4466
Ind. Eng. Chem. Res. 1997, 36, 4466-4475
Production of Ethylene from Hydrous Ethanol on H-ZSM-5 under Mild Conditions Cory B. Phillips and Ravindra Datta* Department of Chemical and Biochemical Engineering, The University of Iowa, Iowa City, Iowa 52242-1219
Due to the dramatic rise in the use of ethers as fuel oxygenates, an alternative process for the production of ethyl tert-butyl ether (ETBE) based solely on biomass-derived ethanol feedstock is proposed. The first step in this process is ethanol dehydration on H-ZSM-5 to produce ethylene. Hydrous ethanol is a particularly attractive feedstock for this step since the production of anhydrous ethanol is very energy/cost intensive. In fact, the presence of water in the ethanol feed enhances the steady-state activity and selectivity of H-ZSM-5 catalysts. Reaction kinetic data were collected in a microreactor at temperatures between 413 and 493 K, at ethanol partial pressures of less than 0.7 atm, and at water feed molar ratios of less than 0.25. A sharp initial decline in catalyst activity observed within a few minutes on stream was attributed to the formation of “low-temperature coke” from ethylene oligomerization. Deactivation occurred at a much slower rate after 100 min on stream, allowing near-steady-state data to be collected. Water in the ethanol feed enhanced the steady-state catalytic activity and ethylene selectivity by moderating the acidity of the catalytic sites, resulting in less extensive deactivation due to coking. Introduction
CH3
The commercial production of fuel oxygenate ethers, such as methyl tert-butyl ether (MTBE) and ethyl tertbutyl ether (ETBE), has increased dramatically over the past decade (Jarvelin et al., 1996). The isobutylene required for MTBE or ETBE synthesis is available from different sources, e.g., as a byproduct from steam cracking, fluidized catalytic cracking (FCC), or field butane dehydrogenation, and in a variety of grades, with weight fractions between 12 and 40% (Ancillotti et al., 1987; Reid and McPhaul, 1996). Often, feed pretreatment is also required to eliminate potential catalyst poisons, such as propionitrile, butadiene, and acetylene. The biomass ethanol-derived ethers, e.g., ETBE, are becoming increasingly attractive due to their partial renewability. In view of the potential limitations of isobutylene availability from the current sources, a completely renewable process for producing ETBE is proposed here, based solely on ethanol. The proposed clean biomass ethanol-to-ether (CBETE) process is shown in Figure 1. The process uses both pure and hydrous ethanol feeds, similar to the biomass ethanolto-ethylene process proposed by Le Van Mao and coworkers (Le Van Mao et al., 1989). This is economically attractive since production of anhydrous fuel grade EtOH is energy and cost intensive. The proposed process is based on three well-known industrial reactions. First, the hydrous EtOH feed is subjected to catalytic dehydration to form ethylene (Tsao and Reilly, 1978)
n–C4H8
CH3
C
CH2
(3)
which is compressed and combined with anhydrous ethanol in an etherification reactor/catalytic distillation unit to produce ETBE CH3
CH3 C2H5OH + CH3
C
CH2
CH3
C
O
C2H5
(4)
CH3
Thus, by summing eqs 1-4, the overall transformation is CH3 3C2H5OH
CH3
C
O
C2H5 + 2H2O
(5)
CH3
This product stream is cooled and dimerized (AlJarallah et al., 1992) to n-butene
It may, of course, be possible to design catalysts that may combine some of the above processing steps. This paper deals with only the first reaction stage, namely, acid catalyzed EtOH dehydration to ethylene. The selective conversion of EtOH to ethylene in mildly acidic homogeneous solution was first detected by Bondt et al. over 200 years ago (Winfield, 1960). The heterogeneous catalysis of this reaction has been practiced before the turn of the 20th century on the industrial scale by passing ethanol vapors at atmospheric pressure over γ-Al2O3 or a supported acid catalyst at 588-668 K (Winter and Ming-Teck, 1976). Ethylene is typically assumed to be produced via a simultaneous parallelconsecutive scheme involving direct ethanol conversion (as shown in eq 1) as well as by the consecutive reaction
2C2H4 f n-C4H8
2C2H5OH f C2H5OC2H5 f 2C2H4 + 2H2O
C2H5OH f C2H4 + H2O
(1)
(2)
The n-butene effluent is next converted by skeletal isomerization (Szabo et al., 1991; Xu et al., 1995; Houzvicka and Ponec, 1997) to isobutylene * Author to whom correspondence should be addressed. E-mail address:
[email protected]. S0888-5885(97)00254-6 CCC: $14.00
(6)
At low temperatures (573 K), ethylene is the dominant product. Alcohol dehydrogenation to produce acetaldehyde can also occur as a side reaction at the higher temperatures, depending upon the impurities in Al2O3 (Berteau et al., © 1997 American Chemical Society
Ind. Eng. Chem. Res., Vol. 36, No. 11, 1997 4467
Figure 1. Proposed clean biomass ethanol-to-ether (CBETE) process schematic.
1987). Unfortunately, due to the difference in physical and chemical characteristics of the various γ-Al2O3 catalysts used, there is little agreement in the literature on the reaction kinetics and mechanism (Mezaki and Inoue, 1989; Szymanski et al., 1994; Hutchings et al., 1994; McCabe and Mitchell, 1984; Hasik et al., 1994; Butt et al., 1962; Miller and Kirk, 1962; Kabel and Johanson, 1962). Studies involving the use of other catalysts for this reaction report concerns of (1) catalyst deactivation due to coke formation or (2) thermal degradation (for the case of polymeric ion-exchange resins) and (3) an adverse effect of water on the reaction rates (Brey and Krieger, 1949; Toptshieva and Romanarsky, 1965). The use of a catalytic material possessing well-characterized sites that are amenable to the control of their number and acid strength, as well as a well-defined pore geometry (van Bekkum et al., 1991), e.g., zeolite, allows a more systematic study of the reaction, which is an objective of this work. Thus, H-ZSM-5 zeolite was chosen here as the catalyst for this reaction. Surprisingly, only a limited amount of research has been reported on using H-ZSM-5 for the selective production of ethylene from ethanol. However, there is a sizable amount of literature on the production of higher hydrocarbon mixtures from primary alcohols (Chang and Silvestri, 1977; Chang et al., 1978, 1979; Chang, 1983; Schulz and Bandermann, 1994). Jingfa et al. (1988) studied the surface reaction mechanism using infrared spectroscopy and concluded that EtOH may be converted into either ethylene or diethyl ether via an ethyloxonium intermediate. Much of the more recent work on this reaction has been done by Le Van Mao and co-workers (Le Van Mas et al., 1989) in their development of the biomass-ethanol-to-ethylene (BETE) process based on ZSM-5 catalysts modified with agents such as triflouromethanesulfonic acid (TFA), Zn-Mn, La/Ce cations, and asbestos (Le Van Mao and Dao, 1987; Le Van Mao and Nguyen, 1989; Le Van Mao, 1989). Nguyen and Le Van Mao (1990) examined the reac-
tion network among EtOH, diethyl ether, and ethylene on steam-treated and asbestos-derived ZSM-5 catalysts using aqueous EtOH feeds (10 wt % EtOH) and suggested the existence of a direct low-temperature route to ethylene. Although direct ethylene formation was not confirmed experimentally at low temperatures (
