Recent Developments
Solid Catalyst Alkylation
Researchers have long sought to develop a solid catalyst alkylation (SCA) process
that would eliminate the safety and corrosion issues associated with HF or sulfuric
acid. Early attempts used supported metal halides, including work by UOP (Kojima
and Kocal 1995) and Total (Nascimento et al. 2001) with chlorided alumina; work
by Mobil with boron trifluoride supported on silica (Chou et al. 1990a), MCM-25
(Huss and Johnson 1993), or beta zeolite (Chou et al. 1990b); and work by HaldorTopsoe (Angelis et al. 2001) with triflic acid supported on silica. More recent
efforts have focused on wide-pore zeolites.
Zeolites have about an order of magnitude lower acid site concentration per gram
than liquid acids and occupy about ten times the volume per gram ($5 cm
3 vs. 0.5 –
1.0 cm
3
/g). This results in significantly larger catalyst volumes required for equivalent reaction. Studies (Nivarthy et al. 1998; Stocker et al. 1994) have shown that
stronger Brønsted acid sites are necessary to effectively catalyze hydride transfer,
which is the reaction step that has the biggest impact on alkylate product quality and
catalyst life. The strength of the acid sites may be modified by dealumination (Corma
and Orchilles 2000) and rare earth exchange (Weitkamp 1980).
A key difference between zeolites and liquid acids is their selective and strong
chemisorption of olefins. Due to the high polarity of the zeolitic surface, especially
in those with a low Si/Al ratio, polar olefin molecules are preferentially adsorbed.
Localized higher olefin concentrations in the pores will result in higher relative
rates of oligomerization than hydride transfer and is believed to be the main reason
why zeolites deactivate so rapidly. Larger molecules are more strongly adsorbed
than smaller ones, so once a heavy molecule is formed, it is unlikely to desorb,
leading to the formation of coke, which will ultimately block the zeolite pore. The
key to running a successful solid catalyst process thus becomes finding a way to
regenerate the catalyst and remove the coke in a timely fashion.
Most SCA processes use a combination of “cold” (liquid phase) and “hot” (gas
phase) regeneration procedures. This requires the incorporation of a noble metal
hydrogenation function in the catalyst. The cold regeneration is essentially a
solvent extraction with isobutane saturated with hydrogen. At low temperatures,
(<100
C) most unsaturated species that block the acid sites are hydrogenated and
can be more easily desorbed from the catalyst. However, the cold regeneration only
partially restores activity, and after a number of cycles, it is necessary to do a hightemperature regeneration with hydrogen. At higher temperatures, there is some
hydrocracking of species that are too bulky to leave the pores into smaller species
that can diffuse out of the pores, thereby restoring pore volume and access to
acid sites.
Several different reactor schemes have been proposed for accomplishing the
solid catalyst alkylation process and regeneration. The CB&I/Albemarle/Neste Oy
AlkyClean
TM process (Gieseman et al. 2006) uses a swing bed strategy. In a typical
unit design, there are five identical reactors. At any one point in time, two reactors
Alkylation in Petroleum Processing
453
Solid Catalyst Alkylation
Researchers have long sought to develop a solid catalyst alkylation (SCA) process
that would eliminate the safety and corrosion issues associated with HF or sulfuric
acid. Early attempts used supported metal halides, including work by UOP (Kojima
and Kocal 1995) and Total (Nascimento et al. 2001) with chlorided alumina; work
by Mobil with boron trifluoride supported on silica (Chou et al. 1990a), MCM-25
(Huss and Johnson 1993), or beta zeolite (Chou et al. 1990b); and work by HaldorTopsoe (Angelis et al. 2001) with triflic acid supported on silica. More recent
efforts have focused on wide-pore zeolites.
Zeolites have about an order of magnitude lower acid site concentration per gram
than liquid acids and occupy about ten times the volume per gram ($5 cm
3 vs. 0.5 –
1.0 cm
3
/g). This results in significantly larger catalyst volumes required for equivalent reaction. Studies (Nivarthy et al. 1998; Stocker et al. 1994) have shown that
stronger Brønsted acid sites are necessary to effectively catalyze hydride transfer,
which is the reaction step that has the biggest impact on alkylate product quality and
catalyst life. The strength of the acid sites may be modified by dealumination (Corma
and Orchilles 2000) and rare earth exchange (Weitkamp 1980).
A key difference between zeolites and liquid acids is their selective and strong
chemisorption of olefins. Due to the high polarity of the zeolitic surface, especially
in those with a low Si/Al ratio, polar olefin molecules are preferentially adsorbed.
Localized higher olefin concentrations in the pores will result in higher relative
rates of oligomerization than hydride transfer and is believed to be the main reason
why zeolites deactivate so rapidly. Larger molecules are more strongly adsorbed
than smaller ones, so once a heavy molecule is formed, it is unlikely to desorb,
leading to the formation of coke, which will ultimately block the zeolite pore. The
key to running a successful solid catalyst process thus becomes finding a way to
regenerate the catalyst and remove the coke in a timely fashion.
Most SCA processes use a combination of “cold” (liquid phase) and “hot” (gas
phase) regeneration procedures. This requires the incorporation of a noble metal
hydrogenation function in the catalyst. The cold regeneration is essentially a
solvent extraction with isobutane saturated with hydrogen. At low temperatures,
(<100
C) most unsaturated species that block the acid sites are hydrogenated and
can be more easily desorbed from the catalyst. However, the cold regeneration only
partially restores activity, and after a number of cycles, it is necessary to do a hightemperature regeneration with hydrogen. At higher temperatures, there is some
hydrocracking of species that are too bulky to leave the pores into smaller species
that can diffuse out of the pores, thereby restoring pore volume and access to
acid sites.
Several different reactor schemes have been proposed for accomplishing the
solid catalyst alkylation process and regeneration. The CB&I/Albemarle/Neste Oy
AlkyClean
TM process (Gieseman et al. 2006) uses a swing bed strategy. In a typical
unit design, there are five identical reactors. At any one point in time, two reactors
Alkylation in Petroleum Processing
453