Desktop version

Home arrow Engineering

  • Increase font
  • Decrease font


<<   CONTENTS   >>

: Natural Gas Reforming to Industrial Gas and Chemicals Using Chemical Looping

Dawei Wang, Titao Zhang, Fanhe Kong, L-S Fan and Andrew Tong[1] [2]

Introduction

Energy production via combustion of carbon-based fossil fuels is the major source for greenhouse gas emissions (International Energy Outlook, 2011). Compared to coal, natural gas emits 50 to 60% less carbon dioxide (CO,) and, thus, has a much lower life cycle greenhouse gas emissions (NETL, 2010). The vast availability of natural gas resources in the United States har e become economically accessible due to advances in production drilling technology. The U.S. Energy Information Administration has projected that the global natural gas consumption will rise by 50% between 2010 and 2035 (International Energy Outlook, 2011).

With the growing abundance of natural gas, research has accelerated to develop efficient methods to utilize natural gas to build long carbon chain chemical and liquid fiielproducts as opposed to the conventional approach of refining and cracking heavy hydrocarbons, such as crude oil. While technologies for directly converting methane (CH,), the main component in natural gas. to chemicals are still under development, industrial chemical production processes rely on a two-step method, converting CH4 to syngas, a mixture of mainly CO and H, with various ratios, and utilizing syngas for chemical synthesis and production. Producing syngas from CH, at an industrial scale is a mature technology, mainly through steam-methane reforming (SMR) and autothennal reforming (ATR). Conventional SMR plants operate at pressures between 14 and 40 atm with outlet temperatures in the range of 815 to 925 °C. Though commercially demonstrated, SMR is an energy intensive process due to the endothermic nature of its reaction and, thus, external heat sources are required for continuously stable operation, resulting in substantial carbon emissions. For conventional ATR, the energy intensive air separation unit greatly reduces the overall plant efficiency, increasing carbon emissions in the plant. For both processes, necessary down-stream processing steps are also required in order to adjust the quality of syngas, such as the H,:CO molar ratio and syngas purity, in order to meet the requirements of various downstream processes. The capital cost of SMR plants is also prohibitive for small to medium size applications. Recently, catalytic partial oxidation of methane (CPOM), where CH, is reacted with O, only to produce syngas over a noble metal catalyst in a simple, one-step reaction scheme, has been treated as a promising alternative and received much attention (Hickman and Schmidt, 1993: Neumann et al.. 2004: Ashcroft et al., 1990; Tsang et ah, 1995). Owing to its operating temperature of over 1000 °C, the CPOM process has a high reaction rate, resulting in an extremely short residence time for the reactants (Hickman and Schmidt, 1993). CPOM offers several advantages over SMR, such as the exothermic nature of the reactions which allows for autothennal operation and, thus, lowers the carbon footprint of syngas production, and the compact reactor design with very high space-time yields owing to its very high reaction rate. However, several issues have limited its commercialization, including the safety concerns related to direct contact of fuel and oxygen at a temperature close to their upper flammability limit, the requirement of an air separation unit (ASU) to produce purified oxygen, and the high cost of producing catalysts from expensive noble metal.

The chemical looping concept provides possible solutions to directly address the issues for the CPOM processes. The spatial separation of the fuel conversion into two or more separate steps avoids any direct contact of feedstock and gaseous oxygen and, thus, alleviates safety concerns. In addition, using metal oxide as oxygen earner eliminates the need for a cryogenic ASU, substantially reducing the plant's parasitic power demand.

Chemical looping was first practiced in the late nineteenth century when Franz Bergmann filed a German patent for producing calcium carbide (CaC,) via reaction between manganese (IV) oxide in the presence of a hydrocarbon fuel and calcium oxide (CaO) (Bergmann, 1897). In 1903, the Lane hydrogen producer, or Lane process, was invented for producing hydrogen in a fixed bed reactor system where syngas was used to reduce iron oxide ores and steam was introduced to produce hydrogen from the reduced non ore from the steam-iron reaction (Hurst, 1939; Gasior, 1961; Teed, 1919). The incomplete conversion of syngas and low recyclability of the iron ores prevented the Lane process from being economically competitive against the increased availability of oil and natural gas in the 1940s. The Lane process was eventually phased out at this time. In the 1950s, Warren Lewis and Edwin Gilliland patented a chemical looping process using non and copper oxides as the metal oxide to produce CO, as the desired product in a countercurrent gas-solids contacting pattern for use in the beverage industry (Lewis and Edwin, 1954). In the 1960s and 1970s, the Consolidation Coal Company (now CONSOL Energy) developed a pilot- scaled CO, acceptor process to produce substitute natural gas (SNG) (Dobbyn et al., 1978). CaS-CaS04 redox cycle observed in the CO, acceptor process was different fr om the typical redox cycles based on metal/metal oxide. In the 1970s, the Institute of Gas Technology (IGT) developed the HYGAS process to produce SNG from coal via the methanation reaction. The required hydrogen for the methauation reaction came fr om the steam-iron reaction, where non (Fe) was obtained fr om iron oxide reduction by syngas produced from coal gasification (U.S. Department of Energy. 1979). Two-stage countercurrent fluidized bed reactor design was used for both the reducer and the oxidizer in order to enhance the fuel gas and non oxide conversions, as well as heat and mass transfer. The HYGAS process was demonstrated at pilot-scale but not commercialized. In the 1980s, Atlantic Richfield Company (ARCO) developed a gas to gasoline (GTG) process using a reactor configuration consisting of a two circulating fluidized beds reactor configuration (Jones et ah, 1987; Sofranko et ah. 1987). The process was designed to convert CH4 to an ethylene-rich intermediate via oxidative coupling of methane for gasoline production using Mobil’s olefin to gasoline and distillate (MOGD) process. Catalytic metal oxide for the oxidative coupling of methane (OCM) reaction was reduced in one reactor while reacting with CH4. and were transported to the other reactor for regeneration. A pilot-scale demonstration of the process was operated. However, the process was discontinued towards the end of 1980s as the crude oil price decreased below the natural gas price. In the 1990s, DuPont developed a process, referred to as the DuPont process, to produce maleic anhydride through the selective oxidation of butane. A vanadium phosphorus oxide (YPO) catalytic metal oxide was used. The selective oxidation for butane conversion to maleic anhydride took place in a lean phase riser where a multifunctional TO metal oxide provided catalytic activity as well as lattice oxygen. Meanwhile, the regeneration reaction of reduced YPO catalytic metal oxide by air took place in a fluidized bed reactor. This process was scaled to commercial demonstration but failed as the VPO catalytic metal oxide was not reactive enough to provide the oxygen and particle integrity was compromised (Dudukovic, 2009; Evanko et al., 2013). A fixed bed reactor system for partial oxidation of CH4 to syngas using CeO, was also attempted in the 1990s. Using CeO, in a chemical looping process to convert CH4 to syngas is thermodynamically favorable, but experimental results showed extensive carbon deposition under high temperanire conditions, which negatively affected the CeO, reactivity, and reduced the reaction kinetics between CH4 and CeO,, resulting in a low CH4 conversion (4% per pass) (Otsuka et al., 1998a). In the 2000s, a chemical looping process to provide the heat necessary to operate conventional steam methane reforming (SMR) reactions using metal oxide oxygen carriers while mitigating carbon emissions from the SMR furnace was developed (Ryden and Lyngfelt, 2006; Adanez et al., 2012; Pans et al., 2013). The SMR tubular reactor is placed inside either the reducer or combustor reactor to serve as the SMR furnace so as to allow the chemical looping reactors to provide the necessaiy heat for the endothermic SMR reaction. In this approach, the CO, produced from the furnace is captured without the need of a post combustion CO, capture unit to mitigate carbon emissions. This process requires the diversion of natural gas in an equal or greater quantity than the conventional approach to fulfill the endothermic heat requirements for the SMR reactions.

Chemical looping processes using a metal-based oxygen earner to perform redox reaction with a carbon-based fuel can be categorized into 2 types of systems: Chemical looping combustion (CLC) for power generation with CO, capture and chemical looping reforming (CLR) for chemical and industrial gas production. The Lewis and Gilliland process represents a CLC system where the carbon fuel is fully oxidized to CO,. CLR systems can be further divided based on the product from the CLR reactor. The Lane Producer and HYGAS processes represent CLR processes for H, production. The DuPont VPO and ARCO processes each represent CLR for selective oxidation systems where the metal oxide oxygen carrier serves to selectively oxidize the reactants to a desired product. CLR for syngas production systems represent chemical looping processes where a carbon-based gaseous fuel, such as natural gas, is partially oxidized to syngas (Fan et al., 2015; Luo et al., 2014; Ryden et al., 2008; Nalbandian et al., 2011; Dai et al., 2006). CLR for selective oxidation systems rely on a multifunctional metal oxide, which possesses catalytic and oxygen transfer properties, to selectively convert hydrocarbon feedstock to chemicals. In the reducer reactor, a catalytic metal oxide reacts with a hydrocarbon feedstock to selectively produce chemicals and reduced catalytic metal oxide. In the combustor reactor, the reduced catalytic metal oxide is regenerated by oxidation with air (Keller and Bhasin, 1982; Contractor, 1999). By directly producing chemicals in the reducer, the generation of syngas as an intermediate is not necessary. A simplified flow diagram of CLR is shown in Figure 1.

