Methanol and dimethyl ether (DME) production from synthesis gas

Methanol and dimethyl ether (DME) production from synthesis gas

12 Methanol and dimethyl ether (DME) production from synthesis gas D. SEDDON, Duncan Seddon & Associates Pty. Ltd, Australia Abstract: The chapter de...

307KB Sizes 4 Downloads 338 Views

12 Methanol and dimethyl ether (DME) production from synthesis gas D. SEDDON, Duncan Seddon & Associates Pty. Ltd, Australia

Abstract: The chapter describes the current technology and process economics for the production of methanol and dimethylether (DME). Alternative routes to the conventional process are described and the status of emerging technology and alternative approaches to methanol synthesis is appraised. Process economics dictates that both methanol and DME are likely to be produced in areas of low-value gas or coal. The introduction of carbon pricing will have a profound impact on production costs and will encourage the migration of production to those regions applying minimal carbon charges. In future, large-scale methanol/DME production may be promoted by the use of DME as an LPG or diesel substitute and the use of methanol as an intermediate in the production of gasoline or olefins. Key words: methanol, dimethyl ether (DME), methanol technology, methanol catalysts, methanol economics, methanol carbon emissions, methanol production costs.



Methanol is an important commodity chemical with the world’s annual production of about 40 million tonnes. The main use of methanol is in the production of formaldehyde (for formaldehyde urea resins) with other main uses being for the production of solvents, acetic acid, biodiesel and the fuel additive MTBE. The dominant methanol synthesis catalyst is based on copper and zinc supported on alumina. This is generally referred to as the low-pressure process (<100 bar) to distinguish it from the older high-pressure process (>300 bar). The synthesis is very selective but there are some impurities, particularly water. These have to be removed by distillation prior to use as a chemical, but fuel methanol can tolerate higher levels of impurity, which in some instances can be beneficial. Most of the world’s methanol is made from natural gas, especially in those countries where gas is available at low price; the product methanol is then shipped to the major economies. In some parts of the world, in particular China, methanol is made from coal by gasification into synthesis gas. As well as its use as a chemical, methanol has been touted as an alternative clean burning fuel both in its own right and as a blend with gasoline. It can also be used as a diesel substitute if a cetane improver is added. Dimethylether (DME) is easily made from methanol. Outside China, the present day uses of DME are few, but in China DME is used as a substitute for 363 © Woodhead Publishing Limited, 2011


Advances in clean hydrocarbon fuel processing

Table 12.1 Properties of fuels of interest

Boiling point (°C)



Propane Butane

Gasoline Diesel







Flash point (°C)







Specific volume (L/t)







Higher heating value (GJ/t)







Lower heating value (GJ/t)







Research octane number






Cetane number





LPG, and DME has been proposed as a substitute for diesel. Although DME is generally made from methanol as an intermediate, direct routes from synthesis gas are under development. The properties of methanol and DME compared to typical fuels are shown in Table 12.1, which includes some properties of conventional fuels of interest. For the use of either methanol or DME as an alternative fuel, large resources of low-cost feedstock, either coal or gas, are required. This gives the necessary economies of scale and low production cost required to be competitive with petroleum fuels.


Process technology and new innovations

Methanol is made from synthesis gas, which is a mixture of carbon monoxide and hydrogen. There is usually some carbon dioxide also present, which is beneficial to the process. Synthesis gas is made from any carbon source by a wide variety of routes. Although synthesis gas can be made from any carbon source, economics forces production from the lowest-price carbon source available, which is usually natural gas, which has little other use; or coal, where coal mining costs are low. From natural gas, synthesis gas can be made by any of the usual production methods ranging from steam reforming to partial oxidation (Seddon, 2006). Autothermal reforming being popular in modern process operations because it generates a synthesis gas with the optimum ratio of hydrogen to carbon monoxide (stoichiometric ratio). From coal, synthesis gas is made by gasification. The main issue is the choice of gasifier, which is dependent upon the nature of the coal. After production the raw synthesis gas is cleaned to remove all impurities and catalyst poisons. Using coal as feedstock it is usually necessary to apply

© Woodhead Publishing Limited, 2011

Methanol and dimethyl ether production from synthesis gas


the water-gas shift reaction to obtain a suitable ratio of hydrogen to carbon monoxide in the synthesis gas; excess carbon dioxide produced is removed by an appropriate operation. After production, the synthesis gas is first compressed and passed to the converter, with the conversion usually taking place at about 250–300 °C. The pass conversion is low, typically 10%. The product gases pass through a condensing system (which raises steam) and condensed products are removed in a knockout pot (Fig. 12.1). The raw methanol from the knockout pot, also called crude methanol, contains large quantities of water and some higher alcohols. It is now passed to a fractionation train where the required methanol grades are produced. There are several designs for the converter. These are usually developed by the individual process licensors. There are many variations including quench converters and various types of tubular reactor. A discussion of the various types is beyond the scope of this paper; Lee (1990) discusses the general layout of proprietary flow-sheets and Abbott (1992) discusses the various designs of methanol converter. Higher temperatures favour higher alcohols (particularly butanols). Older highpressure technology produced methanol containing several percent butanols. An old (now extinct) technology produced high concentrations of isobutanol (the Isosynthesis Process). Methanol containing higher alcohols has some advantages as a fuel over methanol alone; for example it has higher energy density. If methanol were to become an alternative fuel to automotive gasoline then these processes may offer advantages over the current processes.

12.1 Methanol synthesis.

© Woodhead Publishing Limited, 2011


Advances in clean hydrocarbon fuel processing

12.2.1 Advances in the conventional process Most research and development in the area of methanol synthesis concentrates on improving the process. Workers at ICI (now Johnson Matthey) have proposed improved heat exchange and better heat integration with steam reforming (Abbott, 2001; Fitzpatrick, 2002a, 2002b, 2006). Several other groups have also proposed better heat integration of the methanol synthesis and the reformer (Kobayashi and Nagai, 1999; Hirotani et al., 2000; Early et al., 2001; Patel et al, 2004) and/or the distillation train (Kobayashi et al., 2001; Seiki et al., 2005a). Several groups have proposed multiple reactors in series (Konig and Gohna, 1998; Bahnisch, 2003, 2004; Eastland, 2007); in one version the first reactor acts as an absorber of catalyst poisons (Bahnisch, 2005). The synthesis reactor can be a catalytic distillation reactor (Allison et al., 2004). Methanol or DME can be converted into olefins using various processes but including technology developed by ExxonMobil (Seddon, 2006). This has resulted in a series of innovations not only in the conversion of methanol into olefins, the integration of synthesis gas production with methanol synthesis and olefin synthesis (Janda, 2002; Janda and Kuechler, 2002, Van Egmond, 2006a, 2006b, Van Egmond et al., 2007, 2008; Janssen et al., 2007; Borgmann et al., 2007), but also by Lattner (2006, 2007, 2008a, 2008b, 2009a, 2009b) in the synthesis of methanol using different technological approaches such as pressure swing reforming and the use of fluid bed reactors. Finally, the current concern with the production of greenhouse gas emissions has sparked an interest in utilising carbon dioxide in the synthesis and using renewable sources of feedstock rather than fossil fuel-based feedstock (Ihm et al., 2003; Kaneko et al., 2003; Shiroto et al., 2003; Shaw, 2004). Partial oxidation versus steam reforming One of the ways the process could be improved is to use partial oxidation of natural gas to give directly a synthesis gas of the correct stoichiometric ratio (the molar (H2)/(CO + CO2)ratio). This avoids the high hydrogen levels produced by steam reformers. One advantage of the partial oxidation route is that the oxidation can be conducted at the methanol synthesis pressure thereby saving on compression costs (Guillard and Schmidt, 2005, 2006). Another variation is to use air or enriched air as the oxidant rather than pure oxygen. The methanol synthesis gas then contains nitrogen and to avoid build up in a recycle stream, a cascade of methanol converters is proposed (Fraley, 2006).

12.2.2 Dimethyl ether (DME) DME is receiving increased attention as an alternative fuel. The salient properties are presented above in Table 12.1. The key points are:

© Woodhead Publishing Limited, 2011

Methanol and dimethyl ether production from synthesis gas

• • •


DME is a gas at ambient temperature and pressure and is therefore stored and transported at pressure, much the same as LPG. It can be blended with LPG and therefore is an LPG substitute. DME is proposed as a substitute for diesel, for which use it has beneficial environmental properties. Unlike methanol it has a sufficient cetane number without the use of additives or spark assistance. DME has low octane and is not used as a gasoline substitute in contrast to methanol, which has excellent octane for this use.

The calorific value of DME is higher than methanol but lower than conventional fuels. Although the energy density is higher, relative to methanol, transport logistics are higher cost due to the requirement to store and transport under pressure. Like methanol, DME can be used directly in gas turbines for power generation and like methanol it has lower NOx emissions than conventional liquid fuels or natural gas.

12.2.3 DME production DME is produced from methanol by the reaction: 2CH3OH = CH3.O.CH3 + H2O

[Reaction 12.1]

The process is very simple and uses solid acid catalysts. The layout in block form is shown in Fig. 12.2. Methanol, not necessarily of the highest quality, is vaporized and passed to the converter operating typically at 250 °C, which establishes the equilibrium.

12.2 DME synthesis from methanol.

© Woodhead Publishing Limited, 2011


Advances in clean hydrocarbon fuel processing

Product is cooled (not shown) and water removed before methanol is separated from the DME before recycle. This process is well known, and used to produce relatively small quantities of DME for various uses such as a propellant in retail spays or incorporated into processes for converting methanol into gasoline or olefins as a pre-converter, to help in the control of the high level of heat release from these processes. It is also used to produce larger quantities of DME for blending with LPG (Toyo Engineering Corporation, 2005). Because the process is relatively simple and uses mildly acidic catalysts (alumina), the process could be improved using reactive distillation techniques. For the mass production of DME a single-stage production process has been proposed using synthesis gas as the starting feedstock and thereby avoiding the production of intermediate methanol. However, the route is as yet commercially unproven. There are questions concerning the technical viability of the single-stage route. The principal problems to date have been:

• •

Poor catalyst stability with catalyst activity and selectivity falling away with time-on-stream. By comparison, methanol synthesis catalysts last many years. Requirement to reduce carbon dioxide in the synthesis gas for optimum performance. By contrast, methanol requires some carbon dioxide.

Several companies, such as NKK (Shikada et al., 2000), Air Products (Peng et al., 1998, 2000, 2002) and Haldor Topsøe (Haugaard and Voss, 2001) claim to have overcome these problems and offer a single-stage technology for license. These technologies are summarized in Table 12.2 (adapted from Kato, 2002), and are compared to the Toyo Engineering Corporation (TEC) two-step process.

Table 12.2 Comparison of technology for the production of dimethylether Licensor


Air Products Inc. Haldor Topsøe








Liquid phase/ slurry

Liquid phase slurry

Vapour phase – fixed bed

Vapour phase – fixed bed


Methanol synthesis + dehydration

Methanol synthesis + dehydration + shift

Methanol synthesis + dehydration + shift

Methanol synthesis then separate dehydration

Temperature (°C)





Pressure (bar)






>95% DME

30–80% DME


DME only

© Woodhead Publishing Limited, 2011

Methanol and dimethyl ether production from synthesis gas


Note that some processes require synthesis gas of low stoichiometric ratio and produce a mixture of methanol and DME. These processes require a methanol dehydration unit if only DME is to be produced. The method has been the subject of research over many decades because methanol can easily be obtained from DME by reversing the above reaction: CH3.O.CH3 + H2O = 2CH3OH

[Reaction 12.2]

This offers the methanol producer potentially significant lower methanol production costs as a result of higher theoretical conversion (recall that the pass conversion in methanol synthesis is only about 10%).

12.2.4 Liquid-phase methanol process The conventional process for the production of methanol is a gas-phase process which is very good for the high hydrogen content synthesis gas produced from natural gas. However, conducting the reaction in the liquid phase could produce significant benefits (Korosky, 1996):

• • • •

the ability to handle high carbon monoxide content synthesis gas like that produced by coal gasification potential better integration with IGCC power plants providing back-up fuel (methanol) for power generation improved temperature control by a slurry operation which more efficiently removes reaction heat higher conversion of synthesis gas to methanol e.g. 15–18% compared to a typical 10% in the conventional process.

The process involves suspending the methanol synthesis catalyst in hydrocarbon oil. The development was based on 1970s Chem Systems Inc. research into the conversion of coal to methane using liquid-phase reactors. The methanol variant on this was sponsored by Electric Power Research Institute (EPRI). After 1979 Air Products and Chemicals joined the group and in 1984 built pilot operations at LaPorte in Texas producing 5 t/d of methanol (Air Products, 1992). A decade later it was planned to build a demonstration plant at the Texaco Cool Water facility (Brown and Freduto, 1992). The technology is offered for license by Air Products Inc. (2007). There are several variants on the process. Workers at Brookhaven (Mahajan, 2005) use nickel complexes in solution with an alkoxide. The solvent for the process can be methanol (Nielsen et al., 2005).

12.2.5 Integration of ammonia and methanol For some time the major process licensors of methanol technology have realized that there could be energy efficiency gains if methanol production were integrated

© Woodhead Publishing Limited, 2011


Advances in clean hydrocarbon fuel processing

with ammonia production (Badano et al., 2000; Filippi, 2001; Filippi et al., 2005; Davey and Wurzel, 2009). In essence, the excess hydrogen produced by steam reforming is used to produce ammonia and the heat generated from the units and the purge gases integrated into a common steam system. Integration has been accomplished in some plants (Rhodes, 1994).

12.2.6 Other routes to methanol Methane partial oxidation The partial oxidation of hydrocarbons (LPG or naphtha) leads to a mass of oxidized products, including methanol, which can be separated by distillation. The process has been operated by Celanese to produce methanol, as well as acetic acid and other chemicals, from LPG (Weissermel and Arpe, 1994). Ever since methanol was first synthesized from synthesis gas, there has been research on the direct partial oxidation of methane (the principal constituent of natural gas) into methanol (Foster, 1985). The route has received spurts of interest over the decades. Some of these studies have mainly focused on the homogeneous partial oxidation, of which the work by Gesser et al. (1985; Hunter et al., 1990) has been an important contribution. Most of the work has been catalytic; however, despite many studies, development of the process does not seem to have progressed beyond that achieved by Dowden and Walker (1971), who developed the process to a pilot scale. Table 12.3 illustrates the results and the general problems with the route. Dowden’s work concentrated on doped molybdenum oxide catalysts; uranium oxide was particularly effective. In order to avoid the potential of the formation of an explosive mixture, the methane has to be in vastly excessive quantities to that of the oxygen. Consequently, conversion is low. The selectivity to methanol and formaldehyde is below 80%; the other products are the deep oxidation products carbon dioxide and water. The low level of product in the excess

Table 12.3 Typical results of the direct partial oxidation of methane to methanol Catalyst

MoO3.ZnO (MoO3)2.Fe2O3 MoO3.UO2









Pressure (bar)







37 000

46 000

23 200

47 600

25 500

Temperature (°C)






Conversion (%)






MeOH yield (%)






HCHO yield (%)






Space velocity

© Woodhead Publishing Limited, 2011

Methanol and dimethyl ether production from synthesis gas


gases inevitably leads to high extraction costs. Furthermore the high reaction temperature would be intuitively odd considering the large thermodynamic driver for a partial oxidation under catalytic conditions. This high temperature suggests a high activation barrier similar to those found in free radical processes; the catalyst is either catalyzing the free radical process or is suppressing the formation of deep oxidation products. Research in the field continues (for example Sanfilippo et al., 1999). Methanol via methyl formate This is a two-stage route whereby methanol is reacted with carbon monoxide to form methyl formate. This is subsequently reduced to form two molecules of methanol. CH3OH + CO = HCOOCH3

[Reaction 12.3]

HCOOCH3 + 2H2 = 2CH3OH

[Reaction 12.4]

The route was first described by Christiansen (1919) and was later developed by Sorum and Onsager (1984) and later by Wainwright (Monti et al., 1985, 1986) using copper chromite catalysts. Recent developments aim to perform both reactions together using a catalyst of copper chromite, potassium methoxide in a non-polar solvent (Wu et al., 2000).


Basic principles of methanol synthesis

12.3.1 Thermodynamics There are two principal reactions for the conversion of synthesis gas into methanol to be considered; the first is the conversion of carbon monoxide: CO + 2H2 = CH3OH

[Reaction 12.5]

And the second is the conversion of carbon dioxide: CO2 + 3H2 = CH3OH + H2O

[Reaction 12.6]

Both processes are highly exothermic at ambient temperatures with enthalpies of reaction −90.64 kJ/mol and −49.47 kJ/mol respectively. The free energy change for these reactions is shown in Fig. 12.3. Figure 12.3 illustrates that the reaction is favoured by low temperatures. Negative values for free energy are observed for temperatures below about 400 K. Both the reaction from carbon monoxide and from carbon dioxide have similar changes in free energy. However, the conversion of CO to DME shows a higher temperature response and for temperatures below about 500 K, DME becomes more favoured a product than methanol.

© Woodhead Publishing Limited, 2011


Advances in clean hydrocarbon fuel processing

12.3 Thermodynamics of methanol production.

Since there is a volume contraction in all of the reactions, methanol synthesis is favoured by increasing the pressure. However, many potential by-products of methanol synthesis, such as methane, proceed with higher heat release and larger negative values for free energy. The production of methanol is therefore controlled by the kinetics of the process rather than thermodynamics.

12.3.2 Kinetics It should be noted that there is a substantial amount of discussion in the literature concerning the kinetics and the primary step of the reaction over the low pressure catalyst. Many of the differences of opinion lie in the different sources of catalyst being used. These points of view fall into three groups: (1) those leading to the view that the main reaction is via carbon monoxide, (2) those of the view that the primary reaction is via carbon dioxide, and (3) those of the view that both processes are important. Klier (1984) in a series of papers proposed the view that the prevalent reaction was via carbon monoxide. The rate of methanol synthesis from carbon dioxide was much lower. Also observed was that the maximum rate of synthesis was obtained with carbon dioxide at 2%. This was considered a promotional effect with carbon dioxide helping maintain the copper in an active monovalent state (Cu+). The surface of commercial catalysts have been shown to be contain both metallic and oxidized species, with the proportion of oxidized species depending on the CO2/CO ratio of the gas phase (Chinchen and Waugh, 1986). The strength of carbon dioxide absorption is higher than carbon monoxide so that at high carbon dioxide concentration, the rate of reaction falls off.

