Most waste generated in South Africa is sent to landfills for disposal, and
although it is confined in specific areas, it can potentially affect both above and
below ground water resources, impacting environmental and public health. This
is particularly relevant in a country where water supplies are limited and
groundwater resources are prone to pollution. The primary objective of this study
was to assess the performance of an upflow packed-bed bioreactor purposedesigned
for the treatment of leachates produced by landfills in the Durban
Metropolitan Area (DMA). The effect of parameters such as the nature of the
biofilm support matrix, aeration rate and recycle rate on the efficacy of the
system were investigated. Another major aim of the project was to develop a low
maintenance technology that could, nonetheless, bioremediate leachate
effectively at minimum cost. This aspect of process design is a crucial factor in
areas where there is a shortage of both funds and skilled labour.
The glass 132 l packed-bed upflow bioreactor was evaluated by measuring its
efficiency in terms of chemical oxygen demand (COD) and biological oxygen
demand (BOD) reduction and ammonia removal. The bioreactor could be
configured as a batch-type system, which was useful for comparing operating
conditions; or as a continuous cascade system, which was used to assess its
overall performance. Different biofilm support matrices viz. various grades of pine
bark, plastic bioballs and ceramic noodles were evaluated in 22 l batch-type
reactors.
Leachates from five landfill sites were remediated during the course of the study,
and only the leachate from Shongweni landfill, which had a remarkably low
BOD:COD ratio (0.05), was intractable and could not be successfully treated;
even in flask trials designed to test strategies such as augmentation of microflora
and biostimulation. The other leachates investigated were from the Umlazi,
Marianhill, Bisarsar Road (all general sites) and Bul-Bul Drive (a semi-hazardous
site) landfills, all of which were remediated to some degree. Originally, leachate
from the Umlazi landfill site was used, but it became unavailable when the site
closed enforcing the use of other leachates for the remainder of the investigation.
Leachates from Marianhill, Bisarsar Road and Bul-Bul Drive were treated
simultaneously in duplicate operating the six-chambered bioreactor in the batchtype
configuration. The highest COD removal efficiency (49 %) was obtained in
the chambers treating the Bul-Bul Drive leachate, which was therefore used for
further investigations. This leachate had the highest BOD:COD ratio and was
therefore expected to be the most suited to biological remediation.
The bioreactor performed best when plastic bioballs were used as biofilm support
matrix with a relatively low level of aeration, although the uncomposted form of
pine bark was used initially as the support matrix because it is inexpensive and
readily available in South Africa. However, although satisfactory COD reduction
(30 – 61 %) and ammonia removal (87 – 98 %) was achieved when the Umlazi
leachate was treated, the possibility of compounds leaching out of the bark and
affecting the quality of the treated leachate was a concern. Also, pine bark would
be prone to mechanical degradation in a full scale operation. Of the other solid
support matrices tested using the Bul-Bul leachate, COD removal efficiencies
were superior with plastic bioballs (60 %) than with pine bark chips (29 %). The
former therefore became the preferred biofilm support matrix.
Aeration level did influence bioremediation of the Umlazi landfill leachate since
those chambers aerated with an aquarium pump (0.05 – 0.1 litres air/litre
leachate/min; 60 % COD removal) performed better than those aerated with a
blower (0.6 -0.7 litres air/litre leachate/min; 42 % COD removal) and those that
remained unaerated (44 % COD removal).
Recycle rate did not significantly affect bioremediation, but the performance of
the system was higher when operated in batch mode (up to 60 % influent COD
removal), rather than in continuous flow-through (cascade) mode when only 37
% of the influent COD in the Bul-Bul leachate was removed. Under the latter
conditions, most of the reduction occurred in the first four chambers and very
little biodegradation occurred in the final two chambers. The cascade-mode will
require some refinement to enhance the COD removal efficiencies achieved.
However, it did eliminate 89 % of the BOD present in the raw leachate, producing
a treated effluent with a consistent BOD:COD ratio of 0.05.
The COD removal efficiencies achieved covered a wide range from a minimum of
23 % with Marianhill leachate to a maximum of 63 % with leachate from Bul-Bul
Drive. These results are comparable with many of those reported by other
authors treating landfill leachate. Up to 98 % of the ammonia was removed when
the Umlazi leachate was treated. However, ammonia removal from the other
leachates tested was erratic.
Although the treated leachate from this system could not be released into the
environment without further remediation, the reduction in concentration of
pollutants would allow its return to the local water supply via a wastewater
treatment plant. This was achieved without temperature and pH regulation or
addition of extraneous nutrient sources. A cost-effective, low maintenance
technology such as this one would be a useful tool for the treatment of effluents
such as landfill leachate in countries like South Africa where although water
conservation is urgently required, resources for highly sophisticated effluent
remediation are often not readily available. / Thesis (Ph.D.)-University of KwaZulu-Natal, Pietermaritzburg, 2011.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:ukzn/oai:http://researchspace.ukzn.ac.za:10413/8604 |
| Date | January 2011 |
| Creators | Vaughan, Halina. |
| Contributors | Tivchev, George V., Wallis, Frederick Malcolm., Laing, Mark D. |
| Source Sets | South African National ETD Portal |
| Language | en_ZA |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0026 seconds