Energy Conversion and Management 96 (2015) 392–402

Contents lists available at ScienceDirect

Energy Conversion and Management

journal homepage: www.elsevier .com/locate /enconman

Equilibrium thermodynamic analyses of methanol productionvia a novel Chemical Looping Carbon Arrestor process

http://dx.doi.org/10.1016/j.enconman.2015.03.0080196-8904/� 2015 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +61 2 4985 4411; fax: +61 2 4921 6893.E-mail address: Behdad.Moghtaderi@newcastle.edu.au (B. Moghtaderi).

Cheng Zhou, Kalpit Shah, Elham Doroodchi, Behdad Moghtaderi ⇑Priority Centre for Frontier Energy Technologies and Utilization, Discipline of Chemical Engineering, School of Engineering, Faculty of Engineering and Built Environment,The University of Newcastle, Callaghan, NSW 2308, Australia

a r t i c l e i n f o a b s t r a c t

Article history:Received 30 December 2014Accepted 2 March 2015

Keywords:MethanolChemical loopingCarbon arrestorHydrogenEquilibrium thermodynamic

Methanol economy is considered as an alternative to hydrogen economy due to the better handling andstorage characteristics of methanol fuel than liquid hydrogen. This paper is concerned about a compre-hensive equilibrium thermodynamic analysis carried out on methanol production via an innovativeChemical Looping Carbon Arrestor/Reforming process being developed at the University of Newcastlein order to reduce both energy consumption and carbon emissions. The detailed simulation revealedthermodynamic limitations within the Chemical Looping Carbon Reforming process however on theother hand it also confirmed that the new concept is a low energy requirement and low emission optioncompared to other methanol production technologies. Specifically, the mass and energy balance studyshowed that the Chemical Looping Carbon Reforming process typically consumes approximately 0.76–0.77 mole methane, 0.25–0.27 mole carbon dioxide, 0.49–0.50 mole water, and 0.51 mole iron oxide(in a chemical looping manner) per mole of methanol production. Moreover, the energy efficiency ofChemical Looping Carbon Reforming process was found to be �64–70% and its emission profile wasfound as low as 0.14 mole carbon dioxide per mole of methanol, which is about 82–88% less than the con-ventional methanol production process and well below the emission levels of other emerging methanolproduction technologies.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Hydrogen is considered as the key element of the sustainableenergy scenarios of the future. Hydrogen is a clean fuel, has a veryhigh energy density (�120 MJ/kg), and can be utilised efficiently infuel-cells to produce heat and power. However, success of hydro-gen based energy systems/hydrogen economy greatly depends onthe availability of suitable methods for efficient production andstorage of hydrogen [1]. Safe storage of hydrogen remains an unre-solved technical challenge and as such has been the subject ofnumerous studies [2].

Techniques for storing hydrogen include compression, liquefac-tion, physio-sorption (e.g. in carbon nano-tubes), metallichydrides, and complex hydrides. However, all of them suffer froma combination of high cost, high energy demand, technical diffi-culty and inefficiency [1]. There are, however, more suitable meth-ods of hydrogen storage which have been overlooked, for example,the use of hydrocarbon-based H2 carries – methanol (CH3OH). The

use of methanol as energy carrier offers an alternative to hydrogeneconomy – methanol economy.

Methanol has a relatively high hydrogen storage capacity andpossesses higher energy density than that of hydrogen. At roomtemperatures and pressures methanol is in its liquid form and,hence, can readily be stored and transported, making it an excel-lent transportation fuel. Methanol can also be blended with gaso-line (e.g. M15 and M85), and for its application in automobilesector only limited modifications to the existing gasoline basedengine systems are required [3].

The application of methanol in transportation sector is espe-cially important for Australia, where oil security becomes a keyissue due to its relative isolation and reliance on transportationfuels. The growing import dependence also leaves Australia morevulnerable to potential disruptions in overseas crude oil and petro-leum products supply chains. According to a recent IEA report, net-oil imports for Australia have increased from 12,000 bpd in 2000 to519,000 bpd in 2011, the highest on record [4]. In contrast, in 2011Australia’s oil production stood at 484,000 barrels per day (bpd)which is a drop of 41% from the peak of 819,000 bpd in 2000 [4].Such a rapid transformation in the structure of country’s oil sup-ply–demand is quite alarming.

Nomenclature

CLCR Chemical Looping Carbon Arrestor/ReformingCRM, SRM CO2/steam reforming of methaneEp the energy content of the product (i.e. methanol)Ef the energy in the feedstockEa auxiliary energy requirement in the processGHG greenhouse gassesDHc

� the low heat of combustion of fuelsmCO2 CO2 emissionmCH3OH methanol production

SR syngas (molar) ratioSR0 alternative/revised syngas (molar) ratio

Subscriptp producta auxiliaryf feedstock

C. Zhou et al. / Energy Conversion and Management 96 (2015) 392–402 393

Methanol blended gasoline may offer an excellent solution.Methanol is a clean burning, high octane blending component pro-cessed from alternative non-petroleum energy sources such asnatural gas, coal and biomass [3]. It has been commerciallyblended into gasoline at various times and locations since 1980in United States, China, New Zealand and parts of the Europe [5].Because carburetted fuel systems were most prevalent in the roadvehicle fleet at that time and such vehicles had limited ability tohandle high oxygen levels in the fuel, methanol concentrationswere generally limited to 3–5 volume percent of the gasoline blend[6]. However, with today’s modern pressurised fuel injector sys-tems with computerised feedback control loops, blends with ashigh as 15 volume percent methanol (M15) can now be success-fully used in the modern vehicles that are on the road today [7].

The major impediment in realising methanol economy, and inparticular, the large-scale deployment of methanol-gasolineblends, is the high energy costs associated with generating hydro-gen/syngas needed to synthesise methanol. Conventional steamreforming (SRM), dry/CO2 reforming, partial oxidation, oxy reform-ing and auto-thermal reforming of methane, as well as coal/bio-mass gasification are the most common processes used for theproduction of syngas for methanol production [8]. Among them,SRM is the predominant technology used at present for methanolproduction and accounts for approximately 75% of global methanolproduction [9]. In general, the energy required to drive the syngasproduction using the above-mentioned conventional processes isquite high and typically constitutes about 50% of the total energydemand of the methanol production process [10]. This invariablyresults in high GHG emissions during the production phase. Inaddition to large energy and GHG footprints, significant opera-tional/production costs and considerable capital investmentrequired for conventional syngas production processes are amongthe major barriers against their widespread deployment for metha-nol production.

A number of new syngas production methods have been pro-posed in recent years, including the use of plasma and catalytictype reactors. Yao et al. [11] examined the approach of methaneselective oxidation to syngas and finally methanol fuel usingnon-thermal pulsed plasma. Aasberg-Petersen et al. [12] reviewedthe catalytic conversion of natural gas into syngas using variouscatalysts and catalytic processes. Yet, these methods have largelyfailed to make any major attraction primarily because of the lowyields and selectivity of the product streams, as well as complexhardware and hence high capital costs. It is, therefore, time to con-sider an entirely new strategy that firstly meets the essentialrequirement of inexpensive capital, operation and productioncosts, while offering the prospect of higher product yields, lowerenergy demands and smaller GHG footprint.

As a respond to the above technology bottleneck, an alternativemethanol synthesis process based on chemical looping conceptwas proposed and developed at the University of Newcastle,Australia. It is also part of a larger programme being carried out

at the University of Newcastle in which chemical looping concepthas been applied to a number of areas other than combustion.Moghtaderi et al. [13] applied chemical looping concept for oxygenproduction. The working principle behind the chemical looping airseparation (CLAS) process for oxygen production involves the cyc-lic oxidation and reduction reactions of metallic oxide particles intwo separate reactors as a means of separating oxygen from air[14]. The process can be used for either high purity oxygen produc-tion [15] or can be integrated to oxy-fuel combustion and gasifica-tion plant to suffice their need for oxygen by eliminating nitrogenfrom air stream in the air reactor [16]. In the advanced version ofthe CLAS-oxy-fuel integration proposed by Shah et al. [17], effortshave been made to provide coal-natural gas based hybrid powerplant called chemical looping oxy-combustor (CLOC). Moghtaderiet al. [18] also applied chemical looping concept for biomassgasification using construction demolition waste material. The pro-cess can produce hydrogen rich synthesis gas by capturing CO2 viaconstruction demolition waste material. Also, the research teamhas successfully applied chemical looping concept for ventilationair methane abatement. The two processes being developed in thisarea are called chemical looping VAM abatement unit [19] andstone dust looping VAM abatement unit [20]. The former one uti-lises metal oxide oxygen carriers while the later one uses stonedust (limestone). The novel chemical looping based ChemicalLooping Carbon Reforming (CLCR) process proposed in this workutilises natural gas and CO2 as the feedstock and can produce syn-thesis gas with a CO/H2 ratio suitable for methanol synthesis and/or other Fischer–Tropsch processes.

The novelty of the CLCR process lies in its ability to minimisecarbon losses throughout the process by converting CO2 into inter-mediates (CO and carbon) then methanol via a chemical loopingroute followed by carbon gasification. The CLCR process featurescarbon gasification for achieving higher yields and greater util-isation of CO2 (i.e. if deployed at commercial scales the CLCR pro-cess can offer a large sink for CO2 sequestration). In this way,methanol may be produced with less energy requirement andtheoretically negative CO2 emission. The CLCR process also elimi-nates the need for an expensive air separation unit (i.e. oxygenplant) as required in other conventional processes. Apart fromlower energy requirement and greater utilisation of CO2, CLCR pro-cess employs a compact multi-loop-seal modular design to ensurebetter oxygen carrier circulation, more efficient heat transfer and asmaller process plant footprint.

Use of natural gas as the feedstock for CLCR process is also con-sidered quite suitable for Australia where natural gas resources areabundant. If CO2 is sourced from ambient, the process may becomeeven more attractive by mimicking nature’s photosynthesis pro-cess allowing for the chemical recycling of global CO2 emission.To the best of our knowledge, no equilibrium thermodynamicstudy or process simulation has been performed for CLCR process.The only relevant work is the one published by Zeman and Castaldi[9] who solely performed the calculations for a process similar to

394 C. Zhou et al. / Energy Conversion and Management 96 (2015) 392–402

CLCR assuming that all reactions are complete. The calculations intheir work are overly simplified and did not consider equilibriumthermodynamic analyses as well as identification of auxiliarypower requirements.

The objective of this paper is to evaluate the thermodynamicfeasibility and performance of CLCR concept, and provide itsdetailed comparison with other methanol production processes.

2. Methodology

2.1. Process simulation

Fig. 1 shows the schematic diagram of the CLCR process. Asshown the hardware comprises a reactor for CO2 reforming ofmethane (CRM) and a modular carbon reforming reactor consistingof four compartments. The CLCR is a cyclic process and consists ofthe following six steps:

Step-I (CRM): Methane is reacted with CO2 in the CRM reactoraccording to reaction R1 producing a syngas mixture with aH2/CO ratio of 1:1.Step-II (Oxidation/Carbon deposition): The syngas then entersCompartment 3 of the oxidation reactor wherein CO decom-poses as carbon on the surface of metal oxide oxygen carriersand a hydrogen rich product is formed (R2). During this stagethe oxygen carriers are oxidised to a higher oxidation state asshown in R2.Step-III (Carbon gasification): The metal oxides are transferredfrom Compartment 3 to 4 where the carbon deposit on the sur-face of metal oxide particles is gasified by steam (H2O), in turn,producing more syngas with H2/CO ratio of 1:1 (R3).Step-IV (Further oxidation – optional): The metal oxides aretransferred from Compartment 4 to 1 for further oxidisationusing normal air when necessary (see R4). This would takethe metal oxides to their highest oxidation state.Step-V (Reduction): The fully oxidised metal oxides are movedto Compartment 2 for oxidation of methane. The products ofthis reaction are reduced metal oxides, H2, CO and/or CO2

depending on whether methane is partially (R5-1) or fully oxi-dised (R5-2).Step-VI (Synthesis): The reduced metal oxides are transferred toCompartment 3 to repeat the cycle. The product streams fromCompartments 3, 4 and 2 are mixed to produce a syngas streamwith a H2/CO ratio at 2:1, which is then transferred to methanolsynthesis reactor.

CH4ðgÞ þ CO2ðgÞ $ 2COðgÞ þ 2H2ðgÞ ðR1Þ

COðgÞ þMexOyðsÞ $ CðsÞ þMexOyþ1ðsÞ ðR2Þ

CðsÞ þH2OðgÞ $ COðgÞ þH2ðgÞ ðR3Þ

MexOyþ1ðsÞ þ 0:5O2ðgÞ $MexOyþ2ðsÞ ðR4Þ

2CH4ðgÞ þMexOyþ2ðsÞ $ 2COðgÞ þ 4H2ðgÞ þMexOyðsÞ ðR5-1Þ

CH4ðgÞ þMexOyþ2ðsÞ $ CO2ðgÞ þ 2H2ðgÞ þMexOyðsÞ ðR5-2Þ

Process simulation was carried out using the process simulationpackage – Aspen Plus v-7.3 [21]. The aspen model solves all theequilibrium constant equations simultaneously and calculates theequilibrium conditions by minimising the total Gibbs free energyof the system. Moreover, thermodynamic database – HSC chemistry(developed by Outotec) – was also used as an assisting tool. Withthe above tools, energy and mass balance analyses were performedfor CLCR process using Iron Oxides as the oxygen carriers, and

comparisons have been made against conventional SRM and emerg-ing methanol production plants using either methane, coal, or bio-mass as the feedstock.

Fig. 2 shows the Aspen simulation flowsheet of the wholemethanol production process, in which the process is divided intotwo sub-process loops: the chemical looping process and methanolsynthesis process. In CLCR process, methane, carbon dioxide, andwater are used as the feedstock for the chemical looping side reac-tions, which is able to produce syngas with an optimised H2/COratio for methanol synthesis. Waste heat recycling was also fullyimplemented. The syngas product, after removing excess water,is directed to the methanol synthesis process where methanolwas generated and purified. The chemical looping process (seeFig. 3) follows the same principles described in Section 2.1,whereas methanol synthesis process (see Fig. 4) involves mainlythe following four steps: (i) water removal process where excesswater is removed from the syngas product using a user block con-sisting of a condenser and a separator, (ii) gas compression processin which the inlet syngas is compressed to a pressure of 50 bar viaa two-stage inter-cooled compressor, (iii) methanol synthesis reac-tion taking place at 50 bar and 230 �C, and (iv) separation and pur-ification processes where one separator and two distillationcolumns are simulated in order to separate the product from gasesand remove impurities contained in the crude methanol.

2.2. Performance indicators

Several key performance indicators were defined to evaluatethe performance of CLCR process, which includes syngas ratio,energy efficiency, and emission profile.

2.2.1. Syngas ratioFor an optimum methanol synthesis reaction, the ideal syngas

product is expected to have a H2/CO molar ratio (SR) of 2:1, namely

SR ¼H2

CO¼ 2:0 ð1Þ

However, in a conventional methanol production process CO2

usually presents in the syngas product along with CO and H2. Inthis case, an optimum syngas product should be characterised bythe stoichiometric number given below, according to Lurgi’smethanol synthesis process [22],

S0R ¼H2 � CO2

COþ CO2¼ 2:0� 2:1 ð2Þ

The above equation is actually deduced from the stoichiometricnumbers of the feed gases participating in the methanol synthesisthrough the following two reactions,

COþ 2H2 ! CH3OH ðR6Þ

CO2 þ 3H2 ! CH3OHþH2O ðR7Þ

Some literature suggests that CO hydrogenation (R6) is the pri-mary reaction that generates methanol and CO2 hydrogenation(R7) should be avoided since some hydrogen is wasted in the formof water [23]. However, other literature suggests that someamount of CO2 presented in the feed gas is believed to benefitthe overall process and help maintain the catalyst activity [24].On the contrary, Coteron and Hayhurst’s study [25] indicates thatthe main methanol synthesis reaction is CO2 hydrogenation withCO acting as an intermediate that removes oxygen absorbed onthe catalyst surface. It can be seen that the methanol synthesiskinetics and reaction mechanism is still not conclusive in theliterature, and general rules on the contribution of CO and CO2 inproducing methanol is not available [26].

Fig. 1. Schematic diagram of Chemical Looping Carbon Reforming process.

C. Zhou et al. / Energy Conversion and Management 96 (2015) 392–402 395

This thus leaves certain degree of freedom in simulating theCLCR process to achieve an optimum syngas ratio, where the gen-erated syngas may contain a wide range of H2, CO, CO2, and H2O. Ifa significant amount of CO2 is produced, CO2 can either participatein the methanol synthesis reaction via R7 or be completelyrecycled via a gas separation/removal procedure using physicaland/or chemical processes (e.g. chemical absorption, physicalabsorption, membranes, and distillation). The water content inthe synthesis gas product can also be removed before and/or aftermethanol synthesis via a condenser or distillation unit. The ulti-mate goal, however, is still to generate a syngas with a syngas ratiopertinent to that shown in Eq. (2).

2.2.2. Energy efficiencyEnergy efficiency is an important performance indicator since it

reflects how well the system performs in converting feedstock intouseful product. Here, the energy efficiency is defined as the energy

content of the product (i.e. methanol), Ep, divided by the energy inthe feedstock, Ef, and other auxilliary energy requirement in theprocess, Ea (Eq. (3)) [27]. The energy contained in the feedstockand product is normally expressed as the low heat of combustionof those fuels, DHc

� (analogous to the low heating value of coal),whilst the auxiliary inputs are given as the thermal equivalent ofelectric power used.

g ¼ Ep=ðEf þ EaÞ ¼ DH�

c;p=ðDH�

c;f þ DH�

c;aÞ ð3Þ

2.2.3. Emission profileAnother critical performance indicator of any chemical pro-

cesses is the associated emission profile/intensity – primarily CO2

emission. The emission profile of CLCR process is calculated asthe sum of CO2 emission associated with the feedstock (mCO2,f),main process (mCO2,p), and auxiliaries (mCO2,a), divided by methanol

Fig. 2. Aspen process model for CLCR process (black and red lines are mass and energy streams, respectively. R: recycled streams, +Q: unit operations in which heat isreleased/needs to be removed, �Q: units that requires heat input, LTHR: for low-temperature heat recovery).

Fig. 3. Aspen process model for the chemical looping block in CLCR process.

396 C. Zhou et al. / Energy Conversion and Management 96 (2015) 392–402

production. We assume that the auxiliary energy supply for CLCRprocess is met by natural gas combustion.

Emission profile=intensity ¼ ðmCO2 ;f þmCO2 ;p þmCO2 ;aÞ=mCH3OH ð4Þ

3. Results and discussion

3.1. Thermodynamics considerations

The simulation of CLCR process indicated that the feasibility of aCLCR plant is greatly subjected to the thermodynamics of relatedreaction sets, not to mention the kinetics behind each reaction.However, it does not necessarily rule out the possibility of a suc-cessful CLCR process. A critical thermodynamics consideration

when one builds a CLCR plant should be given to the carbongasification reactor (i.e. Step-III) – one of the key hydrogen genera-tion steps. This reactor was found to have thermodynamic conflictwith the previous oxidation reactor (i.e. Step-II Oxidation/Carbondeposition). This conflict is detailed in the following texts.

In the oxidation reactor (OXI in Fig. 3) the oxidation of FeO byCO and the simultaneous conversion of CO into carbon arethermodynamically favourable at low-medium temperatures(<700 �C), above which the oxidation reaction is prohibited. Asper the Gibbs energy of the reaction, the lower temperature below700 �C the reactor operates, the greater extent the oxidation couldachieve. This is, however, contradicted to the preferred highoperating temperature of the subsequent carbon gasification reac-tion, which commonly takes place between 400 �C and 700 �C witha higher temperature yielding a greater carbon conversion rate.

Fig. 4. Aspen process model for the methanol synthesis block in CLCR process.

C. Zhou et al. / Energy Conversion and Management 96 (2015) 392–402 397

The problem here is the difficulty associated with the sep-aration of deposit carbon from Fe3O4, which leads to the presenceof both carbon and Fe3O4 in carbon gasification reaction. To pre-vent any unwanted reversed oxidation reaction from occurring inthe carbon gasification reactor, the carbon gasification reactor willbe limited to operate at a significantly lower temperature (e.g.450 �C) than its ideal operating temperature, leading to a low car-bon conversion rate. For a complete carbon conversion which isessential for producing hydrogen, a two-stage carbon gasificationreactor needs to be implemented (see Fig. 3). The two-stage designoperates in a way such that the unconverted carbon together withFe3O4 left in the first-stage carbon gasification reaction is separatedvia a gas–solid separator, which is then directed into the second-stage carbon gasification for a complete carbon conversion. Theabove process design solves the conflict between the two conjointreactors, but leaves a narrow operating temperature window(about 400–600 �C) for the carbon gasification step to play inCLCR process.

Moreover, since the proposed CLCR process is a new idea andhas not been demonstrated at any scales, model validation usingactual plant data is not applicable. Nevertheless, a comprehensivethermodynamic analysis was performed to compare the perfor-mance indicators of CLCR process against other conventional andemerging methanol production processes including conventionalSRM, CRM, CO2 hydrogenation, and methanol production usingcoal/biomass as the feedstock.

1 Varied inlet gas ratios will be examined as part of the sensitivity and economicanalyses of our future publications.

3.2. Typical operating conditions

A series of typical operating conditions for CLCR process werespecified considering the desired components of products andequilibrium thermodynamic of all reactions. Table 1 lists the typi-cal conditions for the key unit operation blocks used in the AspenPlus model for CLCR process. These typical conditions are deter-mined after considering the favourable operating conditions ofall reactions and any thermodynamic conflictions between them.For example, Table 1 shows the approximate operating tempera-ture ranges for all reactors based on the limitations in equilibrium

thermodynamic of each reaction. With the specified operating con-ditions, mass and energy balance analyses of the CLCR processwere carried out.

3.3. Feedstock and syngas ratio

In the CRM reactor (i.e. REF-CH4 unit in Fig. 3) the inlet gas ratioof CH4:CO2 at 1:1 was taken as the typical value1 with both gaseshaving a flow rate at 10 kmole/h. The CRM reactor is thus capableof generating 20 kmol/h of CO and 20 kmol/h of H2 theoretically.For an ideal methanol synthesis process another 20 kmol/h of H2 isrequired. Depending on the composition of the gas product fromthe reduction reactor, dominated by the methane oxidation extend(see R5-1 and R5-2), the extra amount of H2 can be obtained inthe following two approaches, which are

(i) Partial Oxidation of Methane Approach (R5-1), where theextra amount of H2 is obtained from mainly the carbongasification reaction (REF-C1 & REF-C2 in Fig. 3), while thepartial oxidation of methane in the reduction reactor (REDin Fig. 3) generate extra amount of syngas with a H2:CO ratioat 1:2.

(ii) Complete Oxidation of Methane Approach (R5-2), where therequired H2 is obtained equally from both methane oxida-tion in the reduction reactor and carbon gasification reac-tion. It should be noted that in practice both of theseapproaches may occur simultaneously depending on theactual operating principles and other source of uncertaintiesarose from the complexity of the process. Thus it is impor-tant to investigate both approaches in the present sim-ulation study.

Table 2 presents the summarised mass balance analysis resultsfor CLCR process operating under the typical conditions for bothpartial and complete oxidation approaches. As shown in Table 2,

Table 1Specifications and typical operating conditions of different unit operation blocks used in the ASPEN Plus model.

