72    BIOLOGICAL DENITRIFICATION OF HIGH NITRATE
INDUSTRIAL WASTEWATER
Ruth Hund, Graduate Student
Linda Figueroa, Assistant Professor
Nevis Cook, Assistant Professor
Colorado School of Mines
Boulder, Colorado 80401
Katherine Cutter, Graduate Student
JoAnn Silverstein, Associate Professor
Jennifer Deters, Undergraduate Student
University of Colorado
Boulder, Colorado 80401
INTRODUCTION
The processing of actinides by the Department of Energy's (DOE) Rocky Flats plant produces a
liquid waste stream with high concentrations (greater than 1,000 mg-N03"/L) of sodium and potassium nitrate salts. Nitrates potentially interfere with the cementation process necessary for the disposal of low level radioactive wastes. Therefore, it is necessary to remove the nitrates from the water
in order to properly dispose of the waste. Biological denitrification of high calcium nitrate wastes has
been successfully demonstrated at the Y-12 plant in Oak Ridge, Tennessee, but the rate was variable.'
It was hypothesized that this variable rate was due to lack of pH control.
Denitrification increases the pH by producing alkalinity.2 To keep the pH within the optimal range
of the denitrifiers the pH must be controlled by the addition of acid. Two reactor systems with pH
control and a third system with no pH control were evaluated to test the hypothesis and determine
which one provided the maximum denitrification rate. The first system used carbon dioxide and the
second used hydrochloric acid to reduce the pH. Both systems used an automated controller that
added acid to the reactor when the pH exceeded 7.9. The controller stopped adding acid when the pH
went below 7.4.
The purpose of this study was to determine which pH control system provided the overall best
performance. Best performance was a combination of the highest denitrification rate and a consistent
Mixed Liquor Suspended Solids (MLSS) concentration above 3000 mg/L. It was also hypothesized
that the biological denitrification rate of a saline industrial waste would be zero order with respect to
nitrate and first order with respect to MLSS concentration.
METHODS AND MATERIALS
Three anaerobic Sequencing Batch Reactors (SBRs) were operated from September 14 to October
23, 1992 in the Environmental Engineering Laboratory at the University of Colorado, Department of
Civil, Architectural and Environmental Engineering in Boulder, Colorado. The reactors were seeded
with waste activated sludge from the Boulder Wastewater Treatment Plant in Colorado and immediately fed the industrial strength feed used for the experiment. The bacteria acclimated to the synthetic
waste within three days.
Three 10 L cylindrical reactors were constructed of Plexiglas with five ports bored in the top, one
pH probe port along the wall toward the top, two ports for recirculation, and a spare port toward the
bottom. The reactors had a maximum liquid volume of 9 L. A schematic of the SBR design is
represented in Figure 1.
The pH probes were connected to Cole Parmer 5656-00 pH/ORP controllers that triggered the
addition of acid to the SBRs when the pH exceeded 7.9. The addition of acid ceased when the pH went
48lh Purdue Industrial Waste Conference Proceedings, 1993 Lewis Publishers, Chelsea, Michigan 48118. Printed in
U.S.A.
723