As a first step in understanding the chemistry of toxic elements in sludged soil, this thesis presents a comprehensive model of their chemistry in the digester. A review of the literature had shown that heavy metals were likely to be held in 3 pools: as precipitated and detrital mineral phases 'Particulate'; as complexes with the flocculated biomass 'Biofloc'; and as complexes in solution 'Soluble'. A simple pragmatic fractionation procedure has been offered to separate these 3 pools based on their physical properties in water. A mass-balance between the pools showed that the 'Particulate' fraction held only 5-16% of the heavy metals but contained them in the highest concentration. The 'Biofloc' held 82-94% of the heavy metals. The 'Particulate' material was subdivided by density separation and examined by a combination of analytical SEM and XRD. Thirty-four minerals were identified by XRD, many of which were detrital. Secondary precipitates on the surface of detrital minerals were revealed by the SEM; of these only the sulphides were found to contain detectable levels of heavy metals. Eleven minerals were identified in the 'Biofloc', of which most were clays. The fractionation scheme defined 2 fractions that could hold heavy metals by complexation. The 'Soluble' had a CEC of 8.8 meq/gm and the 'Biofloc' 4 meq/gm. Complexation by the heavy metals and a few other important cations was measured. A thermodynamic model was built which describes the possible solution species, mineral phases and complexation by the biomass in terms of a set of 33 primary components. This model was solved by computer for an 'average' sludge based on published analyses, and considered 313 solution species, 42 exchange reactions and 129 possible minerals. The predicted speciation was broadly in line with observations and suggested that the majority of the heavy metals separating with the 'Biofloc' would have been held as fine enmeshed sulphide precipitates. The model may be used with existing programs such as MINEQL and GEOCHEM. Preliminary studies have shown that with a few additions the model may be used to describe the heavy metals in sludged soil.

Dialysed activated sludge was used as a substrate to test for enzymes which can hydrolyse activated sludge floc. Two hundred and fifty aerobic and anaerobic microorganisms were tested against activated sludge for the presence of hydrolytic enzymes. These included known genera and organisms obtained by various enrichment procedures. Anaerobic digester mixed liquor was contacted with activated sludge agar under anaerobic conditions. None of the microorganisms present in the digester liquor hydrolysed the floc. The following commercial enzymes were contacted with activated sludge singly, in combination, and sequentially under various physical and chemical conditions: protease, lipase, cellulose, pectinase, phospholipase C, trypsin, and glucuronidase. Although commercial enzymes and various microorganisms reacted with known substrate controls neither the enzymes nor microorganisms employed affected the activated sludge floc. Treatment of activated sludge floc with ethylenediaminetetraacetic acid resulted in gross deflocculation and release of humic substances. The floc particle, probably a combination of polymers bound in a complex manner, is resistant to enzymatic degradation. Several workers have reported on the enzymatic hydrolysis of polymers produced by axenic cultures isolated from activated sludge floc. It is unlikely that the results of their work can be extrapolated to the complex structure of activated sludge floc particles.

A total of 78 sludge samples originating from 69 sewage works of South Africa were used in this investigation. Seven of the sludge samples were liquid and the rest were solid. Moisture content, pH and mineral ion determination using ICP-OES were carried out. The methods used to digest the samples were EPA 3050B and 301OA for solid sludge and effluent (liquid sludge) respectively. The moisture content determination showed that fresh wet sludge was composed of water between 40 to 90 %. The minimum moisture content was found to be 2.70 % while the maximum was 88.50 % with a mean value of41 %. The pH results showed that the majority sludges produced had pH values between 4.8 - 6.5. The ICP-OES results which involved analysis and quantification of 22 mineral ions showed that the order ofabundance that was most common to the majority ofthe samples was P, Ca, Fe, AI, Mg, K, Zn, Na, Si, Mn, Cu, Cr, Ba, Pb, Sr, Se, B, Ni, Co, Mo and Cd. Even though phosphorus was the most abundant, 11 of the samples had calcium as the highest element. Looking at the heavy metals, zinc was the highest with cadmium being the least. The order of abundance in majority of the solid samples (64 in total) was Zn> Pb > Ni> Cd with the relative metal concentrations of Cu, Se, B, Cr, Co & Mo varying among the sewage works. The liquid samples also had phosphorus as one of the most abundant elements but was 102 times smaller comparing to solid sludges. In addition, most of the transition elements were found to be below the detection limit. Beryllium was exceptionally found to be below detection limit in all sludge samples. The results have pointed out that industrial effluent have 3 times the level of pollutants when compared to the domestic effluents. However, the methods of preparing sludge have no influence on the content or quantity of mineral ions. The results have been compared with 1989 data. The outcome shows that concentrations of the major nutrients namely calcium, magnesium and potassium remained relatively constant whereas phosphorus increased by more than 3 fold. On heavy metals, the data shows that the mean concentration level of Zn, Cu, Cr, Pb, Ni and Cd declined whereas that of Se, B and Mo showed an increase in 2002, all to a varying degree. The current results were also compared with the current maximum limits as stipulated in the permissible utilization and disposal of sewage sludge government guideline. The amount of Cu, Se, Pb and Zn were found to be above the limit in more than 90 % of the samples. There was no sewage works that met the required limits for all the elements of interest. When these results are compared with the intenational limits, all the elements fall within the acceptable range. It is therefore clear that the current South African guideline limit is too restrictive. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2004.

Graduation date: 1997

An essential step in producing "drinking" water is to precipitate the suspended and dissolved colloids through the addition of flocculents such as lime, ferric chloride, aluminium sulphate and/or poly-electrolytes. The by-product of this process is termed water treatment sludge (WTS) and contains mainly silt, clay and some organic matter. Previously this material was disposed of in landfill but more recently, alternative methods for its disposal are being evaluated. A potential disposal option is land treatment. In this system of waste disposal the inherent properties of the soil are used to assimilate the waste. Although the effect of the land disposal of WTS on soil chemical quality is gaining increasing research attention, few studies have investigated the effects on soil physical quality. This study was originally commissioned by a local water utility to evaluate the effects of the land disposal of sludge produced at their works, on soil quality. At this plant organic polymers are used to both flocculate the material and to thicken the sludge in the water recovery process. Fresh sludge has a consistence approaching that of slurry but dries to angular shaped aggregates of extremely high strength. Nevertheless, sludge aggregates comprise a network of micro-pores and channels and are therefore porous. Because of these properties, the potential use of WTS as a soil conditioner was considered.. Since lime, gypsum and polyacrylamide are wellrecognised soil conditioners, these were included as reference treatments in the study. Two field trials (Brookdale and Ukulinga) and laboratory experiments were designed to investigate the influence of WTS on soil in terms of water retention, hydraulic conductivity, evaporation, aeration, aggregation and strength. Seven rates of WTS are represented at the . Brookdale trial but research efforts were concentrated on the 0, 80, 320 and 1280 Mg ha' treatments. WTS was also applied as a mulch (without incorporation into the soil) at the 320, 640 and 1280 Mg ha" level. Gypsum was applied at rates of 5 and 10 Mg ha", lime at 2 and 10 Mg ha' and anionic polyacrylamide at 15 and 30 kg ha'. At the Ukulinga trial, WTS was mixed with the upper 0.2 m of the soil at rates of 0, 80, 320 and 1280 Mgha'. Only the high rates of gypsum, lime and anionic polyacrylamide being tested at the Brookdale trial are represented at the Ukulinga trial. All treatments in this study were maintained fallow. The laboratory study features an additional two soils to those from the field experiments, chosen to produce a range in clay contents. WTS influenced several soil physical properties. Soil bulk density decreased following the addition of sludge to soil. This caused an increase in porosity (particularly macro-porosity) and therefore water retained at saturation, but only of statistical significance at the 1280 Mg ha" level. Equally an increase in water retention at the wilting point (-1500 kPa matric potential) also occurred, owing to the high microporosity of sludge aggregates. Despite these effects very little change in both the plant available and readily available water content occurred. Neither, gypsum nor lime caused any significant change in water retention. Aslight improvement was noted on the polyacrylamide treatment at the Brookdale site but this effect did not persist for very long after the trial was established. Although in situ field measurements were influenced strongly by natural spatial variability, WTScaused a marked increase in the saturated hydraulic conductivity (Ks). The reasons for this relate to the higher porosity and the inherently stable nature of the sludge aggregates, which imparts a more open structure to the soil and reduces the extent of pore blockage. This finding was corroborated in a laboratory study in which strong positive correlations between sludge content and Ks was found. The water retention curve and saturated hydraulic conductivity was used to predict the unsaturated hydraulic conductivity function (Kw)using the RETe computer model of van Genuchten et al., 1991. The results showed a decrease in Kw on the sludgeamended treatments the extent of which increased with sludge content. This finding was tested in an evaporation study conducted under controlled environmental conditions. More water was conserved on the sludge-amended treatments than the control, because of its lower Kw. The application of the sludge as a mulch was more effective in conserving water than incorporating the sludge with soil. The air-filled porosity at field capacity (-10 kPa matric potential) of the sludge-amended soil remained within a favourable aeration range of 10-15%, which suggests that aeration should not be a limiting factor for plant growth. Air-permeability nevertheless improved substantially. Attempts at using the size distribution of dry soil aggregates to evaluate the influence of the sludge on aggregation proved unsuccessful. Saturated soil paste extracts for selected soil depths beneath the mulch layers at the Brookdale trial, nevertheless, showed significant increases in Ca2+ and Mt+ concentrations, which is encouraging from a soil stability perspective. Due to the inherently strongly aggregated nature of this soil, no meaningful change in aggregate stability, however, was measured. Significant improvements in soil stability were, nevertheless, found when fresh sludge was mixed with soil. If the sludge is not allowed to dry fully beforehand the polymer that it contains remains active and available for bonding of the soil particles together. Upon drying, these polymers become irreversibly attached to the soil substrate and win not become reactivated even upon re-wetting of the soil. This also explains why sludge aggregates found below only a few centimetres of the soil surface maintained their strongly aggregated nature. This suggests that although WTS consists of mainly silt and clay, the risk of this constituent fraction becoming released and clogging water conductive soil pores are, at present, low. Despite the high strength of the sludge aggregates the penetrometer soil . strength (PSS)within the tilled layer was non-significantly different from the control treatment. Below the tilled layer, however, the PSS on the sludge-amended treatments were lower owing mainly to wetter soil conditions. The research completed to date suggests that land treatment as an environmentally acceptable disposal option for water treatment sludge shows promise since soil conditions tend to be improved. / Thesis (Ph.D.)-University of Natal, Pietermaritzburg, 2001