© Woodhead Publishing Limited, 2011

Methanol and dimethyl ether production from synthesis gas


In contrast work using 18O labelling by Kung et al. (1984) showed that synthesis from carbon monoxide and dioxide and the water-gas shift were all possible. They also observed that the rate of methanol formation increased as the proportion of carbon dioxide increased. The conventional wisdom is that over the commercial low pressure catalyst, the reaction proceeds via carbon dioxide rather than carbon monoxide. This is clearly illustrated in 14C labelling experiments (Chinchen et al., 1984, 1987, Kagan Yu et al., 1973)), which show unambiguously that the primary process is via carbon dioxide. If carbon dioxide is not present then this is produced in situ by the water-gas shift reaction; when carbon dioxide free gas is used in the synthesis, the reaction is slower than if a few percent of carbon dioxide is present. The role of water in the reaction is difficult to interpret because of the very rapid water-gas shift reaction over the low pressure catalyst. Generally, water is seen to inhibit the reaction rate. This may be due to catalyst oxidation and hydrothermal processes with water affecting the alumina support of the catalyst. The rate expressions for the production of methanol are generally of the type: Rate = kpxCO pyH2(1−pMeOH/KpCO p2H2)


where p is the partial pressure of the reactant or methanol, K is the equilibrium constant, x is a constant and ranges from 0.1–1.0, y is a constant and ranges from 1.0–2.0. Note that carbon dioxide is not part of the equation despite it being the primary reactant; the rate determining step involves carbon monoxide. The constants for carbon monoxide (x) are smaller than that for hydrogen (y) indicating the larger impact of hydrogen partial pressure. The equilibrium constant in the negative term expresses the fact that as methanol concentration increases the reaction slows down. Other more complex forms taking into account the influence of carbon dioxide and the inhibiting effects of water are discussed by Chinchen et al. (1988). Using equations of the form above, it has been found that the rate constants for the production of methanol from carbon monoxide were: CO + 2H2 = CH3OH

[Reaction 12.7]

Rate = k1 p0.5COp1.5H2(1–pMeOH/K1pCO p2H2) k1 = 0.2032 exp(−2954/RT)


[12.2] [12.3]

and for carbon dioxide: CO2 + 3H2 = CH3OH + H2O

[Reaction 12.8]

Rate = k2p0.5CO2 p1.5H2(1–pMeOH/K2pCO2 p2H2) k1 = 8.893 exp(−6163/RT)


© Woodhead Publishing Limited, 2011

[12.4] [12.5]


Advances in clean hydrocarbon fuel processing

As well as the extensive work on these catalysts in the gas phase, an extensive series of kinetic measurements in the liquid phase has been made (Lee, 1990). This was in support of the development of a liquid phase process.



The original catalysts were based on zinc oxide. The catalyst was active at temperatures over 350 °C which necessitates the use of very high pressures (350 bar) to obtain adequate conversion. Under these conditions zinc oxide is not stable and it was common to include chromia in the formulation which stabilized the zinc oxide and prevented sintering to large (less active) zinc oxide crystals. These catalysts have been largely replaced by formulations based on copper/ zinc oxide/alumina (Bridger and Spencer, 1989; Lee 1990, Cheng and Kung, 1994). These catalysts are active at much lower temperatures and hence can be operated at lower pressure for the same level of conversion as the older formulations; they are generally known as low-pressure catalysts. Stiles and Koch (1995) describe a recipe for both catalysts. Because they are able to operate at lower temperature and pressure, the low-pressure catalysts produce far less by-product than the high-pressure process. Table 12.4 compares the by-products formed from the two catalyst types. The formation of other by-products is influenced by the presence of impurities in the catalyst. Alkalis promote the formation of higher alcohols and acids (such as silica) promote the formation of waxes which block the catalyst. Many impurities are transported onto the catalyst during operation. For instance, nickel and iron can be transported in the gas phase as the metal carbonyls which are formed in up-stream operations when synthesis gas interacts with finely divided metal. These impurities promote the formation of methane, paraffins and waxes in the methanol synthesis loop. Other impurities such as sulphur and chlorine can contaminate the synthesis gas and can permanently decrease the activity of the catalyst.

Table 12.4 By-products of methanol synthesis By-product

High pressure (ppm (w/v)) Low pressure (ppm (w/v))

Dimethyl ether

5000–10 000




Higher alcohols




2% of input carbon


Carbonyl compounds

© Woodhead Publishing Limited, 2011

Methanol and dimethyl ether production from synthesis gas


12.4.1 Other formulations As well as the standard high and low pressure catalysts, there are several other catalyst approaches based on different metals. These have been reviewed by Chinchen et al. (1988). Variations on the standard Cu/ZnO/Al2O3 catalyst There are several claims for improving the activity of the standard copper/zinc/ alumina catalyst by using additives including boron, silver, manganese, rare earths, chromium, molybdenum, tungsten and vanadium. Manganese gave a catalyst that was active at 180 °C; Cr, Mo, W gave catalysts of very high initial rates. High rates have also been claimed for formulations containing zirconia (Sofianos et al., 2000) and a formulation of the standard type with an alumina and zinc oxide coating of copper oxide particles (Fukai et al., 2000). One variation is a catalyst based on Raney copper. These catalysts are produced by caustic leaching of copper/zinc/aluminium alloys (Friedrich et al., 1983a, 1983b; Marsden et al., 1980). Recent developments of the route claim very high activity and hence lowering the pressure or increasing the pass conversion (Kokubu, 2000). Inter-metallic compounds Thorium-copper inter-metallic compounds are active for the conversion of carbon dioxide-free synthesis gas into methanol. Other formulations not involving thorium have been developed (Owen et al., 1987). With an appropriate activation procedure not only are the catalysts active, but the selectivity is similar to the standard catalyst (Short and Jennings, 1984; Bryan et al., 1988). Some of these catalysts are deactivated by small amounts of carbon dioxide in the synthesis gas. Precious metal catalysts Palladium on silica has been shown to be an active catalyst for the synthesis (Poutsma et al., 1978a, 1978b). Additives, particularly calcium, have been found to significantly improve performance. Although these catalysts are less active than the standard catalysts on a weight basis they are much more active from the perspective of the active metal. Research in the field continues; for example Matsumura and Shen (2002) describe Pd/Ce3O3 compositions and Fujimoto et al. (2007) describe Pd methanol catalysts in combination with zeolites to produce LPG from synthesis gas. Other precious metal catalysts have been studied but the rates are much lower than the palladium catalysts.

© Woodhead Publishing Limited, 2011


Advances in clean hydrocarbon fuel processing


Product quality

Traded methanol is generally analyzed by the methods described in ASTM D1152 of the International Methanol Producers and Consumers Association (IMPCA). There are several grades of methanol; the key parameters for these grades are given in Table 12.5. Fuel methanol can be produced in a single column. This product has water content reduced to <1%. Since the presence of higher alcohols is not a problem with fuel grade, the product contains some ethanol. Higher alcohols are favoured by high temperature synthesis, and since higher alcohols have higher energy content, this could be advantageous for some end uses. There have been proposals to produce fuel methanol with several percent higher alcohol (particularly butanol) content. Fuel methanol is not traded but has been produced for various alternative fuel demonstration projects and the use of this grade is common when methanol is being used as an intermediate for the immediate production of another product such as gasoline. Federal Grade A methanol can be produced in a two-column system. This grade is used when certain specifications in the AA grade are not critical, e.g. for the production of formaldehyde. Federal Grade AA is the grade generally traded. It has a high specification, particularly on the ethanol and acetone content. This grade is produced by a three- or four-column distillation system.

Table 12.5 General specifications for methanol

Methanol (min wt%)


Grade A

Grade AA






Specific gravity (20°/20°)




Distillation range† (°C)






Water (max wt%)



Acetone + aldehydes (ppm)



Acetone (ppm)






Ethanol (ppm) Residue (mg/L)



Chloride (mg/kg)


Sulphur (mg/kg)


Acidity (acetic acid; mg/kg)




Permanganate (@ 15 min)





contain 64.6 °C +/− 0.1 °C at 760 mm

© Woodhead Publishing Limited, 2011

Methanol and dimethyl ether production from synthesis gas


The IMPCA specification is primarily aimed at assuring quality control from seaborne cargoes and concentrates on avoiding the risk of seawater contamination. Although the modern low-pressure technology is very selective, contamination of the product can arise from spurious reactions due to contaminated materials in the system, for example:

• •

rust would be reduced to iron which could catalyze the Fischer–Tropsch process to hydrocarbons and wax nitrogen in the system at start-up or shutdown or as a contaminant to the synthesis gas leads to the formation of amines.

The removal of lighter and heavier components is generally performed with a three-column distillation system. Figure 12.1 is a highly simplified representation of the fractionation train, which in practice is complex. The approach to the arrangement and duty of individual columns tends to vary according to the process licensor. Mehta and Pan (1971) have given an overview of the various distillation schemes and describe the approach to the fractionation train design. Recent developments aim to better integrate the heat exchange of the exothermic synthesis with the heat demand of the distillation train (Kobayashi et al., 2007; Seiki et al., 2005b).


Estimation of production costs

12.6.1 Methanol We are principally concerned with the production of methanol from large-scale gas-based plants. For the analysis we are consider the production cost from standalone, green-field facilities, that is – the cost encompasses all of the utilities and off-sites required. Such plants comprise three sections: (1) the production of synthesis gas from natural gas, (2) the methanol synthesis loop and (3) a fractionation train to produce quality product. By far the largest cost item is the production of synthesis gas, with the fractionation train costing the least. Older plants generally used steam reformers to produce the synthesis gas. More modern plant generally uses autothermal reforming, or similar technology, and requires oxygen from an air separation plant (ASU). Although these latter process plants are smaller than steam reformers, the capital cost is very similar due to the cost of the ASU. However, the use of an ASU allows an apparent lower capital cost base because the ASU is often built and owned by a third party such as a merchant industrial gas supplier. With oxygen purchased from the third party, capital is saved at the expense of an increased operating cost. For the economic analysis presented here it is assumed that if an ASU is used it remains within the battery limits and there is insignificant capital cost difference between a steam reformer and the cost of an ASU and auto-thermal reformer.

© Woodhead Publishing Limited, 2011


Advances in clean hydrocarbon fuel processing

There are many variations on the methanol converter available from different process licensors. To a first-order estimate these are assumed to have the same operating and capital cost. The fractionation train is relatively low (typically $25–30 million). This cost can be saved if fuel methanol is to be produced. The methodology used in this analysis has been described in detail previously by Seddon (2006) quoting Stratton (1982). The methodology assumes: • • • • • • • •

Standalone greenfield operation. Treated gas is delivered to the site, water and black start power are available. Base production 850 000 t/y (2500 t/d). Grade AA product. Plant operates at design output immediately at end of construction period. Operational life fifteen years; zero value at end of life. Construction period three years. Return on capital 16.34%; this accounts for capitalizing interest charges during the construction period and with a discounted cash flow rate of return of 10% (see reference).

The operating costs are the cost of gas and the other non-gas operating costs: • • • • •

Labour (3.5% annual charge of capital cost). Maintenance (3.5% annual charge of capital cost). Catalysts and chemicals (2% annual charge of capital cost). Insurance and other costs (1% annual charge of capital cost). Working capital is treated as a debt obligation based on 30 days of methanol production valued at $300/t with a 10% interest rate.

The analysis leads to a fixed variable equation: P=F+C+O


Where: P is the estimated production cost, F is the feedstock (gas) cost, C is the return on capital, O is all of the non-gas operating costs. A plant as described will have a capital cost (US Gulf basis) in 2007 of US $430 million and a gas usage of 32.16 PJ/y; this is the total site gas usage and corresponds to a site thermal efficiency of 60%. The fixed-variable equation is shown in Fig. 12.4. Comparing this figure with the historical prices for traded methanol (Fig. 12.5), we can see the following: •

For many periods in the historical price curve the production cost is below $200/t and sometimes in the region of $100/t. This requires the gas feedstock price at $2/GJ or below. For the higher gas prices usually found in OECD economies (>$5/GJ), production costs are in excess of $300/t. This price is unusual in the historical cost curve.

© Woodhead Publishing Limited, 2011

Methanol and dimethyl ether production from synthesis gas


12.4 Methanol production costs.

12.5 Methanol historical prices.

These observations confirm the trend to produce methanol in areas where large undeveloped reserves of gas are available at low price.

12.6.2 Impact of carbon emissions cost At the time of writing there is a great deal of concern about the emission of carbon dioxide to the atmosphere. Several legislative instruments are being debated for preventing this, including emissions taxes, cap-and-trade schemes and carbon geo-sequestration. All these methods will result in a cost to be applied to the carbon dioxide produced in the methanol production process.

© Woodhead Publishing Limited, 2011


Advances in clean hydrocarbon fuel processing

Starting from natural gas, carbon dioxide emissions will result from the production of synthesis gas. For steam reformers, carbon dioxide emissions result from firing the steam reformer to provide the temperature required and to provide the reaction heat. For synthesis gas produced using partial oxidation or autothermal reforming, carbon dioxide emissions result from the power generation required to produce the oxygen required. As well as these emissions, there may be other emissions resulting from the provision of the natural gas. To properly quantify the level of emissions, a full cradle-to-grave analysis is required for each individual plant. To give some perspective of the impact of carbon dioxide emission costs on the cost of methanol production the following case is developed. • • • • • •

plant as described above (2500t/d methanol; greenfield site) steam reforming of natural gas natural gas cost $2/GJ emission from natural gas production ignored overall site facility operates with a thermal efficiency of 60% only carbon dioxide emissions are considered and contribution to greenhouse gas emissions from methane fugitive emissions and the production of nitrogen oxides is ignored.

With a site thermal efficiency of 60%, 40% of the natural gas is consumed in furnace operations, which result in 0.643 million tonnes of carbon dioxide, which will be subject to emission charges. The impact of these emissions charges is illustrated in Fig. 12.6. Figure 12.6 illustrates that as carbon emissions charges rise to $20/t, the production cost rises to about $215/t from below $200/t. With emissions cost at

12.6 Impact of carbon emissions costs on methanol production cost.

© Woodhead Publishing Limited, 2011

Methanol and dimethyl ether production from synthesis gas


$40/t the production cost is about $230/t relative to below $200/t without an emission charge. This builds in clear advantages to methanol producers in countries without carbon emissions costs.


Future trends

Methanol is a mature commodity chemical with its growth pegged to the growth in gross domestic product. Thus world growth is typically in the region of 3% per annum with world GDP growth at this level. For annual production of about 40 million tonnes the annual increase is in the region of one million tonnes per year which approximates to one world-scale plant. Because methanol is a commodity chemical the production migrates to the cheapest source of base feedstock. Over the past decade this has traditionally been natural gas. However, since 2005 there have been major upheavals in the world gas market with many low-cost sources of gas being diverted to the international pipeline gas (Canada to the USA, Russia to EU) or being diverted to the production of liquefied natural gas (LNG) which has benefited from persistent increase in demand and historically high prices. These trends have resulted in the shutdown of several major methanol production facilities. This loss on the supply side has been offset by the reduction in demand for the production of MTBE in the USA – equivalent to four world-scale plants – and new plants coming on-stream from low-cost producers (Trinidad and the Middle East). In recent years there has been a trend to very large – so-called Mega-Methanol – plants which have outputs of two million tonnes or more. Several are under final stages of construction. Further methanol capacity is coming on-line to produce DME. This is of particular interest to gas or coal rich countries with an established LPG market. The production of DME offsets the import of LPG or allows LPG export at world parity prices which are generally set relative to the prevailing price of crude oil. This demand for DME should not unduly influence world trade as it is unlikely countries would import methanol or DME to offset domestic LPG use. For large-scale production of DME an international trade in the product will be dependent on other uses – either power station fuel or as a diesel substitute. The course of the methanol market in China is of interest. Large volumes are produced to service the rapidly growing domestic market, and there is also heavy use of methanol as a gasoline substitute and for the production of DME to offset LPG imports. This is likely to persist as it gives China the option to use its vast coal reserves to substitute for imported petroleum products. This lead may be followed by other coal (or gas) rich countries. One point is that methanol use in gasoline is vigorously opposed by vehicle manufacturers producing more sophisticated engines.

© Woodhead Publishing Limited, 2011


Advances in clean hydrocarbon fuel processing

The future may see a large increase in methanol/DME production as an intermediate in the production of gasoline and olefins using molecular sieve catalysts. Two new methanol-to-gasoline plants, using coal as the feedstock, are in commissioning or are under construction in China and the USA. These methanol-based routes would require the production of large volumes of intermediate methanol and there would be opportunities for the business owners to sell some of the methanol/DME intermediate.


Sources of further information and advice

The use of methanol is generally opposed by the majority of vehicle engine manufacturers; methanol corrodes lighter construction materials (aluminium, zinc, titanium) used in advanced engines. The engine manufacturers express their view on fuel formulation in the World-Wide Fuel Charter. This is frequently revised and available from the websites of the major engine manufacturer associations. The WWFC gives reference to background SAE Technical Papers and similar to support their arguments. A comprehensive review of the production of synthesis gas by steam reforming and methanol synthesis is given in the Catalyst Handbook (Twigg, 1989). Methanol Production and Use (Cheng and Kung, 1994 ) contains an excellent description by L V LeBlanc, R V Schneider and R B Strait of the production of methanol in a large gas-based Cape Horn methanol plant in Chile. This paper also gives detail of the coal gasification and gas partial oxidation and auto-thermal reforming routes to synthesis gas. Lee (1990) in Methanol Synthesis Technology gives a description of the chemistry, kinetics and thermodynamics of the methanol synthesis reaction. This work also describes the approach to synthesis of the major process licensors. Progress in the field of partial oxidation of methane to methanol is described by Pitchai and Klier (1986) and Brown and Parkyns (1991); this has been updated by Klier et al. (1993, 1995, 1996) in a series of DOE reports. Methanol prices, analysis and market analysis are regularly provided by chemical consultancy groups and reporting agencies. These are often reviewed in trade journals such as Hydrocarbon Processing and the Oil and Gas Journal.



Allison J D, Wright H A, Harkins T H and Jack D S (2004), US Patent 6,723,886 (20 April 2004) to Conoco Phillips Company. Abbott P E J (1992), ‘Methanol Reactor Design Choices’, World Methanol Conference, Monte Carlo, Monaco. Abbott P E J (2001), US Patent 6,255,357 (3 July 2001), to ICI Ltd. Air Products Inc. (1992), Oxygen for Gasification . . ., brochure 280–9101. Air Products Inc. (2007), LPMEOH brochure.

