Effect of FCC Catalyst Density and Attrition Index on Stack Opacity
By Ronald G. McClung
This article is the second in a series of three Catalyst Reports dealing with quantification of FCC physical property effects on emissions measurements. The previous Catalyst Report dealt with stack emissions as affected by density in cyclone operation. This Catalyst Report deals with the measurement of opacity as a function of FCC physical properties. A model is developed that allows quantification of attrition index, particle size distribution and catalyst density effects on the opacity measurement. Catalyst density and attrition index effects on opacity will be covered by this report. The next report will deal with the effect of stack fines particle size distribution (PSD) on opacity.
When stack opacity first became an issue of environmental importance, the measurements were done by a trained smoke observer. The trained reader used a Ringelmann scale to rate opacity from 0 (low) to 5 (high).
A Ringelmann number of 1 is defined as equivalent to the 20% opacity measurement, 2 is 40%, etc. These determinations were subject to a number of sources of error including background haze, windage, presence of steam, and the subjective judgment of the reader. In addition, since the opacity standards today are in the 1/2 to 1 range on the Ringelmann scale, accuracy of such a subjective measurement is clearly inadequate.
The Ringelmann scale still remains the reference for most government regulations, but light transmittance instrumentation has taken away much of the subjectivity and inaccuracies at the low end of the scale.
Opacity measuring devices are calibrated for the same response range in the electromagnetic spectrum as the human eye. The range of wave lengths used in the opacity measurement must exclude those that would be absorbed by CO2 and water. The range used is relatively narrow at 400-700 nanometers which also then allows a determination of the particle size "seen" by the instrument. Such instruments usually give maximum response in the 500-600 nanometer range.
Figure 1 shows the effect of particulate diameter on light attenuation, which correlates with stack opacity. Note that the maximum light attenuation occurs between 0.2 to 0.8 microns, particle sizes which certainly do not exist in fresh FCC catalysts.
Thus, the particle sizes which contribute most to the measurement of opacity are products of catalyst attrition: the lower the attrition resistance of a catalyst the higher the propensity to produce fine particles that increase opacity. Note also that the light attenuation is about the same for a 0.2 micron particle as for a 10 micron particle, giving the opacity measuring device sensitivity to a broad range of fine particle sizes.
A mathematical model has been derived for the general case of opacity as a function of particulate physical properties(Ref. 1).
That expression is as follows:
EQN. 1 OP = 1 - EXP (- BE D)
where D is the stack diameter in units consistent with BE which is defined by the following integral expression:
The n(r) function is relatively simple to obtain from commercial data, illustrations of which are given in Figure 2.
The more complex QE function is pictured in Figure 3 and will be dealt with in a subsequent Catalyst Report giving the effect of particle size distribution on opacity.
When expressed in numerical terms, Equation 2 transforms to the expression given in Equation 3, and the opacity equation to that in Equation 4.
Note that one physical property (ABD) is expressly stated as impacting opacity. The effects of attrition index are implied in the W term and those of particle size distribution, in the QE term. The attrition index direct relationship of attrition index to W, the pounds of particulates per unit of stack gas, is implied by the common knowledge that a more attrition prone catalyst increases stack opacity. Also, the mixture of particulates with the lowest light extinction will give the lowest opacity. For the purposes of this article, will be considered a constant of 0.58.
Model Input Data
The total particulate density (W Ib/cu.ft.) will require use of some commercial data. In practice, this value would be obtained by stack sampling on the FCC unit. There is some guidance for calculating this value in open literature(Ref. 2). The values for total emissions (Ib/hr) and calculated concentrations (dry basis) in the stack flue gas are given in Table 1.
Average fresh catalyst physical property values for the three U.S. FCC catalyst manufacturers are given in Table 2. Density is expressed as both ABD and CBD. Which density is used for comparison among the catalysts is not material, so long as the same measurements are used for each catalyst. The void fraction for all cases will be 0.42. Attrition Index is expressed as two parameters, EAI (BASF Attrition Index) and Roller Number. Both will be used for calculating impact on opacity measurement. The stack diameter chosen will be 3 ft.
Some additional information is required in order to make application. A primary assumption is the relationship between an attrition index and stack losses, or W (total solid particulates per unit volume of stack gases).
The effects on opacity of each of the two properties given in the prior sections, are illustrated in Figures 4, 5 and 6. The base case for all these figures is:
Using the ranges of fresh catalyst properties presented in Table 2 and the foregoing mathematical model, the impact on opacity of density and attrition indices can be calculated. The relative percentage changes in opacity are given in Table 3.
Clearly, attrition index has the largest impact on opacity. Because the exponent n is not as well known for Roller as for EAI, opacity impacts for several values of n are given. When the Roller's corresponding to the EAl's given in Figure 5 are used, the exponent value calculates to n = .625 for the Roller measurement.
1. Pilat, Michael J. and Ensor, David S., "Plume Opacity and Particulate Mass Concentration", Atmospheric Environment, Pergamon Press, 1970, Vol. 4 pp. 163-173
2. Air Pollution Engineering Manually 2nd Edition, U.S. Environmental Protection Agency, Research Triangle, N.C., May 1973, p. 667, pp. 913-915