Magnetic Flocculation of Steel Mill
Waste Waters
HENRY C. BRAMER, Vice President and Director
Data Graphic, Inc.
Pittsburgh, Pennsylvania
WILLIAM L. GADD, Senior Research Chemist
Research and Development Department
National Steel Corporation
Weirton, West Virginia
INTRODUCTION
Previous work by others has indicated that steel mill waste waters containing
suspended solids, such as result from basic oxygen furnace or blast furnace gas
washing and from rolling mill operations, can be successfully treated by flocculation
with polymers and/or magnetic forces. Most of these studies, however, have been
conducted with batch-type testing of samples collected over short time periods. This
type of testing is useful for qualitative determinations of the potentials of such
treatments on a "go," "no-go" basis; such data do not, however, provide sufficient
quantitative information to justify the investments required on the basis of effluent
concentration reductions to be expected.
The present study was conducted on site at the Weirton, West Virginia, and
Ecorse, Michigan plants of National Steel Corporation on samples continuously
withdrawn and treated and immediately analyzed. The procedures used were
designed to simulate plant operating conditions, particularly in the case where
flocculation would be used as an adjunct to an existing operation. Perhaps the
potentially most important application of this type of treatment is improving the
performance of presently overloaded or underdesigned sedimentation basins. In such
instances conventional flash mixing and flocculation mixing are not provided due to
space and cost limitations; treatment is typically accomplished in the inlet feed lines
to sedimentation basins. Such methods are effective because of the generally higher
degree of turbulence within the basins than would be found in conventional designs
such as in sewage treatment plants.
The procedures developed by Bramer and Hoak (1,2,3,4,5) were utilized in the
present study so that the analytical results could be translated into anticipated basin
performance, if the indicated treatments were to be utilized. The "sedimentation
index" (SI) measures the performance of a sedimentation basin and is equivalent to
the time, in minutes, required to achieve a similar degree of separation in an
empirical laboratory settling test. The test is conducted with a suspension that is the
same as that of the basin inlet, i.e., one which has been treated in the same manner.
SEDIMENTATION INDEX DETERMINATIONS
A settling rate curve for use in the sedimentation index determination is
constructed by allowing samples of a suspension to settle quiescently in one-liter
-154-