BIOMASS FOR GREENHOUSE GAS EMISSION REDUCTION 1,2 (BRED)

Size: px
Start display at page:

Download "BIOMASS FOR GREENHOUSE GAS EMISSION REDUCTION 1,2 (BRED)"

Transcription

1 BIOMASS FOR GREENHOUSE GAS EMISSION REDUCTION 1,2 (BRED) Abstract D.J. Gielen, A.J.M. Bos, T. Gerlagh, ECN-Policy Studies PO Box 1, 1755 ZG Petten, The Netherlands tel ; fax ; This paper discusses preliminary results of the BRED project (Biomass for greenhouse gas emission (REDuction). This project focuses on the optimal allocation of biomass for energy and materials production in order to reduce greenhouse gas emissions. The Western European MATTER1. systems engineering model has been applied for this analysis. The results show the important potential of biomaterials, a feature that is often neglected in greenhouse gas emission reduction studies. Moreover, increased use of biomaterials facilitates the use of bioenergy from by-products and waste biomaterials. The model calculations show that new technologies play a key role for future energy and material flows. The study can be considered as a case study that shows how systems engineering models can be used for ex-ante material flow analysis and for development of policies towards sustainability. 1. Introduction Any progress towards sustainable development strongly depends on the availability of methods to describe and analyse the metabolism of human societies. Priorities for substitution and dematerialisation measures can only be set effectively and efficiently, if the status and the current trends of material flows are known. Material flow accounting (MFA) methods provide a tool to monitor and model material flows. However, current MFA methods are either focusing on historical trends or they focus on the current situation. A change to sustainable development will take decades, and many roads are feasible. Ex-ante MFA is required for planning purposes. A tool is required that can select the optimal path towards sustainability. This tool must consider the interactions between improvement options. Within a period of decades, product and service demand will change, technology will change, and resource availability will change. These changes must be considered in the analysis. This paper discusses a case study, where such an analysis tool has been applied. The MATTER 1. model is an integrated energy and materials systems model for Western Europe [1,2]. The cases study focuses on the introduction of biomass in the economy in order to reduce greenhouse gas (GHG) emissions. GHG emissions constitute one of the most important challenges to sustainable development in the next decades, and can serve as an indicator for the sustainability discussion. This paper will show that significant GHG emission reduction can result in much more biomass use in the economy. Because biomass can be produced as a renewable resource, this introduction can simultaneously reduce the consumption of non-renewable natural resources. 1 Paper prepared for the Conaccount meeting Ecologizing Societal Metabolism; Designing Scenarios for Sustainable Materials Management, 21 November 1998, Amsterdam 2 The BRED study is partially funded by the European Union in the framework of the Environment and Climate Research Programme 1

2 2. Method The MARKAL linear programming model was developed 2 years ago within the international IEA/ETSAP framework (International Energy Agency/Energy Technology Systems Analysis Programme). MARKAL is an acronym for MARKet ALlocation. A MARKAL model is a representation of (part of) the economy of a region. The economy is modelled as a system, represented by processes and physical and monetary flows between these processes. These processes represent all activities that are necessary to provide products and services. Many products and services can be generated through a number of alternative (sets of) processes. The model contains a database of several hundred processes, covering the whole life cycle for both energy and materials. The model calculates the least-cost system configuration. This system configuration is characterised by process activities and flows. The database of processes and the constraints for individual processes and for the whole region are defined by the model user. Constraints are determined by the demand for products and services, the maximum introduction rate of new processes, the availability of resources, environmental policy goals for energy use and for emissions etcetera. Processes are characterised by their physical inputs and outputs of energy and material, by their costs, and by their environmental impacts. Environmental impacts are endogenised in the process costs and the costs of energy and material flows between processes. MARKAL is a dynamic model. The time span to be modelled is divided into nine periods of equal length, generally covering a period of decades. Within such a time horizon, technological change will be a major driving force for a changing systems configuration. Changing technology can be modelled through changing parameters in time for individual processes. Another option is the modelling of the future availability of new alternative processes. The model is used to calculate the least-cost system configuration for the whole time period, meeting exogenously defined product and service demands and meeting emission reduction targets. This optimisation is based on a so-called perfect foresight approach, where all time periods are simultaneously optimised. Future constraints are taken into account in current investment decisions. PRIMARY ENERGY FINAL ENERGY MATERIALS OIL GAS COAL URANIUM RENEWABLES REFINERIES POWER PLANTS OIL PRODUCTS ELECTRICITY HEAT GAS COAL INDUSTRY MATERIALS/ PRODUCTS RESIDENTIAL TRANSPORTATION WASTE MATERIALS RENEWABLES OTHER INCINERATION ENERGY FLOWS MATERIAL FLOWS Figure 1: Generic MARKAL energy and materials system model structure [1] 2

3 A MARKAL model is a representation of (part of) the economy of a region. The economy is modelled as a system, represented by processes and physical and monetary flows between these processes. These processes represent all activities that are necessary to satisfy a fixed demand for products and services. Many products and services can be generated through a number of alternative (sets of) processes. The model contains a database of several hundred processes, covering the whole life cycle for both energy and materials with GHG relevance (Figure 1). The model calculates the least-cost system configuration. This system configuration is characterised by process activities and flows. The database of processes and the constraints for individual processes and for the whole region are defined by the model user. Constraints are determined by the demand for products and services, the maximum introduction rate of new processes, the availability of resources, environmental policy goals for energy use and for emissions etcetera. Processes are characterised by their physical inputs and outputs of energy and material, by their costs, and by their environmental impacts. The model covers more than 25 energy carriers and 125 materials. More than 5 products represent the applications of these materials. 3 categories of waste materials are modelled. These materials are characterised by their physical characteristics and by their quality. Different GHG emission penalties have been analysed. These penalties are shown in Figure 2. The base case is run without penalties. In the emission reduction cases, the penalties increase from zero in the year 2 to their maximum level in 22 and stabilise afterwards. [ECU/T CO2] 2 15 "2 ECU/T CO2" " ECU/T CO2" "5 ECU/T CO2" "2 ECU/T CO2" [YEAR] Figure 2: GHG emission penalty scenarios The energy system model structure and energy system model input data are discussed in separate volumes The following materials model discussion is on an intermediate level of detail, focusing on the system structure. Documentation for input data and some analysis results can be found on the Internet [2]. An overview of the model structure is provided in [1]. 3

4 3. Model structure Figure 3 and figure 4 show the model structure that has been used for this study (split into wood crops and other agricultural crops, respectively). A discussion of model input data is provided in [3]. FOREST LAND USE ROUND- WOOD SAWING WOOD CHIPS/ RESIDUES BOARD PROD. ACETYL. PLATONIS. PULPING PULP PAPER PROD. SAWN WOOD BOARDS THW SUBST. PAPER USE (TIME LAG) WASTE WOOD WASTE PAPER INCINERATION DISPOSAL ELECTRICITY HEAT CHEMICAL CONVERSION CHEMICAL CONVERSION CELLOPHANE VISCOSE PUR LIGNINE COMBUSTION/ GASIFICATION FLASH PYROLYSIS ELