Doctoral Dissertations

Date of Award

8-1990

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Civil Engineering

Major Professor

R. Bruce Robinson

Committee Members

Gregory Reed, Wayne Davis, Otto Kopp

Abstract

Manganese is usually present in groundwaters at a concentration of 0.05 to 1.0 mg/L, and groundwaters are seldom found which have manganese levels greater than 7 mg/L. Common aesthetic problems associated with high manganese concentrations include colored water ("red water"), stains, turbidity and taste. To eliminate these problems, EPA recommends that the manganese concentration in drinking water not exceed 0.05 mg/L. Most manganese removal methods involve manganese oxidation to an insoluble higher oxidation state and the separation of the solid phase from the liquid phase by sedimentation and filtration. Although manganese removal eliminates color and turbidity problems, manganese removal is relatively expensive for small communities. An economical alternative would be to control manganese problems by chemical addition without removing it from the groundwater, as manganese is not toxic but is merely an aesthetic problem. The fact that sludge generation could be eliminated and substantial savings in the capital, operation and maintenance costs could be obtained by sequestering the manganese without removing it from the groundwater makes sequestration a viable alternative to other manganese removal techniques. However, several attempts to sequester manganese without removing it from groundwater have not yielded acceptable results. There is no effective technique for sequestering manganese. The physical/chemical principles of this method are not well understood, and the mechanism of manganese sequestration has not been previously investigated. This experimental investigation was aimed at identifying an effective technique to sequester manganese. Sequestration is the addition of chemicals to groundwater to maintain low color and turbidity without removing manganese and iron. The application of this technique to different groundwaters which vary in water quality (pH, presence/absence of calcium, etc.) was tested in this research using model groundwater prepared in the laboratory. Various sequestrant/oxidant combinations were tested: silicate/chlorine dioxide; polyphosphate/chlorine dioxide; polyphosphates and silicates in conjunction with chlorine dioxide; and polyphosphate/hypochlorite. Specific objectives of this experimental investigation are: j to determine the best sequestrant/oxidant combination for effective manganese control; 2. to determine the mechanism of sequestration, i.e. how is manganese sequestered? For example, is a manganese colloid formed and dispersed by sequestrants in successful studies as in iron sequestration or is manganese sequestration different? What is the mechanism by which success is achieved by this technique? 3. to determine the effects of several variables including water quality (pH, presence and absence of hardness) on manganese treatability. Variables tested in this research are explained in Section 1.5, Turbidity, color, and percentage filterable iron were the criteria used to analyze treatment effectiveness. Streaming current detectors were used to determine whether streaming current can predict good treatment. X-ray diffraction and equilibrium modeling studies were done to evaluate the manganese and calcium species formed in sequestration. Results from laboratory experiments indicate that manganese colloids formed in manganese treatment by silicate or polyphosphates with chlorine dioxide were stabilized, and high filterabilities (using 0.1 micron filters) could be obtained. High filterability indicates that the manganese particles did not grow to sizes larger than 0.1 micron, the membrane filter nominal pore size. This indicates that in actual practice, manganese will not precipitate in pipelines or fixtures when treated by this method, as the coalescence, growth, and subsequent settlement of manganese precipitates were limited by the addition of polyphosphates. However, these methods are not recommended for manganese sequestration because high discoloration was produced due to manganese oxidation by chlorine dioxide. Effective manganese stabilization was achieved with polyphosphate and sodium hypochlorite. Polyphosphates inhibited manganese oxidation by hypochlorite, and no noticeable color was produced. This inhibition of manganese oxidation was determined to be an important mechanism in the success of this technique. Manganese retention on 1000 molecular weight (MW) cut-off ultrafiltration membranes indicates that manganese may have complexed with polyphosphate (sequestrant) which had a molecular weight higher than the ultrafilter cut-off level. It is hypothesized that complexation with polyphosphates made the manganese-polyphosphate complex more resistant to oxidation than free Mn2+. In the presence of polyphosphates, manganese was not oxidized by hypochlorite and produced no discoloration for a period of 10 days, whereas in the absence of polyphosphates manganese was slowly oxidized and produced discoloration. Also, in the absence of polyphosphates, manganese was retained on 0.1 micron and 200,000 MW ultrafilters filters. These results imply that polyphosphates would inhibit manganese oxidation and subsequent precipitation in pipelines and fixtures when used with hypochlorite in actual facilities. Thus, the dual purposes of disinfection and manganese stabilization can be achieved with this technique. The effects of two important water quality parameters (pH and calcium) on manganese treatment were studied to determine the applicability of the polyphosphate/hypochlorite technique to natural ground waters. Higher turbidities (proportional to the polyphosphate dosage) were noted in the presence of calcium than in its absence, due to the formation of a calcium-phosphate precipitate. The precipitate could not be identified by x-ray diffraction studies, as it was x-ray amorphous. Equilibrium modeling studies were done using MINTEQ (a geochemical equilibrium model). MINTEQ predicted that hydroxyapatite, Ca5(PO4)3OH precipitation is thermodynamically favored. Even though hydroxyapatite is thermodynamically the most stable precipitate that would form under the experimental conditions, other calcium-phosphate precipitates could initially form and slowly transform to hydroxyapatite. This phenomenon is discussed further in Section 2.2.1 of the literature review. In the presence of calcium, the best treatment was obtained at a dosage of 5 mg/L as total phosphate. The turbidity was below permissible limits (1.0 NTU) at this dosage, and high filterability and low discoloration was obtained. Manganese treatability by the polyphosphate/hypochlorite technique was also tested at various pH levels (6, 7, and 8). The best treatment was obtained at pH 6, due to the absence of calcium-phosphate precipitation.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS