AM-93-17

ISO-OLEFINS FOR OXYGENATE PRODUCTION USING ISOPLUS™

By

J. B. Mclean, Technology Specialist
BASF Corporation
1800 St. James Place Suite 501
Houston, Texas 77056

and

A. Witoshkin, Director of Cracking Technology
D.C. Bogert, Senior Account Manager
BASF Corporation
101 Wood Avenue
Iselin, New Jersey 08830

Presented at the

1993 NPRA
ANNUAL MEETING
March 21-23, 1993
Convention Center
San Antonio, Texas

As refiners enter the reformulated gasoline era, it is clear that the FCCU will play a major role as a producer of iso-olefins to be used as feedstocks for oxygenate production. BASF has developed and commercialized a new family of IsoPlus™ FCC catalysts designed to maximize yields of isobutylene and isoamylenes(Ref. 1). The IsoPlus 1000 series is designed to aid refiners currently operating MTBE and/or TAME units with spare capacity by providing isobutylene and isoamylene yield increases of up to 50 %, compared to previously available USY based catalysts, while fitting into typical unit operating constraints. The IsoPlus 2000 series is designed to offer refiners an option for the future when even higher iso-olefin yields will be required. IsoPlus 2000 series catalysts can increase iso-olefin yields by 100 % or more when used in combination with unit hardware upgrades to ease constraints typical of current operations.

Some significant developments have occurred in the past year to advance this new catalytic technology closer to widespread commercial practice. Laboratory studies at BASF have confirmed the selectivity benefits of the IsoPlus catalysts on a wide variety of commercial FCC feedstocks, and tests conducted by outside laboratories have also supported these results. An economic evaluation of the commercial potential for these catalysts conducted by a major independent contractor showed encouraging results. Both IsoPlus 1000 and IsoPlus 2000 have been manufactured commercially and used successfully in commercial FCC units, with preliminary results confirming expectations. Each of these developments will be discussed in this paper.

At the same time, the past year has been one of uncertainty about future direction in the refining industry. The pace of movement toward the type of operations which reformulated gasoline oriented catalysts are geared to has not been as fast as many anticipated. The market price of MTBE has not increased substantially. It is expected that this is a temporary situation, and significant changes will still be needed to meet the more stringent regulations of the Clean Air Act for 1995 and beyond. BASF feels that IsoPlus catalyst technology will play a significant role, and is continuing R&D efforts to offer refiners choices based on catalytic options for meeting future demands.

Laboratory Feedstock Studies

BASF's development work on IsoPlus catalysts was largely based on testing with a "standard" feedstock, in this case a Mid-Continent gas oil. Realizing that the large yield shifts observed for these catalysts may well be feed-dependent, a microactivity (MAT) scale study was conducted using a range of commercial FCC unit feeds. The catalysts evaluated were the same ones for which results were previously published(Ref. 1). A commercial RE-USY octane catalyst was chosen as the base, and is typical of the type of product refiners with MTBE units are currently using. This was compared with IsoPlus 1000, IsoPlus 2000, and IsoPlus 2100, the highest isobutylene yielding catalyst in the series. The properties of the feeds tested are shown in Table 1. Feed A is the "standard" BASF gas oil. Feed B is a hydrotreated vacuum gas oil from a West Coast refinery processing a high nitrogen crude slate. Feed C, from a Mid-Continent refinery, is a highly paraffinic blend of several gas oils and PDA unit extract, and is much more reactive than the other feeds. Feed D is from a Gulf Coast resid cracker, and is a partially hydrotreated atmospheric resid which was diluted with some standard gas oil for this study to allow easier handling in the fixed bed MAT unit. Feed E is a VGO from a Far East refinery. Together these feeds represent a wide range of geographical sources, chemical compositions, boiling ranges, and processing histories. MAT conditions used were 910 F reactor temperature, 15 WHSV, and variable cat/oil to achieve a range of conversions in order to statistically correlate the yield selectivities.

