A Critical Analysis of Warburg Respirometry
For BO D Determinations of Polluted Waters
T.J. TUFFEY, Assistant Professor
J.V. HUNTER, Professor
Rutgers University
New Brunswick, New Jersey 08903
J.P. HEWITT, Environmental Scientist
EBASCO Services, Inc.
New York, New York
INTRODUCTION
Warburg respirometry has been successfully used to measure carbonaceous BODs on
samples of moderate to high concentration. These would include wastewaters of domestic
and industrial origin. This technique is also effective in studies of a comparative nature, i.e.,
as in determining the i oxicity of potential feed stocks on activated sludge or ascertaining the
treatability or bioavailability of new products. More detailed discussion of these topics is
available (1).
In this paper an evaluation is made of the applicability of Warburg respirometry in
determining the low levels of BOD found in polluted rivers. Errors that would be
insignificant when dealing with BODs of 100 mg/1 or greater become very important in
BODs less than 10 mg/1. This paper compares the error of the Warburg technique to that of
a technique employing standard BOD bottles and a dissolved oxygen probe. Only
carbonaceous BOD is considered.
The potential for significant levels of nitrification exists in nearly all polluted waters. A
critical objective of this research is to determine how accurately that evidence of
nitrification obtained by Warburg respirometry depicts that actually occurring in the water
bodies sampled. Is the physical environment of the Warburg flask so different from that
environment the nitrifying bacteria are exposed to in situ that they exhibit non-
representative growth patterns?
EXPERIMENTAL PROCEDURES
In all studies, Aminco Warburg respirometers were used with 125 ml flasks and 50 ml
sample volumes. 1 ml of 15% KOH was placed in the center well as a C02 absorbent.
Temperature was maintained at 20 C for the duration of the runs, usually 15-20 days.
Standard operating procedure consisted of equilibrating the sample to 20 C and
aerating it to saturation. Manometer readings were recorded to the nearest mm and
incubation time to the nearest 0.01 days for the studies on statistical variability but only to
the nearest day for the nitrification work.
In order to determine the variances of the two BOD techniques, nitrification inhibited
samples, unhibited samples, and thermo-barometers were incubated in triplicate.
Inhibition was achieved with 0.5 mg/1 of allylthiourea replenished at 7 day intervals.
Inhibited and uninhibited samples were also incubated in triplicate at 20 C with the BOD
bottle technique. Standard 300 ml BOD bottles were used and DO was recorded to the
nearest 0.02 mg/1 with a Weston-Stack DO probe calibrated daily against the Winkler (2).
During the measurement of the DO a small volume of sample was lost, less than 1 ml; this
was replaced by adding 1 ml of a similarly incubated sample. The 1 ml addition allowed the
BOD bottle to be resealed without trapping air. DO was maintained above 2 mg/1 at all
times by pulling low DO bottles off line and aerating them.