Unitoperation

Block type Typical operating conditions Operating temperaturerangea (�C)

Function

Chemical looping processREF-CH4 RGibbs P = 1 atm, T = 800 �C,

CH4:CO2 = 1600–900, favoured at hightemperature

CO2 reforming of methane

OXI RGibbs P = 1 atm, T = 450 �C <700, favoured at lowtemperature

Metal oxide oxidation and carbon deposition

RED RGibbs P = 1 atm, T = 600 �C 600–900, favoured at hightemperature

Metal oxide reduction

REF-C1 RGibbs P = 1 atm, T = 450 �C 400–700, favoured at hightemperature

Carbon gasification: the 1st stage reaction

REF-C2 RGibbs P = 1 atm, T = 450 �C 400–700, favoured at hightemperature

Carbon gasification: the 2nd stage reaction

REF-GAS RGibbs P = 1 atm, T = 800 �C 600–1000, favoured at hightemperature

Stoichiometric adjustment unit, water gas shiftreaction

SEP, SEP1,SEP2

Separator** Q = 0 kWt – To separate solids from gases

MIX Mixer Q = 0 kWt – To mix gasesHX1, HX2 Multistream heat exchanger Minimum temperature

approach: 15 �C– To recover/exchange heat

MINH2O Design-Spec – – To find the minimum required steam flow for acomplete carbon conversion

Methanol synthesis processCOOL + SEP Hierarchy block containing a cooler

and a separatorP = 1 atm, T = 2 �C, Q = 0 kWt – For H2O removal

C1, C2 Compressor (two-stage) P = 35, 50 bar – To compress gas stream for methanol synthesisHX3, HX4,

HX5Multi-stream heat exchanger Minimum temperature

approach: 15 �C– To recover/exchange heat

MEOH Rstoicb P = 50 bar, T = 230 �C <300 favoured at lowertemperature

To produce methanol

SEP3 Separator Q = 0 kWt – To separate liquids from gasesTOPPING RadFrac Partial condenser, P = 1.5, 2 atm,

T = 28, 71 �C– For removing ‘light ends’ products

REFINE RadFrac Total condenser, P = 1.5, 2 atm,T = 57, 119 �C

– Purification of methanol

⁄⁄ Assuming 100% gas–solid separation efficiency. It should be noted, however, in practice the separation efficiency is usually less than 100% and at high temperatures(>400 �C) granular bed filters is used instead of cyclone.

a Suitable reaction temperatures from a pure thermodynamic point of view.b It should be noted that Rstoic reactor module was used for the methanol synthesis reaction instead of Plug-flow reactor module. This is to avoid the complexity involved

in modelling methanol synthesis reaction, whose reaction mechanisms remain controversial and existing kinetic models are largely subject to operation conditions and typeof catalyst being used.

Table 2Mass balance calculation of CLCR process under the typical operating conditions.

Parameters Units Complete oxidation approach Partial oxidation approach

Theoretical valuesa Simulation results Theoretical valuesa Simulation results

Chemical looping processFeed gas composition (Total CH4, CO2) kmol/h 15, 10 15, 10 30, 10 30, 10Minimum required H2O kmol/h 10 32 20 56Minimum required Fe3O4 kmol/h 10 10 20 20CO2 production kmol/h 5 8 0 10CO production kmol/h 20 17 40 30H2 production kmol/h 40 43 80 89SR0 – 1.4 1.40 2 1.99

SR – 2 2.53 2 2.97

Methanol synthesis processRecovered H2O kmol/h 0 22.57 0 36.23Recovered CO2 kmol/h 5 4.80 0 0Net CO2 intakeb kmol/h 5 5.20 10 10.00Methanol production kmol/h 20 19.43 40 39.57

a Assuming that all reactions are complete.b CO2 consumed in the CLCR process, which does not take into account any extra CO2 emission associated with supplementary fuel usage.

398 C. Zhou et al. / Energy Conversion and Management 96 (2015) 392–402

the first major difference between the two approaches lies in theamount of feed gas and oxygen carrier used. For the complete oxi-dation approach the mole ratio for methane input in the reductionreactor (RED): minimum iron oxide inventory: methane input inthe methane reforming reactor (CRM) is 0.5:1:1, while such ratiofor the partial oxidation approach becomes 1:1:0.5. With these

ratios, the chemical looping process of CLCR plant was found ableto generate syngas with a SR’ ratio of 1.40 and 1.99 (or a SR ratioof 2.53 and 2.97) for using complete and partial oxidationapproaches, respectively. The SR’ ratio of 1.40 obtained under thecomplete oxidation approach is in a large difference to the desiredSR0 ratio at approximately 2.0–2.1 for most industrial methanol

C. Zhou et al. / Energy Conversion and Management 96 (2015) 392–402 399

synthesis plants. By comparison, partial oxidation approaches ispreferred for direct industrial application where CO2 presence isdiscouraged. However, when catalytic CO2 hydrogenation (R7) alsoforms part of the main methanol synthesis process, the complete

Table 3Heat duty analysis of CLCR process using the complete oxidation approach under typical

Operation units Heat dutya (kWt)/power(kW)

Grade of heatb

(�C)Net heat duty(kWt)

Chemical looping processREF-CH4 �622 800 �622OXI 780 450 780RED �447 600 �447REF-C1 �4 450 �4REF-C2 �5 450 �5REF-GAS �965 800 �965SEP, SEP1, SEP2 0 – 0MIX 0 – 0HX1 256 800 0HX2 618 800 0

Methanol synthesis processCOOL + SEP 212 57 212C1 412 – –C2 29 – –HX3 32 60 0HX4 392 723 108

HX5 303 230 288

MEOH 449 230 449SEP3 0 - 0TOPPING –

Condenser2 53 2

TOPPING – Reboiler �15 71 0REFINE – Condenser 288 71 288REFINE-Reboiler �283 119 0

a Positive numbers are the amount of heat being released or needed to be rejected, whunits, the numbers are the amount of heat being transferred.

b The highest temperature at which the unit operates.

Table 4Heat duty analysis of CLCR process using the partial oxidation approach under typical ope

Operation units Heat dutya (kWt)/power(kW)

Grade of heatb

(�C)Net heat duty(kWt)

Chemical looping processREF-CH4 �622 800 �622OXI 904 450 904RED �1047 600 �1047REF-C1 126 450 126REF-C2 �95 450 �95REF-GAS �1704 800 �1704SEP, SEP1, SEP2 0 – 0MIX 0 – 0HX1 256 800 0HX2 1041 800 0

Methanol synthesis processCOOL + SEP 338 54 338C1 792 – –C2 55 – –HX3 57 60 0HX4 755 746 192

HX5 634 230 592

MEOH 863 230 863SEP3 0 – 0TOPPING–

Condenser0 8 0

TOPPING–Reboiler �42 87 0REFINE–Condenser 572 74 572REFINE–Reboiler �563 119 0

a Positive numbers are the amount of heat being released or needed to be rejected, whunits, the numbers are the amount of heat being transferred.

b The highest temperature at which the unit operates.

oxidation approach which yields a low SR ratio is also a feasibleoption. Such low SR’ ratio was primarily due to the intrinsic CO2

generation during the complete oxidation of methane in the reduc-tion reactor.

operating conditions.

Comments

For preheating cold gasesFor preheating cold gases

For LTHR

For preheating cold gasesProviding heat for the reboiler of REFINE column with the rest for feedwaterheating.Providing heat for the reboiler of TOPPING column with the rest for feedwaterheating.

For LTHR

Met by HX5For LTHRMet by HX4

ilst negative ones are the amount of heat needed for the process. For heat exchanger

rating conditions.

Comments

For preheating cold gasesFor preheating cold gases

For LTHR

For preheating cold gasesProviding heat for the reboiler of REFINE column with the rest for feedwaterheating.Providing heat for the reboiler of TOPPING column with the rest for feedwaterheating.

For LTHR

Met by HX5For LTHRMet by HX4

ilst negative ones are the amount of heat needed for the process. For heat exchanger

400 C. Zhou et al. / Energy Conversion and Management 96 (2015) 392–402

On the other hand, the minimum water requirement for a com-plete carbon gasification process using the two-stage reactordesign was found to be about 2.8 and 3.2 times as that requiredtheoretically for CLCR using complete and partial oxidationapproaches, respectively. Nevertheless, most of the water con-tained in the syngas product is condensed and recycled prior tomethanol synthesis.

Overall, the mass balance analysis found that using eitherapproach CLCR process operating under the typical conditions con-sumes approximately 0.76–0.77 mole methane, 0.25–0.27 moleCO2, 0.49–0.50 mole H2O, and 0.51 mole Fe3O4 per mole of metha-nol production. To evaluate which is a better approach for CLCRprocess, energy analysis covering energy efficiency and carbonemission calculations were performed.

Fig. 5. Emission profile of CLCR process compared to other technology counterparts[29].

3.4. Energy efficiency and emission profile

Tables 3 and 4 present the detailed energy balance calculationresults for the CLCR plant using both the complete and partial oxi-dation approaches, respectively. Both tables indicate that thechemical looping process is the energy intensive process whilstmethanol synthesis process is largely exothermic and requires lim-ited heat input. Moreover, Table 3 shows the heat released in theoxidation reactor is much greater than that required in the reduc-tion reactor. The heat released in the oxidation reactor, however, isat a much lower temperature than that of the reduction reactorand thus impossible to be utilised for the reduction reaction. As aresult, high-quality external heat supply, most frequently met bynatural gas combustion, is required for the reduction reaction.This aspect, unfortunately, has been ignored by many researchers.For example, Zeman and Castaldi’s study [9] ignores the thermody-namic limitations behind each reaction taking part in the CLCR pro-cess and their conclusion that ‘‘the combined oxidation andreduction reactions are exothermic and does not incur an energydemand‘‘ is an invalid statement. Our result indicates that theenergy requirement for CLCR process under the typical conditions,essentially the methane reforming reaction, reduction reaction,and syngas stoichiometric adjustment reaction, reaches to about379 kJ/mol and 307 kJ/mol of methanol production for using thecomplete and partial oxidation approaches, respectively. This num-ber is significantly greater, about 63–102% more than the theoreti-cal value being estimated at 188 kJ/mol [9]. The mismatch ismainly due to the limitations posed by the second-law ofthermodynamics in the chemical looping process of CLCR plant.It should be noted that the above figure comparison does not takeinto account any power generation/consumption of CLCR process.Such aspect will be detailed in the following few paragraphs.

Our equilibrium thermodynamic analyses also suggest that inthe oxidation reactor of CLCR process the oxidation of FeO, if notcomplete, can be completed in the subsequent carbon gasification

Table 5Supplementary fuel requirement and the associated CO2 emission for CLCR process under

Operation units Complete oxidation approach

Supplementary fuel supplya (kmol/h) Added CO2 emission

REF-CH4 2.8 2.8OXI – –RED 2.0 2.0REF-C1 0.0 0.0REF-C2 0.0 0.0REF-GAS 4.3 4.3

Total added CO2 emission – 9.2Process CO2 consumption – 5.2Net CO2 emission – 4.0Emission profile – 0.21 mole CO2/mole

a Assuming natural gas with a gross calorific value at 50 MJ/kg was used as the suppl

reactor without thermodynamic restrictions. For a complete eval-uation of the whole process, the supplementary fuel requirementfor providing heat to various endothermic reactions in CLCR pro-cess and the associated carbon emission were calculated and pre-sented in Table 5 for both the complete and partial oxidationapproaches. As shown, an extra 0.47 mole of natural gas per moleof methanol is needed for CLCR process using either the completeor partial oxidation approach in order to meet the energy demandof the chemical looping process. This corresponds to a largeincrease on the carbon emission for CLCR process, and indeed turnsthe process from a theoretically CO2 negative process to a CO2 posi-tive process. Specifically, the net emission profile for CLCR processusing the complete oxidation approach was found to be 0.21 moleCO2 per mole methanol production, greater than the theoreticalfigure calculated at �0.15 mole CO2/mole methanol. In contrast,CLCR process using the partial oxidation approach was found tohave a net emission profile at 0.14 mole or 33% less CO2 emissionper mole of methanol production, indicating that the partial oxida-tion approach is a low emission approach for CLCR process.

Fig. 5 compares the carbon emission profile of CLCR processwith other conventional and emerging methanol production pro-cesses. As Fig. 5 shows, the carbon emission profile for methanolproduction using the conventional SRM process is about 1.16 moleCO2/mole methanol. If fully replaced by CLCR process, an emissioncut of 82–88% can be achieved among the methanol productionindustry. Also, compared with the methanol production processusing coal as the feedstock, the CLCR process was found more com-petitive in terms of emission intensity and realises an emissionreduction of about 92–95%.

For the methanol production via CO2 hydrogenation, the com-petitiveness of CLCR process will largely depend on the source ofCO2 and H2 used for CO2 hydrogenation reaction. CO2

the typical operating conditions.

Partial oxidation approach

(kmol/h) Supplementary fuel supplya (kmol/h) Added CO2 emission (kmol/h)

2.8 2.8– –4.7 4.70.0 0.00.4 0.47.6 7.6

– 15.5– 10.0– 5.5

methanol – 0.14 mole CO2/mole methanol

ementary fuel to supply the required heat.

Fig. 6. Thermal efficiency of CLCR process compared to other technology counter-parts [29].

C. Zhou et al. / Energy Conversion and Management 96 (2015) 392–402 401

hydrogenation is widely regarded as only attractive when H2 isobtained in a clean way, namely from renewable energy sources,otherwise significant energy and emission penalty would comefrom H2 production process. The lowest emission profile formethanol production can be achieved if ambient CO2 and ‘clean’H2 are used as the feedstock for CO2 hydrogenation reaction(Fig. 5). However, this option is far from reality due to the difficul-ties in achieving an efficient absorption of low-concentrationambient CO2. Alternatively, if CO2 is sourced from flue gas theemission profile of CO2 hydrogenation process is greatly increasedto about 0.6 mole CO2/mole methanol due to the energy and emis-sion penalties associated with CO2 separation process. In compar-ison, CLCR process becomes a superior process with anachievable emission reduction at about 62–77%.

Fig. 5 also shows that CLCR process is comparable with themethanol production process using biomass as the feedstock interms of total life cycle emission. Biomass when used as feedstockis considered as a carbon–neutral source due to the recyclable CO2

for biomass regeneration process. However, the methanol produc-tion process using biomass as the feedstock requires a continuousand stable biomass supply as well as substantial land usages.

Moreover, the emission profile of CLCR process was also com-pared with that of the methanol production process using the com-bined steam and CO2 reforming of methane. The combined steamand CO2 reforming of methane has been proposed to significantlyreduce the emission intensity of the conventional SRM processand was reported to have an emission profile of about 66% lessor 0.4 mole CO2/mole methanol [28]. This number, however, is stillabout 2–3 times greater than that of CLCR process. The low emis-sion profile of CLCR process clearly represents a major advantageover its technology counterparts.

Table 6 gives a summary of power generation capacity and effi-ciency calculation for CLCR process using both the complete andpartial oxidation approaches. As Table 6 shows, the total amountof power that could be potentially generated from the heatreleased in CLCR process, via a conventional regenerative steamRankin cycle, was found to be about 17–21 kW per mole of metha-nol. This amount, however, was found still not enough to offset thepower consumption of the compressor unit of the methanol syn-thesis process. Moreover, if we assume 40% of the low-grade pro-cess heat could be recovered for heating/cooling applications, the

Table 6Supplementary fuel requirements, power generation, and waste heat recoveryassociated with CLCR process under the typical operating conditions.

Operation units Complete oxidation approach Partial oxidation approach

Powergenerationa

(kW)

Waste heatrecoveryb

(kWt)

Powergenerationa

(kW)

Waste heatrecoveryb

(kWt)

OXI 234 – 271 –REF-C1 – – 38 –COOL + SEP – 85 – 135HX4 33 – 58 –HX5 58 – 118 –MEOH 90 – 173 –REFINE–

Condenser– 115 – 229

Net poweroutput

�27 – �190 –

Energy efficiencywithout LTHR

0.64 – 0.66 –

Energy efficiencywith LTHR

0.67 – 0.70 –

a Energy efficiency of 30% and 20% was taken as the nominal efficiency for a newregenerative steam Rankine cycle with a heat source temperature at 450 �C and230 �C, respectively [30].

b Assuming a 40% waste heat recovery level for heating/cooling applications.

low-grade recoverable heat was found to be approximately 9–10 kW t per mole of methanol. Overall, the energy efficiency ofCLCR process was found to reach about 64–70% depending onwhether or not waste heat recovery is taken into account, withthe partial oxidation approach being a more efficient and less car-bon intensive approach. This efficiency result also overwrites thetheoretical efficiency of CLCR process calculated at 88% [9].

In order to better understand the achieved energy efficiencylevel by CLCR process, Fig. 6 compares the energy efficiency of sev-eral representative methanol production processes with CLCR pro-cess. It shows that the overall energy efficiency of CLCR process atabout 64–70% was found quite comparable with that of the con-ventional methanol plants using SRM approach (at about 63.6%[9]), yet more efficient than the methanol production plant usingcoal/biomass as the feedstock (at 50–60%), and much more effi-cient than other parallel low-emission technologies (see Fig. 6).The above results can be mainly attributed to the innovativeCLCR process which minimises the energy demand for syngas pro-duction via a chemical looping approach, whilst other technologiesespecially the low-emission ones need significant amount ofenergy for CO2 capture and syngas production.

4. Conclusion

The current work has investigated the feasibility of a novel CLCRprocess to produce methanol via the use of transitional metal oxidesystems. Comprehensive equilibrium thermodynamic analyseswere performed using ASPEN Plus v7.3. It was identified thatCLCR process is technically superior to both the conventionalmethanol production process and other low-emission methanolproduction processes in terms of emission profile and/or energyefficiency.

Specifically, the thermodynamic analysis indicated that thethermodynamics of carbon gasification and oxidation reactions inCLCR process are working against each other with a relatively nar-row window of workable operating conditions. Also, it was foundthat the heat released in the oxidation reactor could not be recov-ered to supply to heat for the reduction reactor and thus requireshigh-quality external heat supply. Mass and energy balance analy-sis has concluded that compared to the complete oxidationapproach, the partial oxidation approach for CLCR process is a bet-ter option realising greater efficiency and lower carbon emission.