Figure 1 shows the activity responses for each feed and catalyst at a constant cat/oil ratio of 5. The absolute conversions vary considerably due to differences in feed reactivity, but the relative responses of the different catalysts are similar for all feeds. The West Coast Feed B showed less of an activity debit for the IsoPlus 2000 series catalysts than the other feeds. The isobutylene selectivities are shown in Figure 2. On each feed, the catalysts are compared at constant conversion, although the conversion levels vary for different feeds due to the reactivity differences. The relative rankings of the catalysts are very similar in each case, with Feed C showing the highest potential yield for all catalysts due to its higher reactivity. The high isobutylene potential for the IsoPlus catalysts has been attributed to their low hydrogen transfer activity. As discussed in Reference 1, a key index for defining the hydrogen transfer activity is the selectivity ratio of isobutylene to isobutane. This ratio is shown in Figure 3. Once again, the relative rankings of the catalysts are very similar for each feedstock.

Since commercial performance predictions for the IsoPlus catalysts had been primarily based on test data generated on Feed A, the results of this study confirm that these predictions should be valid over a wide range of potential feedstock properties.

Independent Laboratory Evaluation

A sample of IsoPlus 1000 was tested by an independent laboratory as part of a comprehensive multi-client catalyst evaluation program. As indicated in Table 2, the results confirmed BASF's expectations. The coke and dry gas yields were in line with other commercial octane catalysts, while the IsoPlus 1000 had the highest LPG yield and correspondingly lowest gasoline yield. The isobutylene yield was the highest of any catalyst tested, more than 50 % higher than the average octane catalyst. The ratio of isobutane to total C3/C4 olefins was the lowest of the catalysts tested, confirming that the high isobutylene yield is in fact due to low hydrogen transfer activity.

Samples of IsoPlus catalysts have also been tested by most of the major oil company laboratories. While their specific results are proprietary, the general response has been to support the relative yield advantages expected. Taken in combination with the results presented here, it has been confirmed with a variety of feedstocks and testing protocols that the IsoPlus catalysts offer significant potential for maximizing iso-olefin production compared with previously available commercial catalysts.

Economic Evaluation of IsoPlus Catalysts

Commercial performance projections which have been previously reported(Ref. 1) were supplied to a major independent contractor to use as the basis for an economic evaluation of the potential benefits of IsoPlus catalysts. An integrated refinery configuration as shown in Table 3 was assumed for this study. Cases assumed that gasoline production was held constant and quality was consistent with reformulation specifications of the Clean Air Act. All available isobutylene was used to make MTBE in each case, with the balance required purchased at an assumed market price of $48 per barrel (typical of 2Q92 spot price). Revamp capital costs for the FCCU (principally wet gas compressor, gas plant, and in some cases regenerator upgrades including air blower and catalyst cooler) and expanded capacity of downstream MTBE and alkylation units were estimated along with net operating revenue changes, including reduced charges for crude and purchased MTBE for the IsoPlus cases. Estimates were also made for including the installation of an advanced riser termination device to minimize thermal cracking, thus allowing higher LPG and isobutylene yields for the wet gas constrained operation due to lower dry gas production.

Table 4 presents a summary of the results of key case studies. For IsoPlus 1000, a payout period of 1.5 years is indicated for this scenario. This becomes even more attractive when the revamp includes the riser termination modification, reducing the payout to less than one year. The maximum isobutylene case, IsoPlus 2000, is not as attractive in this scenario due to the higher capital costs required. Higher MTBE values and/or product selectivity improvements will be required for favorable economics here. Both are anticipated in the future, due to increased MTBE demand and BASF's continued R&D efforts. Another point to consider is that the FCCU upgrade investments required here could also be used to add flexibility for increased throughput or resid processing, as well as to tailor operation for maximum isoolefins.