© Woodhead Publishing Limited, 2011

Methanol and dimethyl ether production from synthesis gas


Badano M, Fabbri F and Filippi E (2000), US Patent 6,106, 793 (22 August 2000) to Methanol Casale S.A. Bahnisch H-J (2003), US Patent 6,608,113 (19 August 2003) to Krupp Uhde GmbH. Bahnisch H-J (2004), US Patent 6,686,398 (3 February 2004) to Uhde GmbH. Bahnisch H-J (2005), US Patent 6,894,081 (17 May 2005) to Uhde GmbH. Borgmann W, Kunkel J, Fritz H, Lauermann G, Walz R et al. (2007), US Patent 7,238,848 (3 July 2007) to ExxonMobil Chemical Patents Inc. Bridger G and M.S. Spencer (1989) Methanol synthesis’ in M.V. Twigg (ed.), Catalyst Handbook, London, Wolfe Publishing Ltd. Brown W R and Freduto F S (1992), Fuel and power coproduction the integrated gasification/liquid phase methanol (LPMEOHTM) demonstration project, First Annual Clean Coal Technology Conference, Cleveland, Ohio, September 22–24, 1992. Brown M J and Parkyns N D (1991), Catalysis Today, 8, 305. Bryan S J, Jennings J R, Kipling S J, Owen G, Lambert R M et al. (1988), Applied Catal., 40, 1743. Cheng W-H and Kung H H (1994), Methanol Production and Use, New York, Marcel Dekker, 1994. Chinchen G C, Denny P J, Parker D G, Short G D, Spencer M S et al. (1984), Preprints, Amer. Chem. Soc. Div. Fuel Chem., 29(5), 178. Chinchen G C, Denny P J, Parker G D, Spencer M S and Whan D A (1987), Applied Catal., 30, 333. Chinchen G. C, Denny P J, Jennings J R, Spencer M S and Waugh K C (1988), ‘Synthesis of methanol’, Applied Catal., 36, 1. Chinchen G C and Waugh K C (1986) J. Catal., 97, 280. Christiansen J A (1919), US Patent 1,302,011 (29 April 1919). Davey W and Wurzel T (2009), US Patent 7,521,483 (21 April 2009) to Lurgi AG. Dowden D A and Walker G T (1971), British Patent 1,244,001 to ICI Ltd. Early S R, Gamlin T A and Linthwaite, M A (2001), US Patent 6,191,174 (20 February 2001) to Kvaerner Process Technology Ltd. Eastland P H D (2007), US Patent 7,214,721 (8 May 2007) to Davy Process Technology Ltd. Filippi E (2001), US Patent 6,333, 014 (25 December 2001) to Methanol Casale S.A. Filippi E, Rizzi E and Tarozzo M (2005), US Patent 6,946,494 (20 September 2005) to Methanol Casale S.A. Fitzpatrick T J (2002a), US Patent 6,387,963 (14 May 2002) to ICI Fitzpatrick T J (2002b), US Patent 6,433,029 (13 August 2002) to ICI. Fitzpatrick T J (2006), US Patent 7,144,923 (5 December 2006) to ICI. Foster N R (1985), Applied Catal., 19, 1. Fraley L (2006), US Patent 7,019,039 (28 March 2006) to Starchem Technologies Inc. Friedrich J B, Wainwright M S and Young M S (1983a), J. Catal., 80, 1. Friedrich J B, Wainwright M S and Young M S (1983b), J. Catal., 81, 14. Fujimoto K, Asami K, Asaoka S, Li X and Zhang Q (2007), US Patent 7,297,825 (20 November 2007) to Japan Gas Synthesize Ltd. Fukui H, Kobayashi M, Yamaguchi T, Arakawa H, Okabe K et al. (2000), US Patent 6,114,279 (5 September 2000) to YKK Corp. Gesser H D, Hunter N R and Prakash C B (1985), ‘The direct conversion of methane to methanol by controlled oxidation’, Chem Rev., 85, 235 (1985) to Air Liquide. Guillard A and Schmidt E (2005), US Patent 6,881.758 (19 April 2005) to Air Liquide. Guillard A and Schmidt E (2006), US Patent 7,091,251 (15 August 2006) to Air Liquide. Haugaard J and Voss B (2006), US Patent 6,191,175 (20 February 2001) to Haldor Topsøe AS.

© Woodhead Publishing Limited, 2011


Advances in clean hydrocarbon fuel processing

Hirotani K, Nakamura H and Shouji K (2000), US Patent 6,100,303 (8 August 2000) to Toyo Engineering Corp. Hunter N R, Gesser H D, Morton L A, Yarlagadda P S and Fung D P C (1990), Applied Catal., 57, 45. Ihm S-K, Park Y-K, Jeon J-K, Jeong K-E (2003) US Patent 6,376,562 (23 April 2003) to Korea Advanced Institute of Science and Technology. Janda G F (2002), US Patent 6,444,712 (3 September 2002) to ExxonMobil Chemical Patents Inc. Janda G F and Kuechler K H (2002), US Patent 6,486,219 (26 November 2002) to ExxonMobil Chemical Patents Inc. Janssen M J, van Egmond C F Martens L R M and Sher J (2007), US Patent 7,288,689 (30 October 2007) to ExxonMobil Chemical Patents Inc. Kagan Yu B, Liberov L G, Slivinskii E V, Loktev S M, Lin G I et al. (1973), Dokl. Akad. Nauk. SSR, 221, 1093. Kaneko S, Kobayashi Y, Kabata T, Aruga T (2003), US Patent 6,645,442 (11 November 2003) to Mitsubishi Heavy Industries Ltd. Kato K (2002), Hydrocarbon Asia, July/August, 24. Klier K (1984), ‘Methanol and low alcohol synthesis’ in ‘Heterogeneous Catalysis’ (Shapiro ed.), Proc. of the Industry-University Co-operative Chemistry Program of the Department of Chemistry, Texas A&M University, Texas. Kobayashi K and H. Nagai H (1999), US Patent 5,998,489 (7 December 1999) to Mitsubishi Heavy Industries Ltd. Kobayashi K, Osara H, Nagai, H and Ohira, H (2001), US Patent 6,300,380 (9 October 2001) to Mitsubishi Heavy Industries Ltd. Kobayashi K, Osara H, Nagai, H and Ohira H (2007), US Patent 7,189,379 (13 March 2007) to Mitsubishi Heavy Industries Ltd. Kokubu Y, Fujiwara K, Ono H, Sugawara H, Yamamoto K et al. (2000) US Patent 6,028,119 (22 February 2000) to Mitsui Chemicals. Konig P and Gohna H (1998), US Patent 5,827,901 (27 October 1998) to Metallgesellschaft AG. Korosky R M (1996), Pittsburgh Energy Tech. Centre pamphlet (31 July 1996). Kung H H, Lui G and Wilcox D (1984), Amer. Chem. Soc. (Fuel Chem. Div.) Preprints, 29(5), 189. Lattner J R, (2006), US Patent 7,148,261 (12 December 2006) to ExxonMobil Chemical Patents Inc. Lattner J R, (2007), US Patent 7,183,328 (27 February 2007) to ExxonMobil Chemical Patents Inc. Lattner J R, (2008a), US Patent 7,384,958 (10 June 2008) to ExxonMobil Chemical Patents Inc. Lattner J R, (2008b), US Patent 7,470,825 (30 December 2008) to ExxonMobil Chemical Patents Inc. Lattner J R, (2009a), US Patent 7,579,383 (25 April 2009) to ExxonMobil Chemical Patents Inc. Lattner J R, (2009b), US Patent 7,615,578 (10 November 2009) to ExxonMobil Chemical Patents Inc. Lee S (1990), Methanol Synthesis Technology, Boca Raton, CRC Press Inc. Mahajan D (2005), US Patent 6,921,733 (26 July 2005) to Brookhaven Science Associates, LLC.