Metal oxide reaction engineering and particle science and technology are two key technical areas for the development of chemical looping concepts for combustion, gasification, or reforming applications. Understanding these metal oxide issues allows the metal oxide materials to be formulated effectively, synthesized, and used in a sustainable manner for desired chemical looping reaction applications.

A variety of metal oxides have been investigated for chemical looping applications (Fan, 2010; Messerschmitt et al., 1915; Lane, 1913; Ishida et al., 1987). The preliminary screening of the oxygen carrier is based on its thermodynamic properties as illustrated using the modified Ellingham diagram, as shown in Figure 2(left). The modified Ellingham diagram is based on the Gibbs free energy of reactions and its variation with temperature for metal oxides. The diagram can be divided into four sections, namely, combustion section (A), syngas production section (B), carbon deposition section (D) and inert section (C), based on the four fundamental reactions for carbon and hydrogen conversions, as is shown in Figure 2(right). Between reaction lines 1 and 2, a very small section (E) exists, which also produces syngas, however, the syngas will contain a significant amount of H,0 byproduct. Using Figure 2, one can identify metal oxides ideal for CLC systems, i.e., metals that lie in region A, and metal oxides ideal for

Process flow diagram for chemical looping reforming

Figure 1. Process flow diagram for chemical looping reforming.

Modified Ellingham diagram to determine metal oxide performance as oxygen carriers

Figure 2. Modified Ellingham diagram to determine metal oxide performance as oxygen carriers.

CLR for syngas production systems, i.e.. metals that lie in region B. Based on these sections, the metal oxides are identified as potential oxygen carriers for different chemical looping processes.

Metal oxides used in chemical looping combustion applications (i.e., full conversion of carbon fuels) include NiO, CoO, CuO. Fe,0,. Metal oxides lying in region В for CLR for syngas production, such as CeO,, have mild oxidation properties to prevent over-conversion of the carbon-based fuel to CO, and H,0. Metal oxides in carbon deposition and inert sections, such as Cr,0, and SiO„ lack the potential to be used as oxygen carriers and are generally considered to be inert and unfavorable. However, they are good candidates for support materials for active oxygen earner materials, to enhance some physical properties. For example, when TiO„ which lies in the inert region C in Figure 2, is combined with FeO, it will form a FeTiO , complex, which will generate a higher quality of syngas than FeO alone (Li et al.. 2011). Thus, the addition of a support material in this case improves the oxygen earner performance. On the other hand, when considering CLC systems, the addition of supports may hinder the oxidation properties of certain metals alone. Examples are CuO-based oxygen carriers for CLC which lose oxidizing potential when Al,0, is used as a support material due to the formation of CuA1,04 (Arjmand et al.. 2011). It becomes essential to understand how different phases behave in the presence of each other during the development of high performing oxygen carriers. CH4 is thermodynamically unstable at temperatures higher than 750 °C and spontaneously decomposes to form C and H, in the absence of an oxygen source.

It should be noted that the modified Ellingham diagram represents only a thermodynamic analysis of metal oxides to be used as potential oxygen carrier materials. In addition to thermodynamics, many other factors come into play when selecting oxygen carrier materials. With the base metal oxides identified, then physical and chemical properties can then be further characterized. For commercial applicability, the oxygen carrier should possess several properties, including redox reactivity, long-term stability, physical strength, toxicity, and appropriate production cost (Luo et al., 2015). Many oxygen carrier materials har e been studied for CLR syngas production applications (Fan, 2015; Li et al., 2009; Luo et al., 2015). During the early development of these applications, single metal oxides or sulfates were considered to be the active components in oxygen earner materials. Recent research has been directed towards the use of binary and ternary metal composite materials for improved process performance (Luo et al., 2015).

  • [1] William G. Lovne Department of Chemical and Biomolecular Eugineermg, The Ohio State University, USA 43210.
  • [2] Corresponding author: This email address is being protected from spam bots, you need Javascript enabled to view it
 
<<   CONTENTS   >>

Related topics