6.2.1 Steam-methane reforming

Steam-methane reforming is the benchmark process that has been employed over a period of several decades for hydrogen production. The process involves reforming natural gas in a continuous catalytic process in which the major reaction is the formation of carbon monoxide and hydrogen from methane and steam:

${\mathrm{CH}}_4+{\mathrm{H}}_2\mathrm{O}=\mathrm{CO}+3{\mathrm{H}}_2{\mathrm{\Delta H}}_{298\mathrm{K}}=+97,400\mathrm{Btu}/\mathrm{lb}$

Higher molecular weight feedstocks can also be reformed to hydrogen:

${\mathrm{C}}_3{\mathrm{H}}_8+3{\mathrm{H}}_2\mathrm{O}\to 3\mathrm{CO}+7{\mathrm{H}}_2$

That is,

${\mathrm{C}}_{\mathit{n}}{\mathrm{H}}_{\mathit{m}}+\mathit{n}{\mathrm{H}}_2\mathrm{O}\to \mathit{n}\mathrm{CO}+\left(0.5\mathit{m}+\mathit{n}\right){\mathrm{H}}_2$

In the actual process, the feedstock is first desulfurized by passage through activated carbon, which may be preceded by caustic and water washes. The desulfurized material is then mixed with steam and passed over a nickel-based catalyst (730 to 845 °C, 1350 to 1550 °F and 400 psi. Effluent gases are cooled by the addition of steam or condensate to about 370 °C (700 °F), at which point carbon monoxide reacts with steam in the presence of iron oxide in a shift converter to produce carbon dioxide and hydrogen:

$\mathrm{CO}+{\mathrm{H}}_2\mathrm{O}={\mathrm{CO}}_2+{\mathrm{H}}_2$

The carbon dioxide is removed by amine washing; the hydrogen is usually a high-purity (> 99%) material.

Steam reforming of natural gas (sometimes referred to as steam-methane reforming, SMR) is the part of the gas-refining process where the natural gas is converted to syngas, which is further used in the synthesis to methanol or Fischer-Tropsch products. Hydrogen-rich synthesis gas can also be used directly for hydrogen enrichment. The technology for steam reforming is of great interest because this part of the process represents a substantial portion of the investment costs. The reforming section costs about 60 to 80% of the total cost of the entire gas-refining plant. Improvements and cost savings in the reforming section will therefore become very noticeable in the total plant cost.

Steam reforming is an exothermic reaction that is carried out by passing a preheated mixture comprising methane (sometimes substituted by natural gas having high methane content) and steam through catalyst-filled tubes. The products of the process are a mixture of hydrogen, carbon monoxide, and carbon dioxide. To maximize the conversion of the methane feed, primary and secondary reformers are often used – the primary reformer elicits a 90 to 92% v/v conversion of methane. Here, the hydrocarbon feed is partially reacted with steam at 900 °C (1650 °F) at 220 to 500 psi over a nickel-alumina catalyst to produce a synthesis gas in which the hydrogen/carbon monoxide (H₂/CO) ratio is on the order of 3:1. Any unconverted methane is reacted with oxygen at the top of a secondary autothermal reformer, which contains a nickel catalyst in the lower region of the vessel.

In autothermal (or secondary) reformers, the oxidation of methane supplies the necessary energy and carries out either simultaneously or in advance of the reforming reaction (Brandmair, Find, & Lercher, 2003; Ehwald, Kürschner, Smejkal, & Lieske, 2003; Nagaoka, Jentys, & Lecher, 2003). The equilibrium of the methane steam reaction and the water-gas shift reaction determines the conditions for optimum hydrogen yields. The optimum conditions for hydrogen production require high temperature at the exit of the reforming reactor (800 to 900 °C; 1470 to 1650 °F); high excess of steam (molar steam-to-carbon ratio of 2.5 to 3); and relatively low pressures (below 450 psi). Most commercial plants employ supported nickel catalysts for the process.

One way of overcoming the thermodynamic limitation of steam reforming is to remove either hydrogen or carbon dioxide as it is produced, hence shifting the thermodynamic equilibrium toward the product side. The concept for sorption-enhanced methane steam reforming is based on in situ removal of carbon dioxide by a sorbent such as calcium oxide (CaO).

$\mathrm{CaO}+{\mathrm{CO}}_2\to {\mathrm{CaCO}}_3$

Sorption enhancement enables lower reaction temperatures, which may reduce catalyst coking and sintering, while enabling use of less expensive reactor wall materials. In addition, heat release by the exothermic carbonation reaction supplies most of the heat required by the endothermic reforming reactions. However, energy is required to regenerate the sorbent to its oxide form by the energy-intensive calcination reaction:

${\mathrm{CaCO}}_3\to \mathrm{CaO}+{\mathrm{CO}}_2$

Use of a sorbent requires either that there be parallel reactors operated alternatively and out of phase in reforming and sorbent regeneration modes, or that sorbent be continuously transferred between the reformer/carbonator and regenerator/calciner (Balasubramanian, Ortiz, Kaytakoglu, & Harrison, 1999; Hufton, Mayorga, & Sircar, 1999).

Synthesis gas produced from natural gas (or coal or other carbonaceous feedstocks) is the building block in the synthesis of ammonia, methanol, Fischer-Tropsch (FT) fuels, hydrogen for hydrocracking at oil refineries, Oxo-alcohols, and other fine chemicals. The gas composition varies with the intended use of the syngas; ammonia production requires a molar H2/N2 ratio of 3, and for hydrogen production, the H2 content should be as high as possible. Because of the active shift reaction, both carbon monoxide and carbon dioxide are reactants in the methanol synthesis and in high temperature FTS. The syngas composition is therefore specified by a stoichiometric number [SN = (H2 – CO2)/(CO + CO2)] which should be close to 2. On the other hand, only carbon monoxide is a reactant for the low-temperature FTS, and the synthesis gas should have a H2/CO ratio close to 2. These different syngas compositions are achieved by using different types of reactor technology and by varying the amount of added steam and possibly oxygen or air, which is discussed later. The synthesis gas composition is also dependent on the feed gas composition and on the outlet temperature and pressure of the reforming reactor (Chadeesingh, 2011).

The higher molecular weight hydrocarbons that are also constituents of natural gas (Speight, 2007, 2014) are converted to methane in an adiabatic pre-reformer upstream of the steam reformer. In the pre-reformer, all higher hydrocarbons (C_2 +) are converted into a mixture of methane, hydrogen, and carbon oxides:

$\mathrm{C}\mathit{n}\mathrm{H}\mathit{m}+\mathit{n}{\mathrm{H}}_2\mathrm{O}\to \mathit{n}\mathrm{CO}+\left(\mathit{n}+\mathit{m}/2\right){\mathrm{H}}_2$

$3{\mathrm{H}}_2+\mathrm{CO}\leftrightarrow {\mathrm{CH}}_4+{\mathrm{H}}_2\mathrm{O}$

$\mathrm{CO}+{\mathrm{H}}_2\mathrm{O}\leftrightarrow {\mathrm{H}}_2+{\mathrm{CO}}_2$

The pre-reforming process utilizes an adiabatic fixed-bed reactor with highly active nickel catalysts. The reactions take place at temperatures of approximately 350 to 550 °C (650 to 1020 °F), which makes it possible to preheat the steam reformer feed to higher temperatures without getting problems with olefin formation from the higher hydrocarbons. Olefins are unwanted in the steam reformer feed because they generally cause coking of the catalyst pellets at high temperatures. Preheating of the steam reformer feed is of great advantage because the reformer unit can be scaled down to a minimum size (Aasberg-Petersen et al., 2001; Aasberg-Petersen, Christensen, Stub Nielsen, & Dybkjær, 2002; Hagh, 2004).

The reactions are catalyzed by pellets coated with nickel and are highly endothermic overall. Effective heat transport to the reactor tubes and further into the center of the catalytic fixed bed is therefore a very important aspect during design and operation of steam reformers. The reactions take place in several tubular fixed-bed reactors of low diameter-to-height ratio to ensure efficient heat transport in radial direction. The process conditions are typically 300 to 600 psi bar with inlet temperature of 300 to 650 °C (570 to 1200 °F) and outlet temperature of 700 to 950 °C (1290 to 1740 °F). There is often an approach to equilibrium of about 5 to 20 °C, which means that the outlet temperature is slightly higher than the equilibrium temperature calculated from the actual outlet composition (Rostrup-Nielsen, Christiansen, & Bak Hansen, 1988).

In a pre-reformer, whisker carbon can be formed either from methane or higher molecular weight hydrocarbons. The lower limit of the H₂O/C ratio depends on a number of factors, including the feed gas composition, the operating temperature, and the choice of catalyst. In a pre-reformer operating at low H₂O/C-ratio, the risk of carbon formation from methane is most pronounced in the reaction zone where the temperature is highest. Carbon formation from higher molecular weight hydrocarbons can take place only in the first part of the reactor with the highest concentrations of C2+compounds.

In addition, two water-gas shift (WGS) reactors are used downstream of the secondary reformer to adjust the H₂/CO ratio, depending on the end use of the steam reformed products. The first of the two shift reactors utilize an iron-based catalyst that is heated to approximately 400 °C (750 °F), whereas the second shift reactor operates at approximately 200 °C (390 °F) and contains a copper-based catalyst.

The deposition of carbon on the catalyst (coking) can be an acute problem with the use of Ni-based catalysts in the primary reformer (Alstrup, 1988; Rostrup-Nielsen, 1984, 1993). The carbon deposition reactions occur in parallel with the reforming reactions and are undesirable, as they cause poisoning of the surface of the catalyst pellets. This leads to lower catalyst activity and the need for more frequent catalyst reloading. The coking reactions are the CO reduction, methane cracking, and Boudouard reaction, given by the respective equilibrium reactions:

$\mathrm{CO}+{\mathrm{H}}_2+{\mathrm{H}}_2\mathrm{O}\to \mathrm{C}+{\mathrm{H}}_2\mathrm{O}$

${\mathrm{CH}}_4\leftrightarrow \mathrm{C}+2{\mathrm{H}}_2$

$2\mathrm{CO}\leftrightarrow \mathrm{C}+{\mathrm{CO}}_2$

Thus, low steam excess can lead to critical conditions causing coke formation – equilibrium calculations of the coking reactions can be a useful tool for predicting the danger for catalyst poisoning but the reaction kinetics may nevertheless be so slow that coking is no concern. A complete analysis should therefore also involve kinetic calculations, which will be feedstock-dependent expressions for these reactions.

Traditionally, steam reformers have been run with a steam/carbon ratio of 2 to 4 to ensure low coking potential. It is desirable to reduce this ratio for methanol and FTS purposes because it will give great cost savings in form of smaller reformer units with higher methane conversion. Technical developments such as new noble metal catalysts and the use of pre-reformers are continually decreasing the feasible team/carbon ratio.

A successful technique is to use a steam/carbon ratio in the feed gas that does not allow the formation of carbon, but the process has a lower efficiency. Another approach is to use sulfur passivation, which led to the development of the SPARG process (Rostrup-Nielsen, 1984, 2006; Udengaard, Hansen, Hanson, & Stal, 1992). This technique utilizes the principle that the reaction leading to the deposition of carbon requires a larger number of adjacent nickel atoms on the catalyst surface than does steam reforming. A third method is to use Group VIII metals (such as platinum) that do not form carbides.

The most common reactor concept for steam reforming of natural gas is the fired steam reformer. Natural gas and the tail gas from the synthesis loop are burned in a firebox where several reactor tubes are placed in rows with a number of 40 to 400 tubes. The reactor tubes are about 33 to 40 feet with diameters of about 4 to 5 inches. The reactions for conversion of natural gas to synthesis gas take place over the catalytic beds in the reactor tubes. The burners can be located in different places: on the roof, on the floor, on leveled terraces on the walls, or on the walls (side-fired heating).

The top-fired steam reformer must be operated carefully because the tube wall temperature and heat flux show a peak in the upper part of the reformer. The bottom-fired reformers achieve a stable heat flux profile along the tube length, which causes high tube skin temperatures at the reactor outlet. The terrace wall-fired reformer is a modification of the bottom fired reformer and has some smaller problem with high metal temperatures. The side-fired reformer has the most effective design and is also the most flexible reformer, both in design and in operation (Dybkjær, 1995). This configuration has the highest total heat flux possible combined with the lowest heat flux where the tube skin temperature is at its highest. In this type of reformer, it is possible to combine a low steam-to-carbon ratio with a high outlet temperature. The most critical operation parameter is the maximum temperature difference over the tube wall, not the maximum heat flux (Aasberg-Petersen et al., 2001).

6.2.2 Autothermal reforming

Autothermal reforming (ATR) uses oxygen and carbon dioxide or steam in a reaction with methane to form synthesis. The reaction takes place in a single chamber where the methane is partially oxidized. The reaction is exothermic due to the oxidation. The main difference between autothermal reforming and steam-methane reforming is that steam-methane reforming does not use or require oxygen. The advantage of autothermal reforming is that H₂/CO can be varied, which is particularly useful for producing certain second-generation biofuels such as dimethyl ether synthesis, which requires a 1:1 H₂/CO ratio.

The process was developed in the 1950s and is used in commercial applications to provide syngas for ammonia and methanol synthesis. In the case of ammonia production, where high H₂/CO ratios are needed, the process is operated at high steam/carbon ratios. In the case of methanol synthesis, the required H₂/CO ratio is provided by manipulating the carbon dioxide recycle. In fact, development and optimization of this technology has led to cost-effective operation at very low steam/carbon feed ratios to produce CO-rich syngas, for example, which is preferred in FTS. These are the advantages of using the autothermal reactor: (1) the reactor is compact in design and therefore has a smaller footprint; (2) it has flexibility in its operation, with short startup periods and fast load changes; and (3) it is a soot-free operation. In addition, the reactor offers a better economic profile.

In the process, an organic feedstock (such as natural gas) and steam (there may also be low amounts of carbon dioxide in the feed) are mixed directly with oxygen and air in the reformer. The reformer itself comprises a refractory-lined vessel that contains the catalyst, together with an injector located at the top of the vessel. Partial oxidation reactions occur in the combustion zone of the reactor and the gaseous mixture then flows through a catalyst bed where the actual reforming reactions occur. Heat generated in the combustion zone from partial oxidation reactions is utilized in the reforming zone, so that in the ideal case, it is possible that the process can be in complete heat balance.

The autothermal reforming reactor consists of three zones: (1) the burner, where the feed streams are mixed in a turbulent diffusion flame; (2) the combustion zone, where partial oxidation reactions produce a mixture of carbon monoxide and hydrogen; and (3) the catalytic zone, where the gases leaving the combustion zone attain thermodynamic equilibrium. Key elements in the reactor are the burner and the catalyst bed – the burner provides mixing of the feed streams and the natural gas is converted in a turbulent diffusion flame:

${\mathrm{CH}}_4+3/2{\mathrm{O}}_2\to \mathrm{CO}+2{\mathrm{H}}_2\mathrm{O}$

When carbon dioxide is present in the feed, the H₂/CO ratio produced is on the order of 1:1, but when the process employs steam, the H₂/CO ratio produced is 2.5:1.

$2{\mathrm{CH}}_4+{\mathrm{O}}_2+{\mathrm{CO}}_2\to 3{\mathrm{H}}_2+3\mathrm{CO}+{\mathrm{H}}_2\mathrm{O}$

$4{\mathrm{CH}}_4+{\mathrm{O}}_2+2{\mathrm{H}}_2\mathrm{O}\to 10{\mathrm{H}}_2+4\mathrm{CO}$

The risk of soot formation in an ATR reactor depends on a number of parameters, including feed gas composition, temperature, pressure, and especially burner design. Soot precursors may be formed in the combustion chamber during operation, so it is essential that the design of the burner, catalyst, and reactor is such that the precursors are destroyed by the catalyst bed to avoid soot formation.

Many observers consider the combination of adiabatic pre-reforming and autothermal reforming at low H₂O/C ratios to be the preferred layout for production of synthesis gas for large gas-to-liquids plants.

6.2.3 Combined reforming

Combined reforming incorporates the combination of both steam reforming and autothermal reforming. In such a configuration, the hydrocarbon (e.g., natural gas) is first only partially converted, under mild conditions, to syngas in a relatively small steam reformer (Wang, Stagg-Williams, Noronha, Mattos, & Passos, 2004). The off-gases from the steam reformer are then sent to an oxygen-fired secondary reactor, such as an autothermal reactor, where the unreacted methane is converted to syngas by partial oxidation followed by steam reforming.

Another configuration requires that the hydrocarbon feed be split into two streams that are then fed in parallel, to a steam-reforming reactor and an autothermal reactor (gas-heated reforming). This process is an alternative to the fired steam reformer and has been commercially proven. There is also interest for the gas-heated steam reformer in relation to the FTS of hydrocarbons and methanol production.

6.2.4 Partial oxidation

Partial oxidation is the process in which the feed fuel, such as methane or a suitable hydrocarbonaceous fuel, reacts exothermically in the presence of a small amount of air (Vernon, Green, Cheetham, & Ashcroft, 1990; Rostrup-Nielsen, 2002; Zhu, Zhao, & Deng, 2004). Because incomplete combustion occurs, a gas containing hydrogen and carbon monoxide is produced. The hydrogen can be used to extend the lean limit of diesel, for instance, which indicates a higher efficiency of the fuel and lower pollutants emissions.

Partial oxidation (POX, P_OX) reactions occur when a substoichiometric hydrocarbon-air mixture is partially combusted in a reformer:

${\mathrm{C}}_{\mathit{n}}{\mathrm{H}}_{\mathit{m}}+\left(2\mathit{n}+\mathit{m}\right)/2{\mathrm{O}}_2\to \mathit{n}\mathrm{CO}+\left(\mathit{m}/2\right){\mathrm{H}}_2\mathrm{O}$

Thus, for coal or any carbonaceous feedstocks the reaction can be represented simply as (understanding the in reality the reaction is extremely complex):

${\left[\mathrm{CH}\right]}_{\mathrm{coal}}+{\mathrm{O}}_2\to \mathrm{CO}+{\mathrm{H}}_2$

In the process, the feedstock is partially burned in a simple pre-combustion chamber in the presence of a small amount of air and converted into carbon monoxide and hydrogen. Because partial oxidation is an exothermic reaction, some of the heat of combustion is released. The released energy is converted into heat, which brings the temperature of the gas to approximately 870 °C (1600 °F). The temperature of the gas needs to be lowered before entering the combustion engine. Otherwise, the density of the gas is too low to have a good volumetric efficiency. The resulting gas can be burned in a gas engine.

A thermal partial oxidation reactor is similar to the autothermal reactor with the main difference being no catalyst is used. In the process, the feedstock, which may include steam, is mixed directly with oxygen by an injector that is located near the top of the reaction vessel. Both partial oxidation as well as reforming reactions occur in the combustion zone below the burner.

The principal advantage of the partial oxidation process is the ability of the system to accommodate almost any carbonaceous feedstock, which can comprise very high molecular-weight organic constituents such as petroleum residual and petroleum coke (Gunardson & Abrardo, 1999; Speight, 2014). Additionally, because the emission of nitrogen oxides (NOx) and sulfur oxides (SOx) are minimal, partial oxidation does not leave a large environmental footprint.

On the other hand, very high temperatures, approximately 1300 °C (2370 °F), are required to achieve near complete reaction. This necessitates the consumption of some of the hydrogen and a greater than stoichiometric consumption of oxygen (i.e., oxygen-rich conditions).

Partial oxidation cannot be used for gasifying gasoline, diesel, methanol, or ethanol, because of the decrease in energy content of the fuel. However, the hydrogen-rich gas (hence, the preference for this type of process in the petroleum industry) that is produced by partial oxidation may be used to enrich other fuels. For the production of hydrogen, partial oxidation is often used in combination with steam reforming, using the heat of the partial oxidation for the endothermic steam reforming. However, given that steam reforming can be accomplished by using the energy from the exhaust gases coming out of the combustion engine, there is no need to partially oxidize the fuel first. Doing so would result in loss of the heating value of the fuel, and thus an overall energy loss for the process.

A possible means of improving the efficiency of syngas production is by use of the catalytic partial oxidation (CPOX, CP_OX) technology, which has the potential to offer several advantages over steam reforming and thermal particle oxidation, particularly higher energy efficiency (Enger, Lødeng, & Holmen, 2008). The reaction is not endothermic – as is the case with steam reforming – but slightly exothermic. Furthermore, an H₂/CO ratio of close to 2.0 (i.e., the ideal ratio for the Fischer-Tropsch and methanol synthesis) is produced by this technology, which can occur by either of two routes: direct or indirect.

The indirect route comprises total combustion of methane to carbon dioxide and water, followed by steam reforming and the water-gas shift reaction in which equilibrium conversions can be greater than 90% at ambient pressure. However, in order for an industrial process for this technology to be economically viable, an operating pressure in excess of 300 psi would be required. Unfortunately, at high pressures, equilibrium conversions are lower and, because of the exothermic combustion step, process control is more difficult and there is the potential for temperature runaways.

The direct route occurs by a mechanism involving only surface reaction on the catalyst:

${\mathrm{CH}}_4+0.5{\mathrm{O}}_2\to \mathrm{CO}+2{\mathrm{H}}_2$

Compared with conventional synthesis gas production methods, the direct route would drastically reduce the amount of catalyst used, making it possible to use compact reactors.

6.2.5 Membrane reactors

An innovative technology for combining air separation and natural gas reforming processes is using membrane technology, which has the potential to reduce the cost of syngas generation and hydrocarbon products (Carolan, Chen, & Rynders, 2002; Khassin, 2005). The technology (oxygen transport membranes) can combine five unit operations currently in use: (1) oxygen separation, (2) oxygen compression, (3) partial oxidation, (4) steam methane reforming, and (5) heat exchange. The technology incorporates the use of catalytic components with the membrane to accelerate the reforming reactions.

A patented a two-step process for synthesis gas generation has been developed (Nataraj, Moore, & Russek, 2000) that can be utilized to generate synthesis gas from several feedstocks, including natural gas, associated gas (from crude oil production), light hydrocarbon gases from refineries, and medium-weight hydrocarbon fractions such as naphtha. The first stage comprises conventional steam reforming with partial conversion to synthesis gas and is followed by complete conversion in an ion transport ceramic membrane (ITM) reactor. This combination resolves any issues associated with steam reforming for feedstocks with hydrocarbons higher in molecular weight than methane, because the higher molecular weight hydrocarbons tend to crack and degrade both the catalyst and membrane.

By shifting the equilibrium in the steam reforming process through removal of hydrogen from the reaction zone, membrane reactors can also be used to increase the equilibrium-limited methane conversion. Using Pd-Ag alloy membrane reactors. methane conversion can reach as close to 100 % (Shu, Grandjean, & Kaliaguine, 1995).

6.3.1 Heavy residue gasification and combined cycle power generation

Heavy residues are gasified and the produced gas is purified to fuel gas that is free of contaminants (Gross & Wolff, 2000). As an example, solvent deasphalter residuum (deasphalter bottoms) is gasified by partial oxidation method under pressure of about 570 psi and at a temperature between 1300 and 1500 °C (2370 °F and 2730 °F). The high temperature generates gas stream flows into a waste heat boiler, in which the hot gas is cooled and high pressure saturated steam is generated. The gas from the waste heat boiler is then heat exchanged with the fuel gas and flows to the carbon scrubber, where unreacted carbon particles are removed from the generated gas by water scrubbing.

The gas from the carbon scrubber is further cooled by the fuel gas and boiler feed water and led into the sulfur compound removal section, where hydrogen sulfide (H₂S) and carbonyl sulfide (COS) are removed from the gas to obtain clean fuel gas. This clean fuel gas is heated with the hot gas generated in the gasifier and finally supplied to the gas turbine at a temperature of 250 to 300 °C (480 to 570 °F).

In order to decrease the nitrogen oxide (NO_x) content in the flue gas, two methods can be applied. The first method is the injection of water into the gas turbine combustor. The second method is to selectively reduce the nitrogen oxide content by injecting ammonia gas in the presence of de-NO_x catalyst that is packed in a proper position of the heat recovery steam generator. The latter is more effective that the former to lower the nitrogen oxide emissions to the air.

6.3.2 Hybrid gasification process

In the hybrid gasification process, a coal/residual oil slurry is injected into the gasifier where it is pyrolyzed in the upper part of the reactor to produce gas and chars. The chars produced are then partially oxidized to ash. The ash is removed continuously from the bottom of the reactor.

In this process, coal and vacuum residue are mixed together into slurry to produce clean fuel gas. The slurry fed into the pressurized gasifier is thermally cracked at a temperature of 850 to 950 °C (1560 to 1740 °F) and is converted into gas, tar, and char. The mixture of oxygen and steam in the lower zone of the gasifier converts the char to gaseous products. The gas leaving the gasifier is quenched to a temperature of 450 °C (840 °F) in the fluidized-bed heat exchanger, and is then scrubbed to remove tar, dust, and steam at around 200 °C (390 °F).

Ash is discharged from the gasifier and indirectly cooled with steam and then discharged into the ash hopper. It is burned with an incinerator to produce process steam. Coke deposited on the silica sand is removed in the incinerator.

6.3.3 Hydrocarbon gasification

The gasification of hydrocarbons to produce hydrogen is a continuous, noncatalytic process that involves partial oxidation of the hydrocarbon and one of several processes that are used for gasification of carbonaceous fuels to gaseous products (Breault, 2010).

In the process, air or oxygen, with steam or carbon dioxide, is used as the oxidant at 1095 to 1480 °C (2000 to 2700 °F). Any carbon produced (2 to 3% w/w of the feedstock) during the process is removed as a slurry in a carbon separator and pelletized for use either as a fuel or as raw material for carbon-based products.

6.3.4 Hypro process

Due to its abundance and high H/C ratio (highest among all hydrocarbons), methane is an obvious source for hydrogen. The steam reforming of methane represents the current trend for hydrogen production (Hypro process). Other popular methods of hydrogen production include autothermal reforming and partial oxidation. However, if hydrogen is the desire product, all these processes involve the formation of large amounts of unwanted carbon monoxide and carbon dioxide (COx) as a by-product.

Hydrogen production routes, which do not require complex COx removal procedures, are therefore desired for production of high-purity hydrogen. Thus, there is much interest in the catalytic decomposition of natural gas, whose major constituent is methane, for production of hydrogen. Given that only hydrogen and carbon are formed in the decomposition process, the separation of products is not an issue. The other main advantage is the simplicity of the methane decomposition process as compared to conventional methods. For example, the high- and low-temperature water-gas shift reactions and carbon dioxide removal step (involved in the conventional methods) are completely eliminated. Catalyst regeneration is extremely important for the practical application of the clean hydrogen production process.

The hypro process is a continuous catalytic method for hydrogen manufacture from natural gas or from refinery effluent gases, especially the decomposition of methane to hydrogen and carbon (Choudhary & Goodman, 2006; Choudhary, Sivadinarayana, & Goodman, 2003):

${\mathrm{CH}}_4\to \mathrm{C}+2{\mathrm{H}}_2$

Hydrogen is recovered by phase separation to yield hydrogen of about 93% purity, and the principal contaminant is methane.

6.3.5 Pyrolysis processes

There has been recent interest in the use of pyrolysis processes to produce hydrogen. Specifically, the interest has focused on the pyrolysis of methane (natural gas) and hydrogen sulfide.

Natural gas is readily available and offers a relatively rich stream of methane with lower amounts of ethane, propane, and butane also present. The thermocatalytic decompositon of natural gas hydrocarbons (c.f., hypro process) offers an alternate method for the production of hydrogen (Dahl & Weimer, 2001; Uemura, Ohe, Ohzuno, & Hatate, 1999; Weimer et al., 2000):

${\mathrm{C}}_{\mathit{n}}{\mathrm{H}}_{\mathit{m}}\to \mathit{n}\mathrm{C}+\left(\mathit{m}/2\right){\mathrm{H}}_2$

The production of hydrogen by direct decomposition of hydrogen sulfide has also been proposed (Clark & Wassink, 1990; Donini, 1996; Luinstra, 1996; Zaman & Chakma, 1995). Hydrogen sulfide decomposition is a highly endothermic process, and equilibrium yields are poor (Clark, Dowling, Hyne, & Moon, 1995). At temperatures less than 1500 °C (2730 °F), the thermodynamic equilibrium is unfavorable toward hydrogen formation. However, in the presence of catalysts such as platinum-cobalt at 1000 °C (1830 °F), disulfides of molybdenum (Mo) or tungsten (W) at 800 °C (1470 °F) (Kotera, Todo, & Fukuda, 1976), or other transition metal sulfides supported on alumina at 500 to 800 °C (930 to 1470 °F), decomposition of hydrogen sulfide proceeds. In the temperature range of about 800 to 1500 °C (1470 to 2730 °F), thermolysis of hydrogen sulfide can be treated simply:

${\mathrm{H}}_2\mathrm{S}\to {\mathrm{H}}_2+1/\mathit{x}{\mathit{S}}_{\mathit{x}}{\mathrm{\Delta H}}_{298\mathrm{K}}=+34,300\mathrm{Btu}/\mathrm{lb}$

where x = 2. Outside this temperature range, multiple equilibria may be present, depending on temperature, pressure, and relative abundance of hydrogen and sulfur (Clark & Wassink, 1990).

In addition, the steam-iron process is an established process, which was used for the production of hydrogen from cokes at the beginning of the twentieth century. However, the process could not compete with the later-developed steam reforming of methane, and so the process fell into disuse. The renewed interest in the development of the steam-iron process is mainly focused on the use of renewable energy sources, like biomass. In this thesis, the production of hydrogen by the steam-iron process from pyrolysis oil is studied. Pyrolysis oil, obtained from the pyrolysis of biomass, is used to facilitate transportation and to simplify gasification and combustion processes, before being processed to hydrogen. The benefit of the steam-iron process compared to other thermo-chemical routes of biomass is that hydrogen can be produced in a two-step redox cycle, without the need of any purification steps (such as pressure-swing adsorption) (Bleeker, 2009; Bleeker, Kersten, & Veringa, 2007).

6.3.6 Shell gasification process

The Shell gasification process (partial oxidation process) is a flexible process for generating syngas, principally hydrogen and carbon monoxide, for the ultimate production of high-purity, high-pressure hydrogen, ammonia, methanol, fuel gas, and town gas, or for reducing gas by reaction of gaseous or liquid hydrocarbons with oxygen, air, or oxygen-enriched air. Traditionally, petroleum residues have been sold as marine bunker fuel or used on-site as furnace fuel. However, with changing legislation, refineries are under pressure to reduce both their emissions and the sulfur content of their products. In addition, the market for fuel oil is shrinking. The Shell gasification process can be combined with other upgrading and treating technologies to convert a wide range of low-value residue into syngas.

The most important step in converting heavy residue into industrial gas is the partial oxidation of the oil using oxygen with the addition of steam. The gasification process takes place in an empty, refractory-lined reactor at temperatures of about 1400 °C (2550 °F) and pressures between 29 and 1140 psi. The chemical reactions in the gasification reactor proceed without catalyst to produce gas that contains carbon amounting to some 0.5 to 2% by weight, based on the feedstock. The carbon is removed from the gas with water, extracted in most cases with feed oil from the water, and returned to the feed oil. The high reformed gas temperature is utilized in a waste heat boiler for generating steam. The steam is generated at 850 to 1565 psi. Some of this steam is used as process steam and for oxygen and oil preheating. The surplus steam is used for energy production and heating purposes.

6.3.7 Steam-naphtha reforming

Liquid feedstocks, either liquefied petroleum gas or naphtha, can also provide backup feed for the steam-methane reformer, if there is a risk of natural gas curtailments (Breault, 2010; Rostrup-Nielsen & Christiansen, 2011). The feed-handling system needs to include a surge drum, feed pump, a vaporizer (usually steam-heated) followed by further heating before desulfurization. The sulfur in liquid feedstocks occurs as mercaptans, thiophene derivatives, or higher boiling compounds. These compounds are stable and will not be removed by zinc oxide, therefore a hydrogenation unit will be required. In addition, as with refinery gas, olefins must also be hydrogenated if they are present.

Thus, steam-naphtha reforming is a continuous process for the production of hydrogen from liquid hydrocarbons. In fact, it is similar to steam-methane reforming, which is one of several possible processes for the production of hydrogen from low-boiling hydrocarbons other than ethane (Brandmair et al., 2003; Find, Nagaoka, & Lercher, 2003; Muradov, 1998; Murata, Ushijima, & Fujita, 1997). A variety of naphtha-types in the gasoline boiling range may be employed, including feeds containing up to 35% aromatics. Thus, following pre-treatment to remove sulfur compounds, the feedstock is mixed with steam and taken to the reforming furnace (675 to 815 °C, 1250 to 1500 °F, 300 psi, where hydrogen is produced.

6.3.8 Texaco gasification (partial oxidation) process

The Texaco gasification (partial oxidation) process is a partial oxidation gasification process for generating synthesis gas (Breault, 2010). The characteristic of the process is to inject feedstock together with carbon dioxide, steam, or water into the gasifier. Therefore, solvent deasphalted residue or petroleum coke rejected from any coking method can be used as feedstock for this gasification process. The produced gas from this gasification process can be used for the production of high-purity, high-pressurized hydrogen, ammonia, and methanol. The heat recovered from the high-temperature gas is used for the generation of steam in the waste heat boiler. Alternatively the less expensive quench-type configuration is preferred when high-pressure steam is not needed or when a high degree of shift is needed in the downstream carbon monoxide converter.

In the process, the feedstock, together with the feedstock carbon slurry recovered in the carbon recovery section, is pressurized to a given pressure, mixed with high-pressure steam, and then blown into the gas generator through the burner together with oxygen.

The gasification reaction is a partial oxidation of hydrocarbons to carbon monoxide and hydrogen:

${\mathrm{C}}_{\mathit{x}}{\mathrm{H}}_{2\mathit{y}}+\mathit{x}/2{\mathrm{O}}_2\to \mathit{x}\mathrm{CO}+\mathit{y}{\mathrm{H}}_2$

${\mathrm{C}}_{\mathit{x}}{\mathrm{H}}_{2\mathit{y}}+\mathit{x}{\mathrm{H}}_2\mathrm{O}\to \mathit{x}\mathrm{CO}+\left(\mathrm{x}+\mathrm{y}\right){\mathrm{H}}_2$

The gasification reaction is instantly completed, thus producing gas mainly consisting of H₂ and CO (H₂ + CO = > 90%). The high-temperature gas leaving the reaction chamber of the gas generator enters the quenching chamber linked to the bottom of the gas generator and is quenched to 200 to 260 °C (390 to 500 °F) with water.

6.3.9 Recovery from fuel gas

Recovering of hydrogen from refinery fuel gas can help refineries satisfy high hydrogen demand. Cryogenic separation is typically viewed as being the most thermodynamically efficient separation technology. The basic configuration for hydrogen recovery from refinery gases involves a two-stage partial condensation process, with post purification via pressure swing adsorption (Dragomir et al., 2010). The major steps in this process involve first compressing and pre-treating the crude refinery gas stream before chilling to an intermediate temperature (–60 to –120°F). This partially condensed stream is then separated in a flash-drum after which the liquid stream is expanded through a Joule-Thompson valve to generate refrigeration and then is fed to the wash column. Optionally, the wash column can be replaced by a simple flash drum.

A crude liquefied petroleum gas (LPG) stream is collected at the bottom of the column, and a methane-rich vapor is obtained at the top. The methane-rich vapor is sent to compression and then to fuel. The vapor from the flash drum is further cooled in a second heat exchanger before being fed to a second flash drum where it produces a hydrogen-rich stream and a methane-rich liquid. The liquid is expanded in a Joule-Thomson valve to generate refrigeration, and then is sent for further cooling. Next, the hydrogen-rich gas is sent to the pressure swing adsorption unit for further purification. The tail gas from this unit is compressed and returned to fuel, together with the methane-rich gas.

6.4.1 Purification

The processes that have been developed for gas cleaning (Mokhatab, Poe, & Speight, 2006; Speight, 2007, 2008) vary from a simple once-through wash operation to complex multistep systems with options for recycle of the gases (Mokhatab et al., 2006). In some cases, process complexities arise because of the need for recovery of the materials used to remove the contaminants or even recovery of the contaminants in the original, or altered, form.

In more general terms, gas cleaning is divided into removal of particulate impurities and removal of gaseous impurities. For the purposes of this chapter, the latter operation includes the removal of hydrogen sulfide, carbon dioxide, sulfur dioxide, and products that are not related to synthesis gas and hydrogen production. However, there is also need for subdivision of these two categories as dictated by needs and process capabilities: (1) coarse cleaning whereby substantial amounts of unwanted impurities are removed in the simplest, most convenient, manner; (2) fine cleaning for the removal of residual impurities to a degree sufficient for the majority of normal chemical plant operations, such as catalysis or preparation of normal commercial products, or cleaning to a degree sufficient to discharge an effluent gas to atmosphere through a chimney; and (3) ultra-fine cleaning where the extra step (as well as the extra expense) is justified by the nature of the subsequent operations or the need to produce a particularly pure product.

Contrary to the general belief of some scientists and engineers, all gas-cleaning systems are not alike, and having a good understanding of the type of gaseous effluents from coal-based processes is necessary to implementing the appropriate solution. The design of a gas-cleaning system must always take into account the operation of the upstream installations, because every process will have a specific set of requirements. In some cases, the application of a dry dusting removal unit may not be possible and thus requires a special process design of the wet gas-cleaning plant. Thus, the gas-cleaning process must always be of optimal design – one for both the upstream and downstream processes.

Gas processing, although generally simple in chemical and/or physical principles, is often confusing because of the frequent changes in terminology and, often, lack of cross-referencing (Mokhatab et al., 2006; Speight, 2007,, 2008, 2013a, 2014). Although gas processing employs different process types, there is always overlap between the various concepts. And, with the variety of possible constituents and process operating conditions, a universal purification system cannot be specified for economic application in all cases.

Nevertheless, the first step in gas cleaning is usually a device to remove large particles of carryover (entrained) material coal and other solid materials (Mokhatab et al., 2006; Speight, 2007, 2008). This is followed by cooling, quenching, or washing to condense tars and oils and to remove dust and water-soluble materials from the gas stream. Water washing is desirable for simplicity in gas cleaning; however, the purification of this water is not simple.

Clean-up steps and their sequence can be affected by the type of gas produced and its end use (Mokhatab et al., 2006; Speight, 2007, 2008). The minimum requirement in this respect would be the application of low-heat value (low-Btu) gas produced from low-sulfur anthracite coal as a fuel gas. The gas may pass directly from the gasifier to the burners and, in this case, the burners are the clean-up system. Many variations on this theme are possible; also, the order of the cleanup stages may be varied.

The selection of a particular process-type for gas cleaning is not simple. Several factors have to be considered, not the least of which is the constitution of the gas stream that requires treatment. Indeed, process selectivity indicates the preference with which the process will remove one acid gas component relative to (or in preference to) another. For example, some processes remove both hydrogen sulfide and carbon dioxide; other processes are designed to remove hydrogen sulfide only (Mokhatab et al., 2006; Speight, 2007, 2014).

Gas cleaning by absorption by a liquid or adsorption by use of a solid sorbent is one of the most widely applied operations in the chemical and process industries (Mokhatab et al., 2006; Speight, 2007). Some processes have the potential for sorbent regeneration, but in a few cases, the process is applied in a nonregenerative manner. The interaction between sorbate and sorbent may either be physical in nature or consist of physical sorption followed by chemical reaction. Other gas stream treatments use the principle of chemical conversion of the contaminants with the production of “harmless” (noncontaminant) products or to substances that can be removed much more readily than the impurities from which they are derived (Mokhatab et al., 2006; Speight, 2007, 2008).

Any gases, such as hydrogen sulfide and/or carbon dioxide, that are the products of coal processing can be removed by application of an amine washing procedure.

$2{\mathrm{RNH}}_2+{\mathrm{H}}_2\mathrm{S}\to {\left({\mathrm{RNH}}_3\right)}_2\mathrm{S}$

${\left({\mathrm{RNH}}_3\right)}_2\mathrm{S}+{\mathrm{H}}_2\mathrm{S}\to 2{\mathrm{RNH}}_3\mathrm{HS}$

$2{\mathrm{RNH}}_2+{\mathrm{CO}}_2+{\mathrm{H}}_2\mathrm{O}\to {\left({\mathrm{RNH}}_3\right)}_2{\mathrm{CO}}_3$

${\left({\mathrm{RNH}}_3\right)}_2{\mathrm{CO}}_3+{\mathrm{H}}_2\mathrm{O}\to 2{\mathrm{RNH}}_3{\mathrm{HCO}}_3$

There are also solvent extraction methods for producing low-sulfur and low-mineral matter coal, but hydrotreatment of the coal extract is also required. In these methods, the organic material is extracted from the inorganic material in coal. A study has indicated that solvent-refined coal will probably not penetrate the power generation industry on a large scale for several years to come.

In addition to hydrogen sulfide and carbon dioxide, gas streams may contain other contaminants such as sulfur dioxide, mercaptans, and carbonyl sulfide. The presence of these impurities may eliminate some of the sweetening processes because some processes will remove large amounts of acid gas but not to a sufficiently low concentration. On the other hand, there are those processes that are not designed to remove (or are incapable of removing) large amounts of acid gases, yet they are capable of removing the acid gas impurities to very low levels when the acid gases are there in low-to-medium concentrations in the gas stream.

Many different methods have been developed for carbon dioxide and hydrogen sulfide removal, some of which are briefly discussed here. Concentrates of hydrogen sulfide obtained as by-products of gas desulfurization are often converted by partial oxidation to elemental sulfur (Claus process) (Mokhatab et al., 2006; Speight, 2007,, 2013a, 2014).

Pressure swing adsorption units use beds of solid adsorbent to separate impurities from hydrogen streams leading to high-purity, high-pressure hydrogen and a low-pressure tail gas stream containing the impurities and some of the hydrogen. The beds are then regenerated by depressuring and purging. Part of the hydrogen (up to 20% v/v) may be lost in the tail gas.

