Inrernuriond Journal of Applied Rudiurirm und Isnroper Vol. 31. pp. 761 lo 767 0 Pergamon pless Ltd 1980. Printed in Great Britain Radiotracer Measurement of Particle Deposition and Reentrainment in an Electrostatic Precipitator R. M. FELDER* and E. ARCE-MEDINA Department of Chemical Engineering, North Carolina State University, Raleigh, NC 27650, U.S.A. (Received 3 March 1980; in revisedform 22 April 1980) Irradiated fly ash was injected at the inJet of an electrostatic precipitator, and the subsequent activity level on a collection plate was determined as a function of axial position and time. The results provide a direct measure of the particle deposition profile, and of the local rate of reentrainment. This type of experiment may be used to obtain quantitative information on the effects of dust properties and precipi- tator operating variables on dust collection efficiency, and to provide data for testing mathematical models of precipitator performance. Intruduction IN A CONVENTIONAL electrostatic precipitator (ESP), negatively charged vertical wires are suspended between grounded vertical parallel plates. A dust- laden gas passes between the plates. A small number of gas molecules subjected to naturally occurring radiation are ionized. The free electrons in the intense potential field near the wire are accelerated rapidly, and in their motion away from the wire ionize more gas molecules, creating an avalanche effect (the cor- ona discharge). The free electrons produced in the discharge are eventually captured by electronegative atoms ‘in the gas; the negative ions in turn attach themselves to dust particles, which migrate and adhere to the plates. The design and scale-up of electrostatic precipita- tors is generally performed by analogy with similar operating units. However, it is rare to find precisely analogous conditions (gas flow, temperature and pressure, dust loading, composition, electrical resis- tivity, size distribution, etc.) in two different process streams, so that some degree of extrapolation is required. Several empirical correlations of collection efficiency with various experimental parameters exist,“’ but they are imprecise, and fairly conservative overdesign procedures must therefore be used to arrive at a final precipitator design. The ideal basis for design of an ESP would be a mathematical model that predicts collection efficien- cies for specified values of the gas flow rate, dust load- ing, physical properties of the gas and dust, and geo- metrical and electrical characteristics of the precipita- tor. Several such models have been formulated, of which the most prominent is that developed by researchers at the Southern Research Institute.‘2) In their present state of development, however, none of . l To whom correspondence should be addressed. the models can provide more than a qualitative esti- mate of the collection efficiency for a given set of operating conditions. This inadequacy could be due to several factors. The assumptions that must be made regarding electrical field distributions, particle charging mechanisms, and ion and charged dust par- ticle migration rates could lead to errors in the pre- dicted collection of particles. On the other hand, the basic theory for collection on bare plates might be sound, but the effect of the dust layer resistivity on collection might not be properly taken into account by the model. Moreover, non-ideal effects such as reentrainment and sweepage, being extremely difficult to quantify, can only be accounted for by introducing arbitrarily adjustable parameters, thereby decreasing the general utility of the model. A difficulty associated with testing precipitator models is that generally only overall collection ef- ficiencies are available to compare with model predic- tions. Unfortunately, the overall efficiency is a rather gross testing criterion, although it is the quantity of primary interest in judging ESP performance. Most models of even moderate complexity predict local col- lection efficiencies in the absence of m-entrainment. If reasonably accurate local deposition profiles could be obtained, inadequacies in the model could be pin- pointed to an extent not otherwise possible. Radiotracers provide the means to obtain detailed information on the fate of labelled dust injected at the inlet of an electrostatic precipitator. The distribution of tracer collected on either bare plates or plates already coated with a dust layer can be determined by monitoring the activity on the plate following injec- tion of the labelled dust. Furthermore, by monitoring the activity of collected tracer as a function of time, and correcting for natural tracer activity decay, the rate and extent of reentrainment under different ex- perimental conditions can be determined. 761