With the U.S. refining industry entering the reformulated gasoline era, a significant shift in the disposition of refinery light olefins (C3-C5) Must take place. 1 2
In this effort, there will be a substantial realignment of the alkylation unit's olefin feedstock, according to Stratco Inc. of Leawood, Kan.
Isobutene will be upgraded to methyl tertiary butyl ether (MTBE) to help provide the oxygen required for reformulated gasoline. Some of the propene that has historically been polymerized into gasoline will be directed to alkylation.
Stratco sees the major change in light olefins processing as the upgrading of pentenes that are now left in the gasoline product. This will be required for two reasons:
- Pentenes are one of the major ozone precursors in today's gasoline. Upgrading them in the alkylation unit will be a major step toward meeting the new volatile organic compound (VOC) emission reduction specifications.
- Isopentenes are an important source of potential oxygenates via conversion to tertiary amyl methyl ether (TAME).
Stratco undertook a series of laboratory experiments to more accurately predict alkylate quality and acid catalyst consumption for pentene alkylation using sulfuric acid as the catalyst.
Ken E. Kranz and Kenneth R. Masters of Stratco reported on these experiments at American Chemical Society Division of Petroleum Chemistry's symposium on alkylation, aromatization, oligomerization, and isomerization of short chain hydrocarbons over heterogeneous catalysts, Aug. 25-30, 1991, in New York City.
ALKYLATION CHEMISTRY
In general, the olefins are thought to react first with the sulfuric acid to form isoalkyl esters.
These esters later release the olefin to react with the isobutane to form the alkylate product.
Competing reactions, such as olefin polymerization, can result in the production of high boiling point, low-octane product. These reactions take place in an acid-continuous acid/hydrocarbon liquid emulsion.
Acid soluble conjunct polymers, sometimes called acid soluble oils (ASOs), or red oils, are produced by olefin polymerization from isoalkylester decomposition and by reactions between isoparaffins in the alkylation reactor.
Four major chemical mechanisms describing the alkylation process have been discussed by Albright and others. 3-6 Mechanism 4 is a hydrogen transfer self-alkylation reaction, where the olefin is saturated by a hydrogen ion transfer from the ASO or isobutane.
When pentenes are alkylated, a considerable amount of isopentane and trimethylpentanes are produced via Mechanism 4. With sulfuric acid alkylation, Mechanism 4 is only important with pentenes.
PILOT PLANT
A bench-scale, continuous-alkylation pilot plant containing a well-stirred 450-ml reactor and a 1,300-ml settler was used for this study. The hydrocarbon feed for each experiment, including the required excess isobutane, was premixed in an external storage drum and charged through the pilot plant on a once-through basis.
A synthetic sulfuric acid catalyst was prepared by sparging normal butenes through fresh sulfuric acid until sufficient ASO was formed to reduce the acidity to the desired range.
Composition of the alkylate product and its octane ratings are based on periodic analysis of the product with a gas chromatograph. A computer program then calculated the amount of each compound and estimated the RON and MON based on known octane ratings of individual isoparaffin isomers. 7
Historically, these calculated octanes are about 1.01.5 numbers below the measured octanes obtained with the standard ASTM knock engine test. This is because the calculated octanes do not recognize the synergistic effect on the overall octane when the various isoparaffins are mixed into a single alkylate product.
Acid strength was determined by periodic titration of acid samples.
EXPERIMENTATION
A typical refinery's fluid catalytic cracker (FCC) stream contains four olefin isomers with an approximate distribution of 28% isobutene, 22% 1-butene, 30% trans-2-butene, and 20% cis-2-butene. Isobutane is present in approximately the same amount as the total butenes.
Some normal butane is present, but it does not enter into the alkylation reaction. The mixed C4S will also contain approximately 0.250.30% or more 1,3-butadiene, which reacts quite rapidly with the sulfuric acid to form ASO.
To simulate current commercial refinery operations, samples of FCC gasoline were obtained from several refineries and analyzed for all C5 hydrocarbons. Table 1 shows the average of these analyses.
As is the case for a mixed C4 stream, the mixed C5 stream contains nearly equal amounts of saturated and olefin isomers. The major difference is that the C5 stream contains almost 1.0% C5 diolefins.
In commercial operations, additional steps over those currently practiced will need to be taken to reduce the volume of the C5 diolefins allowed to reach the alkylation unit.
RESULTS
Several experiments were conducted using propene, mixed butenes, and mixed pentenes as the olefin feed. These showed substantial differences in the alkylate produced (Table 2).
These comparisons show that propene alkylation produces mainly dimethylpentanes along with some trimethylpentanes and C10 isomers. Butenes produce mainly trimethylpentanes and small amounts of light and heavy isomers.
Pentenes produce nearly equal amounts of trimethylpentanes and trimethylhexanes along with higher amounts of isopentane and C10+ isomers.
It is important to emphasize that the results for the mixed pentene alkylation are based on a diolefin-free, mixed C5 feed, like that shown in Table 1.
A limited number of additional experiments have been made, wherein the relative distribution of pentene isomers is different from Table 1. These experiments show significant differences in the isoparaffin distribution of the alkylate, particularly in the production of isopentane, C9s, and C10+.
One unexpected result from these experiments is the inability to predict the results of mixed pentene alkylation by linear blending of the results from the individual pentene alkylations.
This is probably because the isopentane in the mixed C5 feed suppresses Mechanism 4. The result is a shift in the isoparaffin distribution in the alkylate with slightly lower calculated octanes than would be predicted.
The second major difference is significantly higher-than-predicted acid consumption with the mixed feed.
Stratco's experiments indicate that the mixed pentene alkylate octane and acid consumption are slightly higher than those of propene. The mixed pentene alkylate's ASTM engine octanes range from 90 to 93 RON and 88.5 to 90 MON. Mixed pentene acid consumption is between 0.8 and 1.2 ppg of true alkylate.
Much of the difference in the mixed pentene alkylation acid consumption values reported above and those reported in the past centers around the basis used in the calculations.
Mixed C5 feed, as produced in the refinery, contains significant amounts of isopentane and normal pentane. Because these paraffins act as diluents in the alkylation reactor, they appear in the total C5+ material or whole alkylate measured after the alkylation reactor.
Thus, if the acid consumption calculations are made on a whole alkylate basis, the value is substantially smaller for the mixed pentene case. This difference in volume between whole alkylate and true alkylate is only significant in mixed pentene alkylation, and appears to have led to considerable confusion in the past.
BUTENE/PENTENE PATIO
A series of experiments was conducted in which the ratio of mixed butenes to mixed pentenes in the olefin feed was varied. Based on these experiments, there appear to be no benefits or penalties associated with combined or separate alkylation of the C4/C5 feed (Table 3).
Overall, the acid consumption increases nearly 0.08 ppg of true alkylate for every 10 wt % of pentenes in the feed.
The octanes decrease an average of about 0.44 RON and 0.38 MON for every 10 wt % pentenes.
It is thought that many refineries will operate with about 25-30% mixed pentenes in their alkylation feedstock.
OPERATING CONDITIONS
For butene alkylation, lower reaction temperatures and lower space velocities (longer residence times) result in a higher alkylate octane and reduced acid consumption. Within reasonable ranges (up to 9/1 for sulfuric acid alkylation), higher I/O ratios improve results.
Mixed pentene alkylation seems to follow the same trends.
However, the absolute effect of changes in these operating conditions is significantly different.
For instance, with mixed butene alkylation, a reduction in the reaction temperature from 15 C. to 4.5 C. will result in a marked improvement in the alkylate's octane.
However, the pentene alkylate octane shows only a minor response to reaction temperature changes.
Conversely, the acid consumption with the pentene feed was significantly improved at the lower temperature. With butenes, acid consumption is only marginally affected by reaction temperature.
Lowering the olefin space velocity and increasing the I/0 ratio had little effect on either the alkylate octane or acid consumption for the pentene alkylation. Such changes with butenes will measurably improve alkylate octane and reduce acid consumption.
FEEDSTOCK CHANGES
The combination of an MTBE unit upstream of an alkylation unit has been practiced for several years (OGJ, Nov. 18, 1991, p. 99).
For sulfuric acid alkylation, this combination removes the least desirable butene isomer from the alkylation feed and results in an alkylate octane improvement of up to 1.0 number. There is little change in acid consumption when using MTBE raffinate as the olefin feed.
There is almost no commercial experience with a TAME unit upstream of an alkylation unit. Thus, some limited work was conducted in Stratco's laboratory with simulated TAME unit raffinate.
The simulated TAME raffinate was prepared based on the reported conversion of 90% 2-methyl-1-butene, 60% 2-methyl-2-butene, and 0% 3-methyl-1-butene to TAME. 1
The results of these experiments are opposite from what is experienced with MTBE raffinate. With sulfuric acid catalyzed alkylation, no improvement was found in alkylate octane from processing the TAME raffinate in lieu of the mixed pentenes.
A measurable decrease in acid consumption was noted when processing TAME raffinate. However, oxygenate carryover from the TAME unit to the alkylation unit may reduce that benefit, as it does with an MTBE unit.
REFERENCES
- Cosyns, J., Nocca, J.L., Keefer, P.S., and Masters, K.R., "Ultimate C,/C5 Olefin Scheme for Maximizing Reformulated Gasoline Production," Paper No. AM-91-50, National Petroleum Refiners Association Annual Meeting, Mar. 18-21, 1991, San Antonio.
- Masters, K.R., "Alkylation's Role in Reformulated Gasoline," National Conference on Octane Quality and Reformulated Gasoline, 1991.
- Albright, L.F., Spalding, M.A., Faunce, J., and Echart, R.E., "Industrial Engineering Chemistry Research," 1988, Vol. 27, p. 391.
- Hofmann, J.E., and Schriesheim, A.J., ACS, 1962, Vol. 84, p. 953.
- Schmerling, L., Chemistry of Petroleum Hydrocarbons, B.T. Brooks, et al., editors, Reinhold, New York, 1955, Vol. 3, p. 363.
- Albright, L.F., Kranz, K.E., "Alkylation of Isobutane with Pentenes using Sulfuric Acid as a Catalyst," unpublished manuscript, 1991.
- Hutson, T., and Logan, R.S., Hydrocarbon Processing, September 1975, p. 107.
Copyright 1992 Oil & Gas Journal. All Rights Reserved.