Pressure swing adsorption is generally the purification method of choice for steam reforming units because of its production of high-purity hydrogen. It is also used for purification of refinery off-gases, where it competes with membrane systems.

Many hydrogen plants that formerly used a wet scrubbing process for hydrogen purification are now using the pressure swing adsorption (PSA) for purification (Speight, 2007, 2014). The pressure swing adsorption process is a cyclic process that uses beds of solid adsorbent to remove impurities from the gas and generally produces higher-purity hydrogen (99.9% v/v purity compared to less than 97% v/v purity). The purified hydrogen passes through the adsorbent beds with only a tiny fraction absorbed, and the beds are regenerated by depressurization followed by purging at low pressure.

When the beds are depressurized, a waste gas (or tail gas) stream is produced and consists of the impurities from the feed (carbon monoxide, carbon dioxide, methane, and nitrogen) plus some hydrogen. This stream is burned in the reformer as fuel and reformer operating conditions in a pressure swing adsorption plant are set so that the tail gas provides no more than about 85% v/v of the reformer fuel. This gives good burner control because the tail gas is more difficult to burn than regular fuel gas and the high content of carbon monoxide can interfere with the stability of the flame. As the reformer operating temperature is increased, the reforming equilibrium shifts, resulting in more hydrogen and less methane in the reformer outlet and hence less methane in the tail gas.

Membrane systems separate gases by taking advantage of the difference in rates of diffusion through membranes (Brüschke, 1995, 2003). Gases that diffuse faster (including hydrogen) become the permeate stream and are available at low pressure, whereas the slower-diffusing gases become the nonpermeate and leave the unit at a pressure close to the pressure of the feedstock at the entry point. Membrane systems contain no moving parts or switch valves and have potentially very high reliability. The major threat is from components in the gas (such as aromatics) that attack the membranes, or from liquids, which plug them.

Membranes are fabricated in relatively small modules; for larger capacity, more modules are added. Cost is therefore virtually linear with capacity, making them more competitive at lower capacities. The design of membrane systems involves a tradeoff between pressure drop (or diffusion rate) and surface area, as well as between product purity and recovery. As the surface area is increased, the recovery of fast components increases; however, more of the slow components are recovered, which lowers the purity.

Cryogenic separation units operate by cooling the gas and condensing some, or all, of the constituents for the gas stream. Depending on the product purity required, separation may involve flashing or distillation. Cryogenic units offer the advantage of being able to separate a variety of products from a single-feed stream. One specific example is the separation of light olefins from a hydrogen stream.

Hydrogen recovery is in the range of 95% v/v, with purity above 98% v/v obtainable. In addition to the general description of the purification processes presented earlier, four of the major process techniques for achieving this level of purity are:

(i) Cryogenics plus methanation, which utilizes a cryogenic process whereby carbon monoxide is liquefied in a number of steps until hydrogen with a purity of on the order of 98% is produced. The condensed carbon monoxide, which would contain methane, is then distilled to produce pure carbon monoxide and a mixture of carbon monoxide and methane. The hydrogen stream is taken to a shift converter where the remaining carbon monoxide is converted to carbon dioxide and hydrogen. The carbon dioxide is removed and any further carbon monoxide or carbon dioxide can be removed by methanation. The resulting hydrogen stream typically has purity on the order of 99.7% v/v.

(ii) Cryogenics plus pressure swing adsorption (PSA), which utilizes the similar sequential liquefaction of carbon monoxide until hydrogen having 98% purity is obtained. Again, the carbon monoxide stream can be further distilled to remove methane until it is essentially pure. The hydrogen stream is then allowed to go through multiple pressure swing adsorption cycles until the hydrogen purity is as high as 99.999% v/v.

(iii) Methane-wash cryogenic process utilizes the principle of carbon monoxide absorption in a liquid methane stream so that the hydrogen stream produced contains only low levels (on the order of parts per million) of carbon monoxide but about 5 to 8% v/v methane. Hence, the hydrogen stream may have purity on the order of only 95% v/v. However, the liquid carbon monoxide/methane stream can be distilled to produce a pure carbon monoxide stream and a carbon monoxide/methane stream, which can be used as fuel.

(iv) COsorb process utilizes copper ions (cuprous aluminum chloride, CuAlCl₄) in toluene to form a chemical complex with the carbon monoxide to separate it from hydrogen, nitrogen, carbon dioxide, and methane. This process can capture about 96% of the available carbon monoxide to produce a stream of greater than 99% purity. On the other hand, water, hydrogen sulfide, and other trace chemicals can poison the copper catalyst and must be removed prior to the reactor.

Although the efficiency of cryogenic separation decreases with a content of low carbon monoxide in the feed, the COsorb process is able to process gas streams with low carbon monoxide content much more efficiently.

6.4.2 Oil-water separation

The typical oil-water separation process occurs in a device designed to separate gross amounts of oil and suspended solids from the effluents of petroleum and gas processing. The most common type of separator is the API separator, which is a gravity separation device designed by using the specific gravity difference between the oil and water (depending on the pressure, the gas remains in the volatile state) while the oil and water separate from stream as liquids (Mokhatab et al., 2006; Speight, 2007). Based on that design criterion, any suspended solids settle to the bottom of the separator as a sediment layer, the oil will rise to the top of the separator, and the wastewater will be the middle layer between the oil on the top and the solids on the bottom.

Typically, the oil layer is skimmed off and subsequently reprocessed or disposed of, and the bottom sediment layer is removed by a chain and flight scraper (or similar device) and a sludge pump. The water layer is sent to further treatment consisting usually of a dissolved air flotation unit for further removal of any residual oil and then to some type of biological treatment unit for removal of undesirable dissolved chemical compounds.

Parallel plate separators are similar to API separators but include tilted parallel plate assemblies, and the underside of each parallel plate provides more surface for suspended oil droplets to coalesce into larger globules. Any sediment slides down the topside of each parallel plate. Such separators still depend on the specific gravity between the suspended oil and the water. However, the parallel plates enhance the degree of oil-water separation. The result is that a parallel plate separator requires significantly less space than a conventional API separator to achieve the same degree of separation.