that contact with the silver surface is important and that release of dissolved silver can't alone explain all the effects seen on bacteria. The other kind of silver is ionic silver created from salts such as silver nitrate. Actually, silver nitrate is used in commercial production of colloidal silver. For example, by the large company, Argenol, which produces a nanosilver powder branded as Collargol. Some ceramic filter factories produce their own colloidal silver from silver nitrate in order to save costs. However, it's not clear that the locally produced nanosilver is as effective as commercial varieties. RDI, a leading producer of ceramic pots in Cambodia that uses silver nitrate, reports that they create colloidal silver particles around five microns in size. Such large particles certainly couldn't penetrate inside a bacterial cell, and might have very different effectiveness from smaller nanoparticles. Until the efficacy of silver nitrate is more conclusively demonstrated, it's recommended to use commercial nanosilver products, though this will slightly increase the cost of filter production. Protozoa are fairly well removed by ceramic filters with two to five log reduction values typically seen. This is mainly through physical size exclusion processes. Bacteria, which are a bit smaller, have somewhat lower reduction in the range of one to two or more log reduction values and the calculation of these can be complicated by regrowth and recontamination on the clean side of the ceramics. However, viruses show much poorer removal, with some studies showing no removal at all, and others showing one or maybe two log reductions. It's thought that the main removal processes here are electrostatic rather than size exclusion. Ceramic pot filters have been produced for many years in Latin America and there are mature products in markets there. In the last ten years or so, ceramic pots have seen tremendous growth in Cambodia led by two non-governmental organizations. Resource Development International Cambodia or RDIC, and International Enterprise Development or IDE. IDE began producing ceramic filters in 2001 and RDI a few years later. Both NGOs produce their own pots, and sell them without subsidy through a variety of distribution channels including direct sales, sales to vendors, and sales to NGOs. The price is typically around eight to 15 U.S. Dollars. And in recent years, IDE has partnered with the U.S. NGO PATH to produce a more aesthetically pleasing model, shown here. Their main model is called the Tunsai, which is Khmer for rabbit. And their new model is the Super Tunsai. Although the Super Tunsai costs almost twice as much as the regular Tunsai, the new model has considerably stronger sales. Both RDI and IDE can produce several thousand filters per month and each has sold hundreds of thousands of filters over the last ten years. In 2007, the World Bank's water and sanitation program, together with UNICEF, supported an evaluation of the RDI and IDE filters in Cambodia. The study contains a lot of really valuable information, and I encourage you to read the full report available here. A summary of some of the key findings are that, in total, only 30% of households, which were randomly selected, were using the filters at the time of the visit. However, this was heavily influenced by the time since implementation. Nearly 90% of those who had gotten filters within the past six months were still using them. There was a clear and steady decline of about two percent per month so that after two years, only about half of the filters were in use. The most commonly reported reason for disuse was breakage, either of the ceramic element itself or the tap. Only about five percent of respondents reported slow filtration rates as the reason for disuse. Also, use among those who had paid for a filter was somewhat higher, at 50 percent. Those who were actively using a filter, reported filling the filter almost twice in a day and cleaning two to three times per week. The evaluation found that filters removed about 70% of turbidity and had an LRV of 1.7 for E. coli. Older filters were not less effective at E. coli removal in spite of the manufacture recommendation to replace filters after two years. However, given the studied breakage rates, two years seems like a good estimate of filter life time. Ceramic candle filters like this one, have long been used in middle and upper class families often after boiling to remove calcium and magnesium precipitation. These candles can be screwed into metal or plastic household containers and must be regularly scrubbed clean. Conventional candle filters tend to be out of the price range for low income households. But new products like the tulip water filter are bridging that gap. A tulip filter consists of just the ceramic candle element in a plastic protective housing, connected by a long tube to a safe storage container. By using a long tube, the pressure difference is increased and the flow rates are higher at around four to five liters per hour. Tulip candle filters, unlike ceramic pot filters, can be backwashed with a plastic bulb and of course can also be scrubbed clean. A big advantage is the small size. A tulip filter can easily be packaged and transported without fear of breakage. Tulip has recently produced table top model, which includes a safe storage container. The market isn't as big yet as for ceramic pot filters. But several tens of thousands of these filters have been sold in a number of low income countries. One of the big advantages of ceramic filtration is the high social acceptability and fairly high use rates. However, those are balanced by moderate effectiveness especially for viruses. Most ceramic pot filters include a in-built safe storage element. But have a fairly low flow rate of a few liters per hour. Ceramic filters involve a one-time capital cost, but also need supply chains for distribution and replacement. It's possible to produce ceramic pot filters locally, but the quality can be variable and it's important to have a strong quality assurance program. Finally, ceramic filters do visually improve the water quality by removing turbidity, but they do not provide and residual protection. And there is still the possibility of recontamination or regrowth, especially if filtered water is transferred to another container. Much of the content from this module has been taken from some really excellent resources which are freely available online. The first, is produced by the ceramics manufacturing working group. And is a best practice recommendation for local manufacturing of ceramic pot filters. The second, is a ceramic water filter handbook produced by RDI in Cambodia, and both of them are freely available on the internet. So in summary, ceramic filtration is a fast growing and exciting sector within household water treatment. Ceramic pot filters can be locally produced, though it's important to have a strong quality assurance program. And there are examples and resources to help production facilities do that. Ceramic filtration involves a variety of removal mechanisms, including primarily physical, but also electrostatic and chemical with the use of colloidal silver. However, removal of pathogens is mixed. Protozoa tend to be well removed. Bacteria fairly well removed. And viruses, not terribly well removed at all. Still, ceramic filters are very user friendly, and tend to have high user satisfaction rates.
