Section 13. PHYSICAL/BIOLOGICAL SYSTEMS
CHEMICAL REACTOR DESIGN THEORY AND BIOLOGICAL
TREATMENT OF INDUSTRIAL WASTES—IS THERE A GAP?
Arthur W. Busch, Visting Professor
Graduate Program in Environmental Sciences
University of Texas at Dallas
Richardson, Texas 75080
INTRODUCTION
One of the earliest papers (perhaps the first) to address the importance of soluble organics and rate
aspects (kinetics) in aerobic biological treatment processes was presented by Garrett and Sawyer at
this conference 31 years ago [1], These concepts, followed by introduction of completely mixed reactors [2,3,4] led to extensive work on process characterization and design in terms of stoichiometry and
kinetics based on chemical reactor design theory. Several books are cited as useful summaries of progressive state of the art over the past 20 years [5,6,7,8]. Research focusing on this more basic approach
was greatly facilitated by development of techniques such as total carbon analysis [9] to complement
generic non-specific measurements; e.g., chemical oxygen demand (C.O.D.) and biochemical oxygen
demand (B.O.D.). Now, of course, effective research involves analysis for specific substrate, intermediates, and end products.
As is discussed later in this paper, there remains some confusion over how best to assess performance of a biological reactor; i.e., what is complete removal?
This paper addresses issues arising from application of chemical reactor design theory for first
order reactions to aerobic biological waste water treatment; i.e., activated sludge.
BACKGROUND
Chemical reactor design theory for the case of a single, irreversible, homogeneous, isothermal
(SIHI) reaction leads to two conclusions of interest to this paper:
1. To achieve the same degree of conversion (effluent quality) a continuous, stirred tank reactor (CSTR) must
be larger than a plug flow reactor (PFR) or a batch reactor (BR).
2. Aggregate volume of a cascade, or series, of CSTRs for the same degree of conversion is less than that of a
single CSTR, but greater than that of a PFR, or a BR.
Mathematical proofs of these conclusions are endemic, if not ubiquitous, in the literature [7,8]
and are not repeated here. Denbigh, in 1965 [10] provided the classic explanation, based on considering the average (sic) reaction rate in the two types of reaction environment, CSTR and PFR.
"For the SIHI case, reaction rate decreases as conversion increases; reaction rate is concentration
dependent. Thus in a PFR, concentration gradients decrease along the length of the reactor and the
average reaction rate has a value between the high initial rate and lower final rate. Because in a CSTR
concentration of reactants is everywhere equal to the effluent, the average reaction rate is set by effluent concentration. If the two systems are compared for the same effluent concentration (degree of
conversion) the PFR always has a higher reaction rate and therefore requires a smaller reactor volume."
Since the early 1960's writers have repeatedly observed that aerobic biological CSTRs performed
equally with PFR configurations [6,7,8]; that is, average reaction rates in CSTRs were found to be the
571