KINETIC RESPONSE OF ROTATING BIOLOGICAL CONTACTORS
A. A. Friedman, Associate Professor
Department of Civil Engineering
Tennessee Technological University
Cookeville, Tennessee 38501
R. C. Woods, Environmental Engineer
Water Quality Branch
Tennessee Valley Authority
Chattanooga, Tennessee 37401
R. C. Wilkey, Project Engineer
J. E. Sirrine Company
Greenville, South Carolina 29602
INTRODUCTION
Treatment of municipal and industrial wastewaters by rotating biological contactors
(RBC) has gained wide acceptance within the last decade.   The process is appealing to
designers because of its ability to produce a high-quality effluent, withstand shock loading variations, and its ease of operation.   Of major current interest is the relatively low
power consumption required to produce a given effluent quality.   These investigations
were initiated to evaluate the effects of various loading conditions on RBC performance
and to provide design engineers with a rational approach to RBC pilot plant testing and
prototype design.
Design practice for determining the required area of RBC media is normally based on
hydraulic loading criteria analogous to the volumetric loading rates previously used in
activated sludge system design.   Pilot plant studies are conducted to provide a correlation
between effluent quality and hydraulic loading conditions.   The hydraulic loading rate,
usually reported as gallons per day per square foot (gpdsf) of disc surface, that provides
the desirable effluent from a pilot unit is selected and used for prototype design.   The
design engineer uses his judgment to increase the required media area as a safety factor.
This design process is unsatisfactory and may lead to overdesign or operational problems since little consideration is given to influent concentration variations and their subsequent effect on microorganism kinetics.   Two different kinetic approaches to overcome
this problem have been recently reported [1-3].   The first of these is based on the apparent first-order kinetics observed when substrate concentration is plotted as a function
of stage or hydraulic detention time [1,2].   Figure 1 illustrates this apparent first-order
reaction with a portion of the excellent data collected and reported by Stover and
Kincannon [2].   A first-order reaction can be readily perceived for each independent
loading condition if one neglects the reaction occurring in the first stage.   Based on apparent first-order kinetics, RBC units have been described by the conventional plug-flow
first-order model,
Ce-Cbe^ (1)
where Cb and C   are the influent and effluent substrate concentrations respectively
(mg/1), K   is the reaction rate constant (time"1) and t is the overall detention time in
the system.   The detention time can be related to either the number of stages or discs
in the flow path.   Unfortunately, this model cannot be used to predict first-stage activity, where a major portion of the applied organics are removed.   In addition, a different
experimentally derived rate constant is required for each loading condition.
Schroeder recently proposed an empirical model based on mass transport concepts for
predicting the removal of organics in an RBC unit [3].   Organic removal per disc face
(or stage) is
420