Physical, chemical, and biological aspects of subsurface organic waste injection near Wilmington, North Carolina

Abstract

From May 1968 to December 1972, an industrial organic waste was injected at rates of 100 to 200 gallons per minute (6.3 to 12.6 litres per second) into a sand, gravel, and limestone aquifer of Late Cretaceous age by Hercules Inc. located near Wilmington, North Carolina. This report presents both field and laboratory data pertaining to the physical, chemical, and biological effects of waste injection into the subsurface at this particular site, a case history of the operation, predictions of the reactions between certain organic wastes and the aquifer components, and descriptions of the effects of these reactions on the subsurface movement of the wastes. The case history documents a situation in which subsurface waste injection could not be considered a successful means of waste disposal. The first injection well was used only for 1 year due to excessive wellhead pressure build-up above the specified pressure limit of 150 pounds per square inch (10.3 bars). A second injection well drilled as a replacement operated for only 5 months before it too began to have problems with plugging. Upward leakage of waste into shallower aquifers was also detected at several wells in the injection-observation well system. The multiple problems of plugging, high pressures, and waste leakage suggested that the reactive nature of the waste with the aquifer into which it was injected was the primary reason for the difficulties experienced with waste injection. A site study was initiated in June 1971 to investigate waste-aquifer interactions. The first stage of the study determined the hydrogeologic conditions at the site, and characterized the industrial waste and the native ground water found in the injection zone and other aquifers. The injection zone consisted of multiple permeable zones ranging in depth from about 850 to 1,000 feet (259 to 305 metres) below land surface. In addition to the injection zone, aquifers were found near depths of 60, 300, 500, and 700 feet (18, 91, 152, and 213 metres) below land surface. The aquifers from 300 feet (91 metres) down to the injection zone were flowing artesian with the natural pressure of the injection zone being 65 feet (20 metres) above land surface at the site. The dissolved solids concentration in the native ground water increased with depth to an average value of 20,800 mg/l (milligram per litre) (two-thirds that of seawater) in the water from the injection zone. Sodium chloride was the major dissolved solid, and all of the ground water below 300-feet (91-metres) depth was slightly alkaline. Dissolved organic carbon of the industrial waste averaged 7,100 mg/l and 95 percent of the organic carbon was identified and quantified. The major organic waste constituents in order of decreasing abundance were acetic acid, formic acid, p-toluic acid, formaldehyde, methanol, terephthalic acid, phthalic acid, and benzoic acid. Prior to injection, the waste was neutralized with lime to pH 4 so that the major inorganic waste constituent was calcium at a concentration of 1,300 mg/l. The second stage of the site study involved the observation of waste-aquifer interactions at various wells as the waste arrived and passed by the wells. Water samples obtained from three observation wells located 1,500 to 2,000 feet (457 to 607 metres) from the original injection well gave evidence for biochemical waste transformations at low waste concentrations. Gas that effervesced from these water samples contained up to 54 percent methane by volume. Ferrous iron concentrations as high as 35 mg/l, hydrogen sulfide gas, and sulfide precipitates were additional indicators of biochemical reductive processes in the subsurface environment. Approximately 3,000 organisms per millilitre were found in uncontaminated ground water from the injection zone whereas in waste-contaminated wells, the number increased to levels as high as 1,000,000 organisms per millilitre. High concentrations of waste were found to be toxic to microo