RELATIONSHIP BETWEEN LEVEL OF NITRIFYING BACTERIA
AND AMMONIA REMOVAL IN THE BIOLOGICAL TREATMENT
OF COKE-PLANT WASTEWATER
Daniel D. Jones, Associate Professor
Joseph J. Gauthier, Associate Professor
Randy H. Martin, Graduate Student
Biology Department
University of Alabama in Birmingham
Birmingham, Alabama 35294
Leon W. Wilson, Jr., Senior Research Engineer
Environmental Research Division
Research Laboratory
United States Steel Corporation
Monroeville, Pennsylvania  15146
Carl R. Majors, Project Engineer
Fairfield Works
United States Steel Corporation
Fairfield, Alabama 35064
Many industries, such as those involved in the production of coke or manufacture of
iron and steel, generate wastewaters containing ammonia. Because ammonia can cause
several undesirable effects in the waters receiving such effluents, industries must treat
their wastewaters before discharge to meet increasingly stringent federal and state water
quality standards. Although chemical and physical means of ammonia removal are practiced
by industry, biological nitrification in an activated sludge process appears to offer the most
cost-effective treatment. However, it is sometimes difficult to maintain a stable biological
nitrification  process in both bench- and full-scale systems treating industrial wastewater.
This process instability is generally attributed to the fact that the relatively slow-growing,
autotrophic, nitrifying bacteria have been found especially sensitive to a variety of environmental conditions and to toxic compounds which often occur in industrial wastewaters.
However, due to the complexity of the entire process of biological wastewater treatment,
it is not easy to determine the precise causes for the occasional increases in effluent ammonia.
In the past it has not been feasible to determine whether a breakthrough in ammonia is
due to inhibition of the nitrifying biomass or to a decrease in the nitrifying population.
To correlate population dynamics with observed changes in nitrification rates, one must be
able to enumerate the nitrifying bacteria. Several approaches have been used for determining
the biomass of nitrifiers. These include most probable number (MPN) estimates [1], 14C-
bicarbonate incorporation [2,3], pure culture equivalent activities [4], a modification of
the latter in which pure culture is added directly to mixed-liquor samples [5], and the use
of fluorescent antibody (FA) techniques [6,7].
The MPN procedure, which is a statistical method based on serial ddutions of a sample
in appropriate medium, has been shown to usually be several magnitudes too low in its
estimate, compared to the FA technique [8]. Furthermore, to obtain the best (maximum)
estimate of nitrifying bacteria, incubation periods of over 100 days are required [ 1 ].
92