A second economic study has been conducted as a collaborative effort between BASF and Wright Killen & Co.(Ref. 2). This study also included the additional benefits of increased isoamylene production for processing in a TAME unit. This study also showed favorable payouts of less than one year for cases based on IsoPlus 1000 and typical current economic scenarios, and verified that higher oxygenate values and/or reformulation levels can make IsoPlus 2000 based cases attractive. IsoPlus cases were also compared with the use of a ZSM-5 based additive for light of production . This approach tends to favor propylene selectivity relative to butylenes(Ref. 1). As a result, the IsoPlus cases were economically favored for reformulated gasoline due to higher potential MTBE yield at the same wet gas constraint.

IsoPlus Commercial Status

Both IsoPlus 1000 and IsoPlus 2000 have been produced commercially and successfully used in commercial FCC units. IsoPlus 1000 was used at a Midwest refinery in a UOP stacked unit processing VG0 feed. The prior catalyst in use at this location was a competitive USY octane catalyst. A preliminary summary of the results obtained is shown in Table 5. This refinery does not operate an MTBE unit and does not differentiate isobutylene from total butenes in their yield reports, so only the total C4 olefin results are presented. Due to feedstock changes during the course of the trial, the unit conversion dropped three percent with a corresponding loss in total LPG yield. However, the LPG olefinicity was up substantially, resulting in a slight increase in overall olefin yields. As indicated, the C4 olefinicity and hydrogen transfer index (as indicated by the C4=/iC4 ratio) changed substantially. The gasoline octane also increased despite the lower unit conversion, also an indication of reduced hydrogen transfer. While this trial had not yet achieved full catalyst inventory turnover, the trends evident in the data shown are in full support of the expectations for the IsoPlus 1000 catalyst.

IsoPlus 2000 is being commercially used in a Western refinery in a UOP stacked unit which has installed riser termination technology similar to that discussed in the previous section. This particular refinery runs a very paraffinic feed at very high conversions and recycles slurry oil to extinction. They do not operate an MTBE unit, but send all their C3/C4 olefins to a poly unit. Their objective for use of this catalyst was to increase distillate for winter operation (lower conversion) while maintaining octane, poly unit feed rate, regenerator temperature, and extinction bottoms recycle operation. Figure 4 shows the conversion and LPG yield trends for this trial. In this high conversion range there is substantial overcracking, so that gasoline yield is maintained even as unit conversion is reduced.

Figure 5 shows the isobutylene and isobutane yield responses. Isobutylene yield increased slightly (even at lower conversion) while isobutane decreased dramatically. This is due to the reduction in the catalyst's hydrogen transfer activity. In Figure 6, the Hydrogen Transfer Index (iC4=/iC4 ratio) is plotted, showing a dramatic change as the trial proceeded. Also plotted in Figure 6 are the ratios based on MAT testing of the equilibrium catalyst samples from the unit. The absolute values of the MAT numbers are lower than the commercial unit data, due to the inherently higher hydrogen transfer effect in the MAT unit, but the trends track very well. This confirms that the unit yield changes observed are due to changes in the catalyst selectivity and not due to feed or operating condition changes. The isobutylene approach to equilibrium (iC4=/tC4='s) also increased (Figure 7), closer to the thermodynamic equilibrium limit of 0.45. All of these trends support expectations. A more complete evaluation of this trial including pilot plant testing of equilibrium catalysts and process modelling to account for operating condition changes will be conducted when a full inventory turnover is obtained.

Summary

Several significant developments have occurred in the past year which combine to advance IsoPlus catalyst technology closer to it's expected role in the reformulated gasoline oriented FCC marketplace of the future. Additional laboratory studies have verified the potential to produce very high iso-olefin yields using a variety of feedstocks and test conditions. Economic studies indicate that there are significant potential benefits for using these catalysts in a reformulated gasoline scenario. Finally, initial results from commercial trials have verified expectations based on predictions from laboratory testing.

References

1. "Reformulated Gasoline Catalyst's Impact on FCCU", J.B. McLean, G.S. Koermer, R.J. Madon, and W.S. Winkler, 1992 NPRA Annual Meeting, Paper AM-92-45

2. "Maximizing Isobutylene Production in the FCC Unit", J.B. McLean, W.S. Winkler, R.G. McClung, and M. Feldman, The Catalyst Report, BASF Publication EC-6573, 1992