Moreover, the CLCR process operating at the typical conditionswas found to emit as low as 0.14 mole CO2/mole methanol, whichis about 82–88% less than the conventional SRM methanol produc-tion process and well below the emission levels of other paralleltechnologies. The energy efficiency of CLOC process found at 64–70% was also among the highest efficiency band among manymethanol production processes. Overall, it has been concluded that

402 C. Zhou et al. / Energy Conversion and Management 96 (2015) 392–402

CLCR process can be a promising step change solution for theadvancement of methanol economy.

Acknowledgement

The authors wish to acknowledge the financial support theyhave received from Australian Research Council (ARC) and theUniversity of Newcastle, Australia.

References

[1] Felderhoff Michael et al. Hydrogen storage: the remaining scientific andtechnological challenges. Phys Chem Chem Phys 2007;9(21):2643–53.

[2] Chalk Steven G, Miller James F. Key challenges and recent progress in batteries,fuel cells, and hydrogen storage for clean energy systems. J Power Sources2006;159(1):73–80.

[3] Liao S, Jiang D, Cheng Q, Huang Z, Zeng K. Effect of methanol addition intogasoline on the combustion characteristics at relatively low temperatures.Energy Fuels 2006;20:84–90.

[4] Oil and Gas Security. Emergency Response of IEA countries – Australia section– International Energy Agency (IEA) report; 2011.

[5] Yang C-J, Jackson RB. China’s growing methanol economy and its implicationsfor energy and the environment. Energy Policy 2012;41:878–84.

[6] Cherry RS, Aumeier SE, Boardman RD. Large hybrid energy systems for makinglow CO2 load-following power and synthetic fuel. Energy Environ Sci2012;5:5489–97.

[7] Shayan SB, Seyedpour S, Ommi F, Moosavy S, Alizadeh M. Impact of methanol–gasoline fuel blends on the performance and exhaust emissions of a SI Engine.Int J Automot Eng 2011;1.

[8] El-Bousiffi M, Gunn D. A dynamic study of steam-methane reforming. Int JHeat Mass Transf 2007;50:723–33.

[9] Zeman F, Castaldi M. An investigation of synthetic fuel production via chemicallooping. Environ Sci Technol 2008;42:2723–7.

[10] Chmielniak T, Sciazko M. Co-gasification of biomass and coal for methanolsynthesis. Appl Energy 2003;74:393–403.

[11] Yao S, Takemoto T, Ouyang F, Nakayama A, Suzuki E, Mizuno A, et al. Selectiveoxidation of methane using a non-thermal pulsed plasma. Energy Fuels2000;14:459–63.

[12] Aasberg-Petersen K, Dybkjaer I, Ovesen CV, Schjodt NC, Sehested J, ThomsenSG. Natural gas to synthesis gas – catalysts and catalytic processes. J Nat GasSci Eng 2011;3:423–59.

[13] Shah K, Moghtaderi B, Wall T. Selection of suitable oxygen carriers forchemical looping air separation: a thermodynamic approach. Energy Fuels2012;26:2038–45.

[14] Song H, Shah K, Doroodchi E, Wall T, Moghtaderi B. Analysis on chemicalreaction kinetics of CuO/SiO2 oxygen carriers for chemical looping airseparation. Energy Fuels 2013;28:173–82.

[15] Shah K, Moghtaderi B, Zanganeh J, Wall T. Integration options for novelchemical looping air separation (ICLAS) process for oxygen production in oxy-fuel coal fired power plants. Fuel 2013;107:356–70.

[16] Shah K, Moghtaderi B, Wall T. Effect of flue gas impurities on the performanceof a chemical looping based air separation process for oxy-fuel combustion.Fuel 2013;103:932–42.

[17] Shah K, Zhou C, Song H, Doroodchi E, Moghtaderi B. A novel hybrid chemicallooping oxy combustor process for combustion of solid and gaseous fuels:thermodynamic analysis. Energy Fuels 2014. http://dx.doi.org/10.1021/ef502389t.

[18] Moghtaderi B, Zanganeh J, Shah K, Wu H. Application of concrete anddemolition waste as CO2 sorbent in chemical looping gasification of biomass.Energy Fuels 2012;26:2046–57.

[19] Zhang Y, Doroodchi E, Moghtaderi B. Chemical looping combustion of ultra lowconcentration of methane with Fe2O3/Al2O3 and CuO/SiO2. Appl Energy2014;113:1916–23.

[20] Shah K, Zhang Y, Doroodchi E, Moghtaderi B. Development of novel stone dustlooping process for mitigation of ventilation air methane. In: 3rd internationalconference on chemical looping, 9–11 September 2014, Gothenburg, Sweden.

[21] Aspen Plus User Manual Version 7.3. Aspen Technologies Inc., Cambridge(MA); 2012.

[22] Lurgi GmbH. Lurgi megamethanol. <http://www.zeogas.com/files/83939793.pdf> retrieved 10.11.14.

[23] Barranon DCC. Methanol and hydrogen production: energy and cost analysis.Master thesis. Lulea University of Technology; 2006.

[24] Hamelinck CN, Faaij APC. Future prospects for production of methanol andhydrogen from biomass. J Power Sources 2002;111:1–22.

[25] Coteron A, Hayhurst AN. Kinetics of the synthesis of methanol from CO + H2

and CO + CO2 + H2 over copper-based amorphous catalysts. Chem Eng Sci1994;49:209–21.

[26] Rahman D. Kinetic modeling of methanol synthesis from carbon monoxide,carbon dioxide, and hydrogen over a CU/ZnO/Cr2O3 catalyst. Master thesis. SanJose State University; 2012.

[27] Bromberg L, Cheng W. Methanol as an alternative transportation fuel in theUS: Options for sustainable and/or energy-secure transportation. Final report;2010.

[28] Lim Y, Lee C-J, Jeong YS, Song IH, Lee CJ, Han C. Optimal design and decision forcombined steam reforming process with dry methane reforming to reuse CO2

as a raw material. Ind Eng Chem Res 2012;51:4982–9.[29] Galindo Cifre P, Badr O. Renewable hydrogen utilisation for the production of

methanol. Energy Convers Manage 2007;48:519–27.[30] Rout IS, Gaikwad A, Verma VK, Tariq M. Thermal analysis of steam turbine

power plants. J Mech Civ Eng 2013;7:28–36.