Methanol-Managing greenhouse gas emissions in the production chain by optimizing the resource base

: The growing demand for methanol as fuel and global competition for resources are key drivers behind the need to find new routes for the production of bulk chemicals such as methanol. Widening the resource base is also linked to the increasing concentrations of methane in the atmosphere. Furthermore, managing greenhouse gas emissions is vital in developing new technologies. This paper compares production routes for methanol based on a cradle-to-gate life cycle assessment (LCA). The LCA is limited to the impact categories of global warming potential (GWP 100 ) and energy use. The highest GWP 100 value of 2.97 kg CO 2 eq/kg CH 3 OH is for methanol from coal, and the lowest, negative emission of 0.99 kg CO 2 eq/kg CH 3 OH is for methanol in co-production with renewable corn ethanol. A comparison of production routes is performed using the carbon dioxide equivalent abatement cost, and the production cost of methanol. The best performing technology on both production cost and GWP 100 is methanol produced by gasification from wood biomass. The factors affecting the results are addressed.


Introduction
Global competition for resources and increasing population, especially in the developing countries, are the trends behind the need to more actively develop technological processes which balance greenhouse gas emissions (GHG), increase the use of renewable resources, and boost the circular economy in the production of bulk chemicals.One of these chemicals is methanol, with a current production rate of 95 million metric tons per year [1].China dominates the global production and consumption of methanol with over 50%.China is included in Asia Pacific in Figure 1a,b.The demand for methanol has shown significant growth in recent years, resulting in over 100% growth of global production between 2005 and 2016.This growth in demand has mainly originated from the increased use of methanol as fuel and raw material for methanol-to-olefins (MTO) plants [2].In spite of the globally distributed production, every day 80000 metric tons of methanol is shipped long distances, adding to its environmental load.In 2015, both China and the US imported 5 million metric tons of methanol.The key exporters of methanol are located in the Middle East and in Latin America [3,4].[3]; (b) World methanol demand of 90 million metric tons by region in % in 2018 [4].
The global importance of methanol is widely recognized and several studies have assessed its use as both marine and land-based transport fuel [5] and platform chemical [6] as well as power storage media [1].This has increased the research on both synthesis gas generation and methanol synthesis.A low carbon resource base and renewable energy options are possibilities for finding feasible solutions to replace the traditional production routes of methanol.
Earlier research in the 1980s and 1990s concentrated mainly on optimizing the mass and heat balance of the conventional methanol production routes.The traditional feedstock for methanol synthesis is fossil natural gas transformed into synthesis gas prior to the actual synthesis of methanol.In addition, especially in China, the coal gasification route for producing synthesis gas is widely in use.Coke oven gas and shale gas are also fossil resource based raw materials that are used for the synthesis gas route to methanol.A recent comparison of a coke oven gas derived and natural gas based methanol production revealed that the former would be a better alternative in terms of CO 2 emissions [7].However, all these processes result in fossil GHG emissions, raising the level of both carbon dioxide (CO 2 ) and methane (CH 4 ) in the atmosphere and thus call for widening the resource base in the production of methanol.The need to widen the resource base is also linked to the increasing concentrations of CH 4 in the atmosphere.Approximately two-thirds of global CH 4 emissions are attributable to anthropogenic activities, offering increasing opportunities for climate change mitigation [8].
Several researchers have studied the kinetics and technological solutions to improve the overall yield and thermal efficiency of methanol synthesis.Riaz et al. [9] concluded in their review that fluidized bed reactors, membrane reactors and thermally coupled reactors are promising solutions to improve the yield of methanol.
Carbon capture and utilization (CCU) is gaining more support than the earlier promoted carbon capture and storage (CCS).This has also made the exhaust gases of power stations potential sources of raw material for the new methanol synthesis units based on CO 2 and hydrogen (H 2 ).The Lurgi/Air Liquide methanol technology is the most advanced with a potential of daily operating capacity of up to 5000 tons [1,6].Anicic et al. [10] compared the direct synthesis of methanol from CO 2 and H 2 with a reverse water gas shift (RWGS) reaction and subsequent methanol synthesis.Their modelling results indicated that direct methanol synthesis in terms of energy and economic efficiency is more favourable than the RWGS route.An adiabatic RWGS reactor prior to methanol synthesis as in [1] favourably prevents catalyst deactivation.CO 2 capture from the ambient air for methanol synthesis is also an option [11].
The lower than 30% conversion rates of single pass rectors have raised interest in different polygeneration and looping systems for methanol synthesis [12].In addition, various oxy-fuel technologies in power generation are suitable for combined power and methanol production [13].The fossil CH 4 emissions of methanol production can be considerably reduced by utilizing tri-reforming technology, combining the CO 2 in flue stack gases with CH 4 as a feed stream [14].A combination of power stations and methanol plants using natural or synthetic based CH 4 is also possible in the catalytic tri-reforming process using CO 2 reforming, steam reforming and partial oxidation in the production of methanol.Minutillo and Perna [15] found this to be the most suitable for flue gases from a steam cycle power plant with low oxygen concentrations.
In addition to distiller grains and solubles (DGS), renewable fuel processes, such as in a corn ethanol production facility, also produce CO 2 as a by-product which similarly can be utilized for methanol production in direct synthesis with H 2 .Matzen and Demirel [16] assessed the
environmental impacts of this production route using a life cycle assessment (LCA) with an economic allocation for CO 2 and direct impacts of electrolysis for H 2 .
Research on the gasification of biomass for the synthesis gas route to methanol is extensive.In flexible integrated gasification polygeneration facilities producing either electricity or chemicals from coal and wood pellets, methanol production considerably reduces the CO 2 in flue gas streams.When the biomass fraction in the feed exceeds 40% on the energy basis, the production has been considered CO 2 neutral [17].Van Rens et al. [18] compared the production of methanol, dimethylether and hydrogen from a wood mixture of 10.4% bark, 14.8% wood chips and the remaining tops and branches.Methanol achieved the highest efficiency of production.Melin et al. [19] compared stand-alone methanol and ethanol production with combined production of both from softwood pine loggings.The study was further developed to cover separate lignin and carbohydrate processing for methanol and methane production [20].The water content of biomass increases the energy demand.Pre-treatment of biomass with torrefaction improves the overall energy economy of methanol production [21].
Different technological choices and process integration in production clusters offer possibilities to improve the feasibility of methanol production from biomass.Holmgren et al. [22,23] concluded that heat integration in particular is a critical factor, and the magnitude of the GHG emissions reductions is dependent upon the reference energy systems and associated emission factors.One of the potential platforms for the production of methanol is a pulp and paper mill (PPM).Andersson et al. [24] modelled methanol production in an existing PPM using a pressurized entrained biomass gasification unit for the generation of the synthesis gas (syngas).
Various side-streams and by-products need valorization to improve the overall feasibility of biodiesel produced by the transesterification of fatty acids, for instance.The reforming of the by-product glycerol to syngas suitable for methanol synthesis has been widely studied.Up to 0.270 kg methanol/kg glycerol can be obtained from the syngas produced by supercritical water reforming with an overall energy efficiency of 38.0%.In addition, a sequestration rate of 0.38 kg CO 2 /kg of glycerol is possible [25].
Thermal integration analysis enables the identification of the possibilities for improving the efficiency of production chains.Bludowsky et al. [26] compared a conventional high temperature biomass gasification process and an aqueous-phase reforming (APR) scheme to produce the syngas stream for methanol synthesis.The latter requires an RWGS unit prior to the methanol synthesis, and had an external energy demand of over 55% higher than that of the biomass gasification process.The carbon balance of both processes showed over 50% carbon losses.The biomass feedstock was modelled as a pure glucose thus reducing the value of the results to an indicative theoretical level.
Solar energy pathways to produce methanol are currently considered unfeasible options.There are several reasons for this: (1) solar energy as a renewable electricity source for H 2 production would require large solar ponds to collect the solar energy [27]; (2) large amounts of water would be required in regions which suffer from water shortages [28]; (3) solar gasification of biomass has been assessed as not competitive with the coal-based methanol production technologies [29].Moreover, the design of large-scale solar thermochemical reactors is challenging and no commercial applications are foreseen in the near future.
The production of syngas by gasification from different resources including biomass, direct captured CO 2 and polygeneration systems have dominated the research on methanol during the last two decades.An overview of different future options can, for example, be found in Bertau et al. [30].
Less attention has been focused on how to optimize the resource base and technological solutions in terms of greenhouse gas (GHG) emissions, and how to implement circular economy approaches in the production of methanol.This paper compares different emerging possibilities to widen the current resource base of conventional methanol production technologies and analyses the overall GHG emissions utilizing life cycle assessment.

Methods
The aim of this study is to compare the life cycle GHG balance of methanol produced from different resources and technological combinations on a cradle-to-gate basis.The final use of methanol as a chemical or transport fuel is not included because of the various possible end uses.
Life cycle assessment (LCA) is used to assess the amount of GHG expressed in carbon dioxide equivalents (CO 2 eq), and for the comparison of different methanol processes included in the study.LCA is a common standardized tool [31,32] used for the environmental assessment of fuels and chemicals [5,33].A GHG management matrix developed by one of the authors [34] is used as an additional tool to help the screening of reviewed studies, and for the scoping of LCA.The reference base case is the production of methanol from coal as a feedstock.
The selected method is that of attributional LCA (ALCA) on a cradle-to-gate basis, reflecting the production systems as described in the references.Any use of allocation, either energy or economic, is included.The functional unit considered is 1 kg of methanol produced.The assessment is limited to the impact categories of global warming potential (GWP 100 ) and energy use.The energy use is estimated based on either life cycle inventory (LCI) results or on the overall efficiency, and the highest value is selected.
In addition, a preliminary evaluation of the carbon dioxide equivalent abatement and methanol production cost estimates are made using a capital recovery factor (CRF) for the first 10 years of investment.CRF = i(1 , where i is the interest rate and n is the number of annuities received for a 10-year period including 35-40% of the total investment cost for longer lifecycles, depending on the overall payback period of the investment, with 5% and 10% annual discount rates.The avoided amount of CO 2 eq is estimated as the difference between the GWP 100 impact and the lower reference value on an annual basis.
The robustness of key assumptions, uncertainty in the obtained data and sensitivity of the obtained results are assessed, and the limitations of the method used are discussed.

Life cycle inventory data (LCI)
This section briefly describes the production routes of methanol selected for comparison.A simplified flowchart of the six processes is illustrated in Figure 2, and an overview of the methodological choices is shown in Tables 1A-F.

Case A-methanol from fossil natural gas
Fossil methane (CH 4 ) from natural gas and fossil hydrogen (H 2 ) produced by the steam reforming of CH 4 are the raw materials used in this traditional process.The LCA inventory data for this methanol production is based mainly on Matzen et al. [35], Brynolf et al. [5] and Minutillo and Perna [15,36].
The key reactions for the steam reforming of natural gas are: The key reactions for methanol production from synthesis gas, including the reverse water gas shift (RWGS) reaction, are: The key process units include a steam reformer, a methanol reactor and a raw methanol distillation unit.
The feedstock is average natural gas (NG) extracted in Norway and transported 648 km to the production site of methanol located in Sweden.The methanol production efficiency at the lower heating value (LHV) is 66% and the purity of the final product is over 99% [5].Norwegian electricity mix is used for the NG extraction and preprocessing, and the average Swedish electricity mix is used for the methanol production site [5].The methodological choices for this case are shown in Table 1A.
Table 1A.Methodological choices for Case A.

Investigated options
Natural gas (NG) extracted and preprocessed in Norway.Liquefied natural gas (LNG) produced in Algeria for alternate case [28].

Functional unit 1 kg of methanol produced Time horizon 2010-2020 Geographical boundaries
Norway for NG and Gothenburg, Sweden for methanol production.

System boundaries
The study includes all production steps shown in the flowchart (Figure 2) Manufacturing of capital goods and auxiliary materials are not included.

Allocation
No allocation used.Impact categories CO 2 equivalents (CO 2 eq) using global warming potential factors for a 100-year period (GWP 100 ).Energy use.

Limitations
Greenhouse gas emissions connected to land use change are not considered in this study.

Case B-methanol from coal gasification
Synthesis gas from coal and fossil hydrogen (H 2 ) produced by the steam reforming of CH 4 or by the water gas shift (WGS) reaction are the basic raw materials used in this traditional process.The LCA inventory data for this case of methanol production is based on Matzen et al. [35], Trudewind et al. [28] and Trop et al. [21].
The key summary reaction for coal gasification and the key reaction for water gas shift are: The key reactions for methanol from synthesis gas are: The key process units include a coal gasification unit, a gas cleaning unit, a steam reformer or a WGS reactor, a methanol reactor and a raw methanol distillation unit.
Coal is gasified in an entrained flow coal gasifier and the cooled synthesis gas produced is treated for sulphur removal in a Rectisol process.The released heat is used to heat the synthesis gas after desulphurization and to produce low pressure steam.The mixture of low pressure steam and purified syngas is conveyed to the water gas shift (WGS) process.After the WGS reactor, the conditioned gas is ready for the methanol synthesis unit and further purification in the distillation unit.This well-to-tank LCA ends at the methanol storage facility.The mining, coal conditioning and transport chain starts in South Africa.Coal is transported 540 km to the coast by diesel engines, loaded and transported 12,975 km by freight to the port in Rotterdam, the Netherlands, and from there a further 668 km by barge to the gasification plant in Germany.The methanol plant is assumed to be in the same location.The methodological choices for this case are shown in Table 1B.

Investigated options
Coal extracted and preprocessed in South Africa

Functional unit 1 kg of methanol produced Time horizon 2010-2020 Geographical boundaries
Coal production in South Africa, methanol production in Germany.System boundaries, allocation, impact categories, included primary pollutants and limitations are similar to Case A.

Case C-methanol production by flue gas tri-reforming integrated with power production
Fossil methane (CH 4 ), fossil coal, fossil CO 2 and fossil H 2 are the raw materials used in this process.The LCA inventory data for this methanol production utilizing a flue gas stream of a coal-fired/CH 4 -fired power plant, is based mainly on Minutillo and Perna [15,36] and Trudewind et al. [28].The key reactions for methane tri-reforming are: The reactions for methanol synthesis are the same as in Case B, i.e. reactions 3B and 4B.The flue gases from the power plant, which contain N 2 , CO 2 , H 2 O and O 2 , are used as reactants (reactions 1C, 2C and 3C) to convert the methane into a syngas with the optimal H 2 /CO ratio equal to 2 required for methanol synthesis.
The key process units include a coal-fired power plant, a tri-reforming unit connected to a CH 4 -fired catalytic burner, a heat recovery system, a syngas compressor, a methanol reactor and a distillation unit.For a detailed technical description of the process units, please refer to Minutillo and Perna [15].The methodological choices for this case are shown in Table 1C.The study includes all production steps shown in the flowchart (Figure 2).Manufacturing of capital goods and auxiliary materials are not included.Flue gases from a steam cycle power plant are used for methanol production.

Allocation
Allocation of the emissions between net power and methanol production.Impact categories, included primary pollutants and limitations are similar to Case A.

Case D-methanol in co-production with renewable corn ethanol
Biogenic CO 2 and renewable H 2 are the raw materials used in this process.The LCA data for methanol production from a CO 2 side stream of renewable ethanol production is from Matzen and Demirel [16], with the key assumption that economic allocation is used to account for the impact of CO 2 emissions.Each gallon (3.79 litres) of corn ethanol produces 3.08 kg of CO 2 and 2.56 kg of distiller grains and solubles (DGS).Wind power generated electrolytic H 2 data is used for the LCA.The methanol production data is from Matzen et al. [35].
The key reactions for this methanol route are: The key process units include a wind power driven electrolysis unit for splitting water into hydrogen and oxygen, a hydrogen compressor, a feed system for CO 2 from the corn ethanol unit, a methanol synthesis reactor of Lurgi type and a distillation unit.
Hydrogen for this small-scale methanol synthesis is produced by the wind-powered electrolysis of water.Oxygen is considered as a by-product of electrolysis with an output of 1.530 kg O 2 /kg methanol produced.Wind power is selected because its cost is around 38% lower than that of solar photovoltaic power [35].CO 2 is recovered as a by-product of renewable ethanol production.The raw material for the ethanol production is dry milled corn.The allocation of environmental load from ethanol production is based on the economic values of ethanol, DGS and CO 2 .
Emissions of the compression and transportation of the CO 2 are assumed to be negligible.The overall energy efficiency for the renewable methanol is around 58%.In addition to methanol, the process produces waste water, flue gas and steam used either for electricity production or for process heat.The flue gas stream contains mostly CO 2 , with less than 0.5% of the produced methanol being lost.The waste water stream contains less than 1% methanol.For a detailed process description, please refer to Matzen and Demirel [16].The methodological choices for this case are shown in Table 1D.The study includes all production steps shown in the flowchart (Figure 2) Production and operation of wind power and electrolyser are partially included.Other manufacturing of capital goods and auxiliary materials are not included.Allocation Economic allocation used.Impact categories, included primary pollutants and limitations are similar to Case A.
2.2.5.Case E-methanol from coal power plant CO 2 and H 2 from electrolysis Fossil CO 2 and renewable H 2 are the raw materials used in this process.The LCA inventory data for methanol production from a purified CO 2 side stream of a coal-fired power station in combination with a wind power driven electrolysis unit is mainly from Matzen et al. [35], Pontzen et al. [6] and Räuchle et al. [1].The key reactions for this methanol route are: The key process units include a CO 2 capture unit, an electrolysis unit for H 2 production, a methanol reactor unit and a crude methanol distillation unit.
Overall reductions in CO 2 emissions can only be achieved if the H 2 is produced from renewable resources, i.e. the electrolysis uses solar-, wind-or biomass-based electricity.Räuchle et al. [1] made a detailed analysis of methanol plants by AirLiquide/Lurgi, concluding that the best overall efficiency with renewable H 2 reached 51% (assumed on an LHV basis).H 2 is generated from surplus electric power by electrolysis and recycled CO 2 comes from a CCU unit of the power station.The methodological choices for this case are shown in Table 1E.

Investigated options
Coal extracted and preprocessed in Germany.Hydrogen and methanol production in Germany.

Functional unit 1 kg of methanol produced Time horizon 2010-2020 Geographical boundaries Germany Allocation
Allocation of emissions used, power station operational emissions excluded.System boundaries, impact categories, included primary pollutants and limitations are similar to Case A.

Case F-methanol from wood biomass
This case is a stand-alone methanol production unit from wood biomass.The LCA inventory data is mainly based on Holmgren et al. [22,23], Andersson et al. [24].The process is similar to that of coal gasification (Case B), except that wood biomass is the feedstock.The synthesis gas from biomass gasification is the basic raw material used in this process.The hydrogen (H 2 ) content is adjusted in the water gas shift (WGS) reactor.Additionally, the result is compared with a methanol production unit integrated into an existing pulp and paper mill.
The key reactions for this methanol route are given below.
The key summary reaction for biomass gasification and key reaction for water gas shift are: The key reactions for methanol from synthesis gas are: The average wood biomass composition differs from fossil resources in its oxygen content and original humidity of 40-50%.The molecular composition of Nordic softwood pine, (wt.%) of C 53.3, H 6.1, O 41.2, N 0.1, S 0.01, ash 0.7, illustrates the oxygen content well [19].
The key process units include a cryogenic air separation unit (ASU) for oxygen production, a pressurized entrained biomass gasification unit (PEBG), a sulphur tolerant water gas shift (WGS) reactor, an acid removal unit, a methanol synthesis reactor and a distillation unit.The methodological choices for this case are shown in Table 1F.

Investigated options
Biomass harvested and preprocessed in Sweden.Electricity produced in Sweden Functional unit 1 kg of methanol produced Time horizon 2010-2020 Geographical boundaries Sweden System boundaries, allocation, impact categories, included primary pollutants and limitations are similar to Case A.
The gasification unit includes a biomass rotary drying unit where the biomass is dried to a water content of 8%.The grinding unit reduces the dried biomass to an average particle size between 0.6 and 0.8 mm prior to being fed to pressurized entrained flow biomass gasification (PEBG) with pure oxygen as a gasification agent.The thermal energy input to both the rotary dryer and grinding unit are included and recalculated as electricity in the overall power consumption [24].The same applies to the cryogenic air separation unit (ASU).Unreacted syngas is recycled to the methanol reactor with a 5% purge.

Results
This section includes the life cycle inventory results for all six cases.The cradle-to-gate environmental impact analysis covers greenhouse gas emissions shown as GWP 100 impact and energy use.The CRF calculation results for the preliminary carbon dioxide equivalent abatement cost estimation and the methanol production costs are listed in Section 3.3.The uncertainty of results is addressed including the assumptions made, quality of data and limitations arising from the methodology used.

Life-cycle inventory analysis
The energy and material balances are calculated in a uniform manner using the functional unit of 1 kg methanol (CH 3 OH).The energy use units differ from case to case depending on the original values available for the case.The shown key inputs generate the given key output and related primary pollutants.If a primary pollutant is missing from the table, it indicates that it was not considered in the reference used.
The energy value for methanol used in all calculations is 21.832 MJ/kg methanol [28] located somewhere between the higher heating value (HHV) of 23.8 MJ/kg methanol and the lower heating value (LHV) of 21.1 MJ/kg of methanol used by Minutillo and Perna [15] The cradle-to-gate energy use for producing methanol is either estimated from LCI tables or from the overall efficiency, and the highest value is chosen.
When available, the GWP 100 impact is given separately for raw material extraction and processing, transport, power production and methanol synthesis.The GWP 100 value is the sum of the carbon dioxide equivalents (CO 2 eq) of CO 2 , CH 4 and N 2 O.The summary GWP 100 is listed for the functional unit (1 kg CH 3 OH) using the unit of kg CO 2 eq.The CO 2 eq of CO 2 equals one.The values for CH 4 and N 2 O are calculated as CO 2 eq using the multiplying coefficients of 25 for CH 4 and 298 for N 2 O from the IPCC AR4 report (2007), resulting in the equation: (GWP 100 in kg CO 2 eq = kg CO 2 + (kg CH 4 ) × 25 + (kg N 2 O) × 298).In addition to the LCI results listed in Tables 4A-F, alternative cases impacting the GWP 100 results are considered for each case.The inventory results are further compared to the reference base case of methanol from coal as a feedstock.

Case A-methanol from fossil natural gas
The NG extracted in Norway is transported to the production site in Sweden, where syngas is generated by the steam reforming of NG and transformed to methanol, as described in Section 2.2.1.The overall efficiency on an LHV basis is 66% The GWP 100 for methanol from fossil NG is 0.462 kg CO 2 eq/kg methanol, and the corresponding energy use is 11.5 MJ/kg methanol.The inventory results for this case are shown in Table 2A.
An alternative case for the production of methanol is NG extracted in Algeria and transported to a production site in Germany [28].In this case the GWP 100 for methanol varies from 0.873 to 0.881 kg CO 2 eq/kg methanol (Alternative 1A and 2A in Figure 6).The primary energy demand varies from 33.277 MJ/kg methanol [5] to 37.114 MJ/kg methanol [28], increasing the uncertainty of energy use by over 10 per cent if the latter reference is used.

Case B-methanol from coal
Methanol from coal is mostly produced in China.However, here we selected coal from South Africa with an LHV of 24.7 MJ/kg [28] and complemented the results with Czech coal from Ledvice with an LHV of 21.73 MJ/kg [21] in the synthesis part.The methanol output is based on 57.7% cold gas efficiency (CGE) where the CGE was calculated using higher heating values (HHV), Trop et al. [21].
The GWP 100 for methanol from coal is 2.965 kg CO 2 eq/kg methanol, and the corresponding energy use (without transport fuels) is 22.8 MJ/kg methanol.The inventory results for this case are shown in Table 2B.
Alternatively, the GWP 100 impact of methanol produced by coal gasification can be considerably lowered by replacing 40% [17] to 50% [21] of the coal energy with biomass (Alternative 1B and 2B in Figure 6).

Case C-methanol by flue gas tri-reforming integrated with power production
The tri-reforming plant produces 990 MW methanol and 52.3 MW net power, and energy allocation is used to split the emissions between power and methanol production.The electrical efficiency of the coal-fired power plant is 38.6% calculated on an LHV basis [15].The overall efficiency of the power plant and methanol production is 56% on an LHV basis.

AIMS Energy
Volume 6, Issue 6, 1074-1102.The natural gas (NG) composition (molar basis) is: 90.0%CH 4 , 4.4% C 2 H 6 , 0.1% C 3 H 8 , 1.3% CO 2 , 4.3% N 2 [15].The heating values of NG are 44.8MJ/kg (LHV) and 49.7 MJ/kg (HHV).The assumption is that the production losses and transport distances for the Pennsylvania Upper coal to Italy do not exceed the values used in Case B. Therefore, the primary pollutants for coal mining, conditioning and transport are calculated based on the reference values of Trudewind et al. [28], and included as upstream primary pollutants for the power production.The natural gas production losses are estimated to be 0.27% for the methanol production amount of NG [37].The assumption is that NG transport emissions do not exceed the values used by Trudewind et al. [28], where 58% of the methane is transported from Algeria by pipeline and 42% by shipment in LNG freight ships to Europe.
The GWP 100 for methanol by flue gas tri-reforming is 1.392 kg CO 2 eq/kg methanol, and the corresponding energy use is 17.7 MJ/kg methanol.The inventory results for this case are shown in Table 2C.Without the allocation of emissions between power and methanol production, the methanol production would be loaded with 0.31 kWh/kg CH 3 OH electricity emissions, meaning additional emissions of 0.961 kg CO 2 eq/kg CH 3 OH with the given GWP 100 for net power
production (Table 2C).This would increase the total GWP 100 for methanol production by tri-reforming to 2.353 kg CO 2 eq/kg CH 3 OH (Alternative 1C in Figure 6).

Case D-methanol in co-production with renewable corn ethanol
This process produces corn ethanol, DSG and methanol.The methanol is produced from the biogenic CO 2 captured from ethanol production and renewable H 2 , and economic allocation is used to divide the environmental load between ethanol, DSG and biogenic CO 2 .
The GWP 100 for methanol in co-production with corn ethanol and DSG is −0.989 kg CO 2 eq/kg methanol, and the corresponding energy use is 15.8 MJ/kg methanol.The inventory results for this case are given in Table 2D.
The economic allocation factor of total production emissions and energy use for CO 2 is 6.1%.This percentage gives a GWP 100 load of 0.0784 CO 2 eq/kg CO 2 used for methanol production (Table 2D).If we change this percentage to 100% with the 1.43 kg CO 2 needed for one kg methanol produced, the GWP 100 load from CO 2 production will change to 1.8379 kg CO 2 eq/kg methanol produced.The overall GWP 100 load of methanol production would change from the negative value of −0.989 to 0.8489 kg CO 2 eq/kg methanol produced.This demonstrates the impact of economic allocation on the LCA result for case D.

AIMS Energy
Volume 6, Issue 6, 1074-1102.The GWP 100 value for methanol in co-production with corn ethanol in Table 2D is taken from Matzen and Demirel [16].If the summary value for CO 2 , N 2 O and CH 4 is used, the GWP 100 value will change to −1.008 kg CO 2 eq/kg methanol (Alternative 1D in Figure 6).
The hydrogen production by electrolysis has a GHG impact of 0.97 kg CO 2 eq/kg H 2 of which 78% are attributed to wind turbine production and operation.The GWP 100 for H 2 production emissions shown in Table 2D includes 14.7% of these emissions [35].Including all the H 2 emissions would change the GWP 100 for methanol by 0.159 units from −0.989 to −0.83 (Alternative 2D in Figure 6).
Changing the H 2 production from electrolysis to NG-or coal-based would considerably change the emissions.Starting from NG the emissions are 7.33 kg CO 2 /kg H 2 , and correspondingly from coal 29.33 kg CO 2 /kg H 2 using 75% energy efficiency [35].The use of biogenic CO 2 and H 2 from NG in methanol production would change the negative GWP 100 value for methanol in Table 2D to a positive value of 0.418 kg CO 2 eq/kg methanol (Alternative 3D in Figure 6).The use of biogenic CO 2 and H 2 from coal would change the negative GWP 100 value for methanol in Table 2D to a positive value of 4.642 kg CO 2 eq/kg methanol (Alternative 4D in Figure 6).This indicates clearly that the results of the LCA are altered drastically by changing the raw material production route or other assumptions.Volume 6, Issue 6, 1074-1102.

Case E-methanol from captured CO 2 and H 2
This methanol production route is based on the same type of technology as in Case D with captured CO 2 from a coal-fired power station, and the emissions of power production are excluded from the GHG balance.The overall efficiency of methanol production is assumed to 51% on an LHV basis.The hard coal mining, processing and transport emissions are estimated as in Case B, but include only a short transport distance of 541 km inside Germany.The assumption is that these emissions are a maximum of 10% of the corresponding emissions for Case B, and total 0.052 kg CO 2 eq/kg CH 3 OH.The emissions of flue gas cleaning, CO 2 CCU unit and methanol synthesis are adapted from Case D, and corrected with the efficiency difference between cases D and E, i.e. by multiplying the corresponding values from Case D by 58/51 = 1.1373.
The GWP 100 for methanol from the captured CO 2 and H 2 is −0.752 kg CO 2 eq/kg methanol, and the corresponding energy use is 15.8 MJ/kg methanol.The inventory results for this case are shown in Table 2E.
The negative GWP 100 for methanol production derives from the capture of fossil CO 2 and the use of renewable wind power for the production of H 2 .
Fossil CO 2 can be replaced by renewable CO 2 from a biomass-fired power station.This would further increase the biogenic carbon capture and enforce the negative value of GWP 100 for methanol production.The replacement of renewable H 2 with fossil H 2 generated from natural gas (NG) or by gasification of coal would change the overall GWP 100 for methanol as shown in Case D.
Similarly, a different source, i.e. the electrolysis of water, increases the overall GWP 100 for methanol.Electricity from the grid would increase the GWP 100 for methanol by 3.933 kg CO 2 eq/kg CH 3 OH with the average electricity emissions of 0.56 kg CO 2 eq/kWh in Germany [38].For Case E, this would change the initially negative overall emissions to a positive load of 3.181 kg CO 2 eq/kg CH 3 OH (Alternative 2E in Figure 6) exceeding even the value for Case B.
The methanol plant could also be located inside a petroleum refinery complex and utilize the H 2 produced by the refinery.The impact of the CO 2 stream would remain unchanged compared to Case E, and the H 2 stream would increase the GWP 100 for methanol by 2.82 kg CO 2 eq/kg CH 3 OH.For Case E this would change the overall GWP 100 for methanol to 2.068 kg CO 2 eq/kg CH 3 OH (Alternative 1E in Figure 6).The H 2 emissions produced by the petroleum refinery are based on Moretti et al. [39].3.1.6.Case F-methanol from renewable wood biomass This process uses 338 MW wood biomass input with 8% humidity and 41 MW external electricity to produce 187 MW methanol and 55 MW district heat.The moisture content of the harvested wood biomass is assumed to be 40%.The biomass-to-methanol overall efficiency is 55.3% assumed on an LHV basis [24].
The GWP 100 for biomass harvesting is assumed to be 3.89 g CO 2 eq/MJ biomass including the transport emissions to and inside the production site [20].The transport distance is assumed to be max.400 km.The electricity emissions are assumed to be 12.78 g CO 2 eq/MJ electricity from the Swedish grid [38].The biomass heating value is assumed to be 12.68 MJ/kg on an LHV basis with a moisture content of 40%.The sulphur recovered in the Claus recovery unit was neither included in the efficiency calculations nor in the GWP 100 impact.The actual methanol synthesis emissions are adapted from Trop et al. [21], and corrected to 55.3% efficiency from the 57.7% used for Case B.
The GWP 100 for methanol from renewable wood biomass is −0.914 kg CO 2 eq/kg methanol, and the corresponding energy use is 17.7 MJ/kg methanol.The inventory results for this case are shown in Table 2F.The negative GWP 100 for methanol production derives from the use of renewable wood biomass and is better than that of power production from 50:50 co-gasification of coal and biomass (Alternative 1F in Figure 6, Trop et al. [21]).Part of the impact is due to the very low electricity emissions in Sweden.
Biomass drying has a significant impact on the overall energy balance.The gasification unit produces excess heat, increasing the cooling demand by 60% when drying of the biomass is not integrated with the process.Oxygen gasification produces an exit gas with a heating value of 10-14 MJ/Nm 3 [22].A humidity increase from 40% to 50% of the wood feedstock reduces the biomass-to-methanol overall efficiency from 55% to 50% [23].An alternative for Case F is methanol production from biomass/black liquor integrated with a PPM.According to [24], integration of methanol production with a PPM improves the system efficiency by 7%.
The methanol yield can be increased by adding hydrogen to the syngas prior to methanol synthesis.The impact on the GWP 100 depends on how the H 2 is produced.The source of electricity is vital; power produced in a condensing coal power plant emits 925 kg CO 2 eq/MWh and 419 kg CO 2 eq/MWh in an NG combined cycle plant [23].Using electricity produced from natural gas would, in Case F, change the electricity emissions to 116.39 g CO 2 eq/MJ from the 12.78 g CO 2 eq/MJ used for base case F in Table 2F.Correspondingly, the GWP 100 for methanol production from wood biomass would change from −0.914 to −0.418 kg CO 2 eq/kg methanol (Alternative 2F in Figure 6).
The difference between similar technologies (Cases D and E) is less than one would anticipate between biogenic and fossil CO 2 .The absolute difference in the GWP 100 impact between Case D and Case E is 0.237 kg CO 2 eq/kg CH 3 OH.50.2% of this can be attributed to the efficiency difference with the rest originating from the different allocation methods (economic and energy) and the benefits of biogenic CO 2 .

Carbon dioxide equivalent abatement cost and methanol production cost
A preliminary CO 2 eq abatement cost estimation for cases A, B, C, D, E and F was performed using the CRF, as defined in Section 2. The investment and operational cost data, the annual production capacity of methanol and avoided amount of CO 2 eq compared to the reference value are shown in Table 4.The reference value for all cases is 2.6 kg CO 2 eq/kg methanol (i.e. the coal-based process) [35].The methanol from gasification of coal (Case B) does not generate avoided emissions.The CO 2 abatement cost is shown in Figure 4, and the production cost of methanol in Figure 5.
The two discount rates indicate the sensitivity of abatement and production cost for changes in the investment cost and profitability rate required.The production cost of methanol with a 5% discount rate varied between €197 and €305 per metric ton of methanol for most of the cases.The only exception is methanol in co-production with corn ethanol (Case D) with €810 per metric ton of methanol.This is explained by the high operational cost combined with a small production capacity.Correspondingly, with a 10% discount rate, the variation was from €220 to €321 per metric ton of methanol, and accordingly Case D achieved €844 per metric ton of methanol.The lowest production cost is for methanol produced from wood biomass inside a PPM.At current methanol prices of US$500 per metric ton of methanol, all cases with the exception of case D can be considered viable options.The example calculations of discounted abatement cost and discounted production cost of methanol are given in the Supplementary material.

Sensitivity and limitations of results
The results are impacted by the robustness of the assumptions made, the quality of the obtained data, and the limitations arising from the methodology used.The ALCA method used in this evaluation reflects the production systems as they are described in their respective references.In a sense, this is a clear limitation, and bounds the results to the technologies, efficiency and allocation factors, feedstock resources, and geographical locations used in the assessment.
The uncertainty of results attributable to different efficiencies and initial process data for LCI tables are partly normalized using one energy value for methanol in all cases to extract the GWP 100 impact.This gives a variation between 1.7% and 12.8%, and reflects the uncertainty of results.The energy use for Case D and E was estimated using efficiency values.
Alternative cases for GWP 100 impact are listed in Table 5 and the sensitivity of results is illustrated in Figure 6.The reference value for all the cases is methanol production from coal with a GWP 100 of 2.6 kg CO 2 eq/kg CH 3 OH.

Discussion
The evaluation of optimal system solutions requires, in addition to environmental and energy efficiency issues such as GWP 100 impact and energy use, an economic evaluation of the technologies.Instead, we have included a preliminary estimate of the CO 2 eq abatement cost for the GWP 100 reduction compared to a reference case of methanol production from coal.In addition, we evaluated the preliminary production cost of methanol for each case.The GHG emissions connected to land use change are not considered in this study.We selected attributional LCA in combination with GHG abatement and methanol production cost evaluation as an assessment tool.Many other tools exist, and one of the more complex methods is a 3-dimensional assessment including LCA, life cycle costing (LCC) and social life cycle assessment (S-LCA) [41] The results of this study implicate the importance of comparing different solutions in a uniform manner.Especially, the case of tri-reforming (Case C: Producing energy by steam reforming, using the flue gas together with natural gas for methanol production) illustrates that there is a trade-off between the positive GWP 100 impact and overall resource efficiency (Table 2C).It can be argued that the energy or economic allocation of both resource and energy use, in this case, between net electricity, district heat and methanol gives overall benefits.Therefore, when assessing co-production facilities, it is vital to remember that overall feasibility depends on the allocation of environmental and energy loads between different products and side streams.
Our results are comparable with previous research, although we have not identified similar research with an assessment of a wide range of technologies concentrating on GWP 100 and energy use.Compared with similar cases in China [2], where methanol production from methane has a conversion efficiency of 59% on an LHV basis, and that from coal 39% on an LHV basis, our results show lower GWP 100 and energy use due to the efficiency difference.The results for methanol from CO 2 and H 2 (Cases D and E) agreed well with the results of Van-Dal and Bouallou [42], who reported a negative emission of 1.15 kg CO 2 eq/kg methanol in the synthesis part without deducting emissions in CO 2 and H 2 production.The tri-reforming route (Case C) practically halved the GWP 100
impact in comparison with Case B, and this is supported by the results of Zhang et al. [14] for methane-based tri-reforming with flue gas input, although the processes themselves are not comparable.The wood biomass route (Case F) with 55.3% overall efficiency is in line with the 53.9% reported by Van Rens et al. [18].Thermodynamically feasible solutions might not always be environmentally optimal.Our results (Case D and E) are comparable with the findings of Dumont et al. [43] concluding (1) that GWP values for CO 2 exceeding 0.276 kg CO 2 eq/kg CO 2 make CH 4 a better option than CO 2 for methanol synthesis, and (2) the GWP for the H 2 used in the synthesis and values below 0.53 kg CO 2 eq/kg H 2 overrule any other carbon source than captured/recycled CO 2 .Environmentally feasible solutions are not always economically viable solutions, as demonstrated by our case of methanol in co-production with corn ethanol (Case D).
Transporting resource materials and intermediates between different geographic locations has a considerable impact on the overall GWP 100 performance (Case B).Especially, the leakage of methane, carbon dioxide and nitrous oxide are related to the transport distance.Similarly, the emissions of power generation easily change the positive impacts to unexpectedly high negative results, as in the case of hydrogen production (Case D).This study shows that results of the LCA are altered drastically by changing the raw material production route or other assumptions.The resource for H 2 production and GHG load of electricity in electrolysis are both factors impacting the GWP 100 .
Global climate change has no regional or local environmental boundaries.In the case of methanol production, CO 2 is the main contributor to the impact of GHG.Positive development is ongoing also in enhancing the current production route of methanol from NG.A recent modelling study showed a net potential for CO 2 emissions reduction of 0.36 ton CO 2 /ton methanol [44].Producing methanol by gasification from wood biomass (Case F) can compete in positive GWP 100 impact with co-firing coal and biomass for electricity and district heat production (Alternative 1F).Yet, wood biomass or any other biomass has many competitive uses.The production of methanol from wood biomass would compete with the production of FT liquids, other chemicals and with the standard products of PPMs.The competition in raw materials makes the captured CO 2 routes more suitable for the production of methanol, provided that the production costs remain competitive.However, with limited fossil or renewable carbon resources, it is more feasible to produce chemicals than to use the same resource as fuel.Storing renewable energy resources in methanol produced from hydrogen and captured CO 2 closes the carbon loop, and facilitates the circular economy approach in the production of valuable bulk chemicals.Growing demand in methanol is mainly in Asia Pacific where constraints in biomass supply may restrict the viability of biomass based methanol production.
However, there are several reasons and barriers that slow down the implementation of low carbon methanol processes and circular economy approaches in the production of bulk chemicals.These include the relatively long cycle of introducing new technological processes, subsidizing of the fossil resource base, and normative and legislative restrictions in the introduction of new raw materials and processes.

Conclusions
The life cycle environmental performance of methanol, on a cradle-to-gate basis, for six different production routes has been assessed using an attributional LCA.The assessment is limited

Figure 1 .
Figure 1.(a) World methanol production capacity of 110 million metric tons by region in % in 2018 [3]; (b) World methanol demand of 90 million metric tons by region in % in 2018 [4].

Figure 2 .
Figure 2. A simplified flowchart of the six different methanol production routes.

Figure 5 .
Figure 5. Methanol production cost in € per metric ton of methanol.

Figure 6 .
Figure 6.Sensitivity of GPW 100 impact assessment in kg CO 2 eq/kg methanol.

Table 1C .
Methodological choices for Case C.

Table 1D .
Methodological choices for Case D. options Biomass harvested and preprocessed in Nebraska.CO 2 feed extracted from corn ethanol process unit and methanol produced in southeast Nebraska.

Table 2A .
LCI results for methanol from fossil natural gas (Case A).

Table 2B .
LCI results for methanol from coal (Case B).

Table 2C .
LCI results for methanol by tri-reforming integrated with power production (Case C).

Table 2D .
LCI results for methanol in co-production with corn ethanol (Case D).

Table 2E .
LCI results for methanol from captured CO 2 and H 2 (Case E).

Table 2F .
LCI results for methanol from wood biomass (Case F).

Table 4 .
Input data for CO 2 eq abatement cost and methanol production cost calculations.

Table 5 .
Alternatives used for the assessment of sensitivity in kg CO 2 eq/kg methanol.