Urbanization has led to rapid and uncontrolled growth of informal housing settlements in many developing countries. As most slum growth is unplanned, these areas tend to lack basic infrastructure including sanitation. The high user rates, lack of water and electricity infrastructure, space limitations, and scant financial resources make sanitation provision a major challenge in slums. As most decentralized sanitation technologies fail when applied in these environments, better technologies need to be developed that cater to the specific needs of slum dwellers. One promising technology, the membrane bioreactor (MBR) is routinely used in developed countries when a compact and resilient treatment system is required. However, the energy requirement of existing MBRs is high, as most are aerobic systems which require aeration. Anaerobic MBRs (AnMBR), which do not require aeration, have led to an improvement of the energy profile of MBRs. As research into the technology is still in its infancy, little is known regarding its applicability in high-density urban environments. This body of research is aimed at understanding the AnMBR’s treatment performance and overall reliability in challenging circumstances similar to those encountered in slums.
The appropriateness of an AnMBR was investigated with pilot and full-scale systems treating real wastewater in field conditions. The first investigation, discussed in Chapter 3, was used to determine the resilience of AnMBR treatment when subjected to periods of disuse and high fluctuations in incoming feed strengths. Decentralized systems often see much higher variations in feed composition than centralized systems as they lack large collection systems which homogenize the influent wastewater. Depending on the application, periods of low and no flow are also possible. During this long-term study it was observed that the membrane served an important role in controlling the effluent quality, especially when environmental conditions and feed characteristics varied so significantly as to upset biological stability. The system achieved an average COD removal efficiency of 88.2% throughout the study. It was also observed during this study that the system had higher removal efficiencies when treating higher COD concentrations. Higher strength wastewaters can routinely be found in decentralized applications where dilution water is minimal. These locations include water-efficient buildings, direct coupling to public toilets, and fecal sludge treatment plants. It was also found that the AnMBR was capable of rapidly recovering from extended periods of disuse. This ensures that the AnMBR can be applied to areas, such as schools and hotels that experience large seasonal variations and periods of disuse.
The second investigation, described in Chapter 4, examined how fluctuations in ambient temperatures affect fouling resistance. In small decentralized applications, operating the reactor at ambient temperatures is the most likely scenario, as controlling the reactor temperature would incur a high energy demand. Operating at ambient temperatures means that variations can be high, and that temperatures can drop below ideal ranges. Temperature is known to affect biological treatment and to a lesser extent membrane filtration, but the interactions between the two are not fully understood. To determine the effect of temperature on operation, a pilot scale AnMBR was used to treat wastewater with fluctuating ambient temperatures. Three trials were conducted during summer and winter conditions, as well an artificially heated period. It was found that membrane permeability can be greatly affected by operating temperature but its effect varied depending on the fouling state of the membrane. Virgin, or recently cleaned membranes were not affected by low temperatures, while the permeability of slightly fouled membranes was negatively correlated to changes in temperature. When slightly fouled, a membrane TMP could increase by 2.4 times with a 10oC drop in temperature. The magnitude of the TMP increase could not be explained by changes in water viscosity alone. The effect of temperature on TMP decreased when fouling became severe and normal operating pressures were high. These results suggest that seasonal adjustments to AnMBR operation would be necessary to prevent sharp and excessive increases in operational TMP during cold spells.
Chapter 5 investigated the feasibility of recovering water, nutrients, and energy in an off-grid and decentralized AnMBR. This investigation performed an energy, nutrient, and mass balance for a theoretical AnMBR treating water from a public toilet in a high density setting. What was concluded from this study is that complete water recycling can be accomplished in such an environment. Onsite water recycling would allow the system to be applied in arid urban areas as well as places lacking regular water provision. The study also concluded that the energy content of wastewater in a high density area would be sufficient to power an AnMBR and electronic toilet. For areas where low wastewater strengths would be expected, food waste addition to the wastewater would improve the energy profile of the system. As many urban areas of developing countries struggle with solid waste management, there is the opportunity to link food waste management with wastewater treatment. This study also highlighted the potential problems that ammonia and salinity buildup could have on a system that achieves complete water recycling.
Once the system specifically designed for urban areas was deemed theoretically feasible, a full-scale, solar-powered, prototypical system was constructed in Florida and tested in India (Chapter 6). This system, which was applied in Kerala, India, was investigated for its treatment and membrane performance as well as energy consumption. During the first four months of operation, the system was able to produce high quality product water that could be used for toilet flushing. This was achieved despite the low strength of the incoming feed water and higher than anticipated wastewater production rates. The wastewater strength was low due to the system’s application in a school setting and high levels of dilution water. The reliance on multiple anti-fouling mechanisms allowed the system to operate for 4 months without a significant change in TMP. The average energy consumption per unit of produced water depended on the amount of water treated per day. On average the energy consumption was 1.52 kWh/m3, but that value dropped to 0.83 kWh/m3 when volumes greater than 200 liters were treated per day. The lowest value measured during this trial was 0.16 kWh/m3 when 1,394 liters were produced. All of the energy used by the system was produced by onsite photovoltaics, with minimal carbon footprint. While the system was capable of meeting the water demand of the toilet system, further improvements in the energy demand of the system will be necessary to make the system more cost-effective, robust and reliable.
These results suggest that AnMBRs can be applied in high density urban areas for the dual objectives of wastewater treatment and resource recovery. Their reliable treatment in the face of large fluctuations in feed concentration, volume, and temperature suggests they are appropriate for decentralized applications. Membrane filtration allows water to be reliably recycled onsite with minimal operator oversight. The low energy requirements of the system allow for onsite renewable energy sourced, such as photovoltaics to be used to power the system. AnMBRs are able to address many of the challenges that traditional sanitation technologies cannot, which makes them a promising technology to address the problems encountered in slum sanitation.
| Identifer | oai:union.ndltd.org:USF/oai:scholarcommons.usf.edu:etd-7370 |
| Date | 07 July 2016 |
| Creators | Bair, Robert Alonso |
| Publisher | Scholar Commons |
| Source Sets | University of South Flordia |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Graduate Theses and Dissertations |
| Rights | default |
Page generated in 0.0022 seconds