Direct Synthesis of Dimethyl Ether in Microstructured Reactors: The Interactions Between Methanol Synthesis and Methanol Dehydration
Doctoral thesis
Permanent lenke
http://hdl.handle.net/11250/2399476Utgivelsesdato
2016Metadata
Vis full innførselSamlinger
Sammendrag
The growing concerns about the climate change, energy security and the diminishing
oil reserves, stimulate the search for alternative energy sources and non-petroleum-based
fuels. Natural gas in the short-term, and biomass in the long-term, have the potential
to satisfy an increasing portion of the energy demands and reduce the dependence on
carbon-intensive coal and oil. However, exploitation of highly distributed biomass resources
and a considerable amount of the natural gas reserves existing as small deposits
calls for technology breakthrough in order to become competitive (without government
incentives and/or penalties) with large scale production that benefits from economies of
scale. Microprocess engineering seems to possess the characteristics necessary for development
of highly efficient, compact, flexible, safe and relatively cheap technologies for
on-site conversion of stranded gas and bio-based feedstock into easily transportable fuels.
Among the leading candidates, i.e. hydrogen, methane, methanol, ethanol, and Fischer-
Tropsch hydrocarbons, dimethyl ether (DME) is one of the best alternative fuels in terms
of feedstock diversity, fuel versatility, energy efficiency, environmental impacts, safety,
availability of technology and infrastructure, and economics. With properties similar to
liquefied petroleum gas (LPG), DME benefits from the already existing LPG market and
infrastructure. Considering the high cetane number and clean burning properties, the
transportation sector is a big potential market for DME as a diesel substitute.
DME can be produced from coal, natural gas and biomass via synthesis gas, i.e a
mixture of H2, CO and usually CO2. Syngas conversion to DME is conventionally conducted
through a two-step process, i.e. methanol synthesis over a Cu-Zn-based catalyst
and methanol dehydration to DME over a solid acid catalyst such as γ-Al2O3 or zeolites.
Alternatively, DME can be synthesized directly from syngas in a single unit over a dual
catalyst system comprising both methanol synthesis and dehydration functions. The latter
production route is thermodynamically and economically favored. Syngas conversion to
methanol is highly limited by equilibrium and further conversion of methanol to DME
during the direct DME synthesis shifts the equilibrium towards higher syngas conversion.
Additionally, combining two reaction steps in a single unit opens up opportunities for
process intensification and cost saving, while making the DME production less affected
by price variations in the methanol market.
The aim of this project was to broaden the understanding of the direct DME synthesis.
Special focus was directed towards the possible interactions between methanol
synthesis and methanol dehydration. A packed-bed integrated microstructured reactorheat
exchanger was applied to study the reaction, in which the high surface-to-volume
ratio and micro-range flow dimensions lead to enhanced heat and mass transfer and create
isothermal conditions throughout the reactor. Cu-Zn-based catalysts and H-ZSM-5 were
used as the primary methanol synthesis and dehydration catalysts, and physical mixture
of these was applied as the direct DME synthesis catalyst. The influence of operating
conditions (space velocity, temperature, pressure, time-on-stream and syngas composition)
on the activity, selectivity and stability of the catalysts was studied. Data obtained from methanol synthesis and methanol dehydration experiments were compared with
those from direct DME synthesis under, respectively, methanol-formation- and methanoldehydration-
controlled regimes.
The results confirm the obvious advantage of the direct DME synthesis over the twostep
process in terms of alleviating the thermodynamic restrictions on syngas conversion.
However, combining methanol synthesis and dehydration produced a negative effect on
the methanol formation kinetics, which could be observed at conditions where syngas
conversion to methanol is less affected by the equilibrium. Higher steam pressure, created
by methanol conversion to DME and H2O, was identified as the most likely explanation.
DME over the methanol synthesis catalysts, syngas over the methanol dehydration catalysts,
and interactions between the methanol synthesis and dehydration functions of the
hybrid catalysts do not seem to affect the reaction kinetics.
Deactivation of both functions of the hybrid catalysts during direct DME synthesis was
also investigated under relevant industrial conditions. Copper sintering appeared as the
cause of the Cu-Zn-based catalyst deactivation, with no apparent effect from the methanol
dehydration step under the conditions applied. Accumulation of hydrocarbon species,
formed on strong acid sites and trapped in zeolite pores, was identified as the cause of HZSM-
5 deactivation. Synthesis gas composition, i.e. H2/CO ratio and CO2-content (which
directly affects steam pressure), seems to influence the zeolite deactivation. The methanol
synthesis and dehydration functions of a hybrid catalyst prepared by physical mixture of
pre-pelletized methanol synthesis and dehydration catalysts do not seem to interact.
Following previous investigations of the micro packed-bed reactors characteristics in
our group, flow distribution in the reactor was investigated experimentally using a hotwire
anemometry technique and the results considered in relation to the reactor performance
for methanol synthesis under relevant industrial operating conditions. Large catalyst
size with narrow size distribution produced the most even flow distribution. However,
the influence of the particle size on flow distribution might be small at low space velocities
corresponding to methanol synthesis at high reactor pressure. The reaction slit geometry
needs modification in order to obtain narrow residence time distribution inside each slit.