Increasing Motor Octane by Catalytic Means, Part 2
As discussed in Part 1, the chemical composition of the gasoline determines its octane quality. High octane components are represented by branched olefins, isoparaffins and aromatics. To maximize octane through catalytic means, the refiner has to judiciously choose a catalyst that will increase the concentration of these hydrocarbons.
Since octane is of greatest concern, an "octane" catalyst would be selected. The octane enhancement that the refiner will obtain is derived from the "octane" cataIysts ability to reduce the ratio of hydrogen transfer to cracking reactions. In turn, this shift will change the gasoline composition and octane. The ratio of these important reactions can be varied by changing specific physical and chemical properties of the catalyst.
Catalyst properties that are critical in controlling the principal reactions are:
Controlling these properties in equilibrium catalyst can significantly change the gasoline octane.
Since the catalyst is composed of two basic components, zeolite and matrix, their contributions to octane will be discussed separately.
In the recent past, several investigators (Ref. 1,2,3) reported on the zeolite properties and their effects on catalyst selectivities and product quality. These investigations provide the mechanism for generating high octane hydrocarbons. It has been demonstrated that zeolite unit cell size (UCS) has a strong effect on catalyst selectivity and resultant product quality (Ref. 4). The measurement of the UCS parameter indicates the molar silica/alumina (Si/AI) ratio in the zeolite and this in turn sets the number, density and acid site strength (Ref. 5). Once the cracking sites are defined, the hydrogen transfer and cracking reaction ratio is established. This would suggest that measurement of equilibrium cataIysts' UCS defines the hydrogen transfer/cracking properties of the zeolite in the FCC unit. Measurement of this parameter can serve as a guide to catalysts' octane making potential. A refiner should be able to observe octane differences when comparing operations between 24.4 A and 24.3 A UCS of equilibrium catalysts. The lower the unit cell size zeolite, the lower the acid site density and the higher the proportion of strong acid sites. This lowers bimolecular or hydrogen transfer reactions. As a consequence, the gasoline octane is higher.
Correlations relating UCS to Motor Octane Number (MON) are illustrated in Figures 1 and 2. Figure 1 illustrates the dependence of full boiling range gasoline MON on zeolite UCS while Figure 2 illustrates this effect on light and heavy fraction gasolines.
Parameters Affecting UCS
Catalysts as manufactured without ultrastabilization have a UCS of approximately 24.72A. Those that have gone through additional processing steps for ultrastabilization will have a UCS of 24.45-24.65A. These catalysts will dealuminate in the unit with the resultant UCS reduced and stabilized in 24.27-24.40A range. The stabilization in the unit occurs rapidly as shown in Figure 3. The other point of interest in the data is that the catalysts tested had initial UCS varying from 24.48 to 24.72A. However, all equilibrated at the low values of 24.28 to 24.33A. This observation implies that regardless of the initial starting UCS, the catalysts equilibrated at similar levels if their chemical compositions were similar. The noted equilibrated UCS differences of the above samples would be reduced if they had similar Na2O levels.
Rare Earth Oxides
Rare earth is exchanged into the zeolite to either stabilize it or promote hydrogen transfer reactions for maximum gasoline yield, however, rare earth oxides (ReO) exchanged into the zeolite structure prevent dealumination. The resultant UCS of equilibrium catalyst is higher relative to non-ReO catalyst as illustrated in Figure 4. The data plotted indicated that UCS is directly proportional to ReO level of the catalysts. The scatter observed reflects Na2O differences between various catalysts tested. The higher ReO level increases the number and density of acid sites in the zeolite. The higher density increases the hydrogen transfer reactions and as a consequence, gasoline octane number decreases. Pilot plant data illustrating the octane response to ReO oxide level on light and heavy gasoline fractions are shown in Figure 5. There is a MON benefit for both fractions of the gasoline. The laboratory data suggests octane gain of 0.6 MON per 1(wt %) change in ReO level. The MON/ RON ratio is approximately 0.6.
Sodium oxide present in the zeolite neutralizes the strongest acid sites and prevents dealumination. Evidence suggests that the former is desired in producing octane by cracking and isomerization reactions and the latter controls the hydrogen transfer reactions. Therefore, low sodium zeolite is highly desirable to improve gasoline octanes. Pilot plant data (Ref. 6), shown in Figure 6 illustrates the MON sensitivity to sodium oxide level in the zeolite. The data show two different octane responses to the sodium. Sodium lowers octane at a rate of only 0.02 MON/0.1% Na2O in the region exceeding 0.3 wt % Na2O on catalyst. However, at sodium levels below 0.3 wt %, the loss in octane is 0.7 MON/0.1% Na2O.
Some investigators (Ref. 1) observed this phenomenon in their testing and postulated that the break in the curve occurs at a point where all of the highly acidic sites are neutralized by soda. The response to sodium is high when these sites are being neutralized but once that is accomplished the octane penalty for sodium diminishes.
Commercially, octane losses of approximately 0.3 MON/0.1 wt % change in Na20 have been observed. The MON/ RON ratio was approximately 0.4.
Matrix cracking sites are more amorphous in nature and crack hydrocarbons with lower hydrogen transfer capability than zeolite sites. As a result, the gasoline produced through this mechanism is normally more olefinic. Pilot plant data suggest 0.15 MON/ 10m/g change in equilibrium matrix surface area. The MON/ RON ratio for this effect is approximately 0.4.
During the past few years, the use of ZSM-5 technology has been tested commercially in numerous units for the purpose of improving gasoline octanes. The results indicate gains in MON of 1-1.5 numbers, depending on the concentration of the ZSM-5 additive in the circulating inventory, feedstock quality and initial MON.
ZSM-5 is a shape selective zeolite with a super cage opening of 5-5.6A. The small opening restricts access of high octane aromatics, naphthenes and branched chain hydrocarbons to internal cracking sites. However, low octane straight chain precursor molecules in the light gasoline are preferentially cracked to propylene and butylenes. The low octane precursor components in the heavy gasoline and front end of light cycle oil are preferentially cracked to light gasoline. The selective cracking of low octane components out of gasoline boiling range, results in concentration of higher octane branched hydrocarbons and aromatics. The net result is an increase in octane. Commercial data at 1.7% concentration of the ZSM-5 additive in the unit have typically shown 1.2 MON increase and a MON/ RON ratio of 0.7. Similar observations have been noted in other trials. The octane response is proportional to the concentration of the ZSM-5 additive in inventory as shown in Figure 7.
There are several options available to the refiner on how to improve MON. The most responsive variables were:
The effect reviewed considered one variable at a time and assumed no unit limits existed. The refiner should consider unit limitations prior to undertaking changes in order to gain octane.
1. K. Rajagopalan and A. W. Peters,Journal of Catalysis 106, p.410 (1987).
2. K. Rajagopalan, A. W. Peters and G. C. Edwards, Applied Catalysis 23, p. 69 (1986).
3. A. Corma, E. Herrero, A. Martinez and J. Prieto, paper presented at A.C.S. Symposium on Advances in FCC, New Orleans Meeting, Aug. 30 -Sept. 4, 1987.
4. L. A. Pine, P. J. Maher and W. A. Wachter, Journal of Catalysis 85 (1984).
5. A. Corma, V. Fornes, F. V. Melo and J. Merrero, Zeolites, Vol.7, pp.559-563 (1987).
6. S. M. Brown, W. J. Reagan and G. M. Woltermann, U.S. Patent 4,325,813.