© Woodhead Publishing Limited, 2011

Methanol and dimethyl ether production from synthesis gas


Marsden W L, Wainwright M S and Friedrich J B (1980), ‘Zinc-promoted Raney copper catalysts for methanol synthesis’, Ind. &Eng. Chem. Product Research and Development, 19, 551. Matsumura Y and Shen W-J, (2002), US Patent 6,342,538 (29 January 2002) to Agency of Industrial Science & Technology, Tokyo, Japan. Monti D M, Wainwright M S, Trimm D L and Cant N W (1985), ‘Kinetics of the vapourphase hydrogenolysis of methyl formate over copper on silica catalysts’, Ind. & Eng. Chem. Product Research and Development, 24, 397. Monti D M, Kohler M A, Wainwright M S, Trimm D L and Cant N W (1986), ‘Liquid phase hydrogenolysis of methyl formate in a semi batch reactor’, Applied Catal., 22, 123. Nielsen P E H, Jaksland C and Perregaard J (2005), US Patent 6,881.759 (19 April 2005) to Haldor Topsøe A/S. Owen G, Hawkes C M, Lloyd D, Jennings J R, Lambert R M et al. (1987), Applied Catal., 33, 405. Patel N M, Wang S-I and Genkin E S (2004), US Patent 6,706,770 (16 March 2004) to Air Products and Chemicals Inc. Peng X-D, Parris G E, Toseland B A and Battavio P J (1998), US Patent 5,753,716 (19 May 1998). to Air Products and Chemicals Inc. Peng X-D, Wang A W and Toseland B A (2000), US Patent 6,069,180 (30 May 2000) to Air Products and Chemicals Inc. Peng X-D, Diamond B W, Tsao C-T and Bhatt B L (2002), US Patent 6,458,856 (1 October 2002) to Air Products and Chemicals Inc. Pitchai R and Klier K (1986), ‘Partial oxidation of methane’, Catalysis Reviews in Science and Engineering, 28(1) 13–88. Poutsma M L, Rabo J A and Risch A P (1978a), US Patent 4,119 656 (10 October 1978) to Union Carbide. Poutsma M L, Elek L F, Ibarbia P A, Risch A P and Rabo J A, (1978b), ‘Selective formation of methanol from synthesis gas over palladium catalysts’, J. Catal., 52, 157. Rhodes A K (1994), Oil & Gas Journal, Aug. 8, 65. Sanfilippo D, Basin L and Marchionna M (1999), US Patent 5,856,585 (5 January 1999) to Snamprogetti SpA. Seddon D (2006) Gas Usage and Value – The Technology and Economics of Natural Gas Use in the Process Industries, Tulsa, PennWell. Seiki Y, Imai T, Kobayashi K, Osora H, Kuwanda C et al. (2005a), US Patent,6,875,794 (5 April 2005) to Mitsubishi Heavy Industries Ltd. Seiki Y, Imai T, Kobayashi K, Osora H, Kuwanda C et al. (2005b) US Patent, 6,894,080 (17 May 2005) to Mitsubishi Heavy Industries Ltd. Shaw J M (2004), US Patent 6,736,955 (18 May 2004) to Technology Convergence Inc. Shikada T, Ohno Y, Ogawa T, Mizuguchi M, Ono M et al. (2000), US Patent 6,147,125 (14 November 2000) to NKK Corporation. Shiroto Y, Kawazuishi K, Tauchi M and Shimura M (2003), US Patent 6,656,978 (2 December 2003) to Chiyoda Corporation. Short G D and Jennings J R (1984), European Patent 117,944 (12 September 1984) to ICI. Sofianos A, Armbruster E, Frei O and Heveling J (2000) US Patent 6.054,497 (25 April 2000) to Lonza AG. Sorum P A and Onsager O T (1984), Proc. of 8th. Int. Cong on Catalysis, Berlin, II, 233. Stiles A B and Koch T A (1995), Catalyst Manufacture, New York, Marcel Dekker, Inc.

© Woodhead Publishing Limited, 2011


Advances in clean hydrocarbon fuel processing

Stratton A (1982) ‘A Simplified Approach to Calculating Product Cost’, Technical Note 3, Economic Assessment Service, IEA Coal Research, London. Toyo Engineering Corporation (2005) ‘Dimethyl ether (DME)’, in ‘Petrochemical Processes 2006’, Hydrocarbon Processing, CDROM. Toyo Engineering Corporation, China. Twigg M V (1989) Catalyst Handbook, London, Wolfe Publishing Ltd. van Egmond C F (2006a), US Patent 7,151,198 (19 December 2006) to ExxonMobil Chemical Patents Inc. van Egmond C F (2006b), US Patent 7,161,051 (9 January 2007) to ExxonMobil Chemical Patents Inc. van Egmond C F, Argo A, Xu T, Janssen M J and Sher J (2007), US Patent 7,196,239 (27 March 2007) to ExxonMobil Chemical Patents Inc. van Egmond C F, Nicoletti M P, Nelson E D and Jenkins C D (2008), US Patent, 7,396,972 (8 July 2008) to ExxonMobil Chemical Patents Inc. Wu Y, Luo S, Liu X, Chen W, Jai C et al. (2000), US Patent 6,090,741 (18 July 2000) to Chengdu Institute of Organic Chemistry.

© Woodhead Publishing Limited, 2011