Hydrocarbon Pool Mechanism

Background

The reactions involved in the MTO process are catalyzed by the zeolite itself and by compounds trapped within the voids. These compounds, called the “hydrocarbon pool," are trapped because they are large relative to the size of the pores. The hydrocarbon pool is formed during a kinetic induction period and must build up from reactions of methanol, DME and other species produced initially. At steady-state the hydrocarbon pool continually reacts with methanol to yield olefins via elimination reactions. A simplified reaction network involving the hydrocarbon pool at steady-state is illustrated below:

hydrocarbon pool

Reactive species in the hydrocarbon pool include polyalkylated aromatics, large alkylated olefins, and carbenium ions. A carbenium ion has a trivalent carbocation center with three bonds and a net positive charge and is thus highly reactive. Some examples of hydrocarbon pool species are given below:

carboniumhpool1hpool2

Carbocation center (left), 3-methyl-3-hexene (center), p-diethylbenzene (right)

Benzenoid compounds in the hydrocarbon pool react with DME, methanol or other species to form higher homologues such as xylenes and ethylbenzenes. These can undergo additional reactions, including elimination to yield light olefins that can leave the catalyst pores. The original species are then re-alkylated to complete the catalytic cycle. An example of one possible step in this overall mechanism is the methylation of toluene by DME to produce xylene and methanol. The reaction path has been investigated by Svelle et. al. and is represented below. The “transition state” in the diagram is the highest energy state in the conversion of reactants to products.

tst

Schematic diagram of the reaction path for methylation of toluene by DME (Svelle et. al. J. Phys. Chem. B. 2005, 109, 12874.)

Characterization

The formation of the hydrocarbon pool as a function of temperature was characterized at steady state over the zeolite H-SAPO-34. The conversion of 13C-enriched methanol was monitored using gas chromatography and MAS-NMR spectroscopy (link to experimental techniques page). At temperatures less than 523 K, only DME and methanol were detected via 13C NMR, with 58% yield of DME. Above 548 K, alkyl, aromatic and olefinic signals were detected, indicating formation of the hydrocarbon pool. A corresponding increase in the yield of light olefins was noted in the gas chromatogram. The 13C NMR signals and percent yields from gas chromatography are labeled below:

h-pool peaks

Greater than 673 K, nearly total conversion of methanol occurred. To investigate the species trapped in the catalyst pores, the reactor was purged with inert gas. The spectra before and after the purge are shown below. The spectra are qualitatively similar and exhibit strong signals in the alkyl and aromatic regions—evidence of the presence of alkylbenzene species in the hydrocarbon pool.

c2

Catalytic Role

To investigate the catalytic role of the hydrocarbon pool, the reactant flow was switched from 13C-enriched methanol (13CH3OH) to methanol with natural isotopic abundance (12CH3OH). If the alkyl groups of the aromatic species are involved in olefin generation via elimination reactions, a decrease would be expected in the 13C abundance of the alkyl groups after the feed change. The elimination of 13C-rich ethene from a hydrocarbon pool species, followed by subsequent re-methylation, is illustrated in the scheme below:

c3

The intensity of the 13C spectrum in the region of the alkyl carbon atoms was monitored. Spectra from before and after the feed change are given below. A 42% decrease was observed in the integrated intensity of the alkyl region peaks. The 1H NMR spectrum showed no significant change, suggesting that the methyl group carbon atoms were still present, but contained 12C rather than 13C. Based on these results, it was concluded that the aromatic species—specifically, ethylbenzenes—play an active role in the catalytic cycle of methanol conversion on H-SAPO-34.

c4

Seiler et. al. Catalysis Letters. 2003, 188, 187.
Song et. al. J. Am. Chem. Soc. 2001, 123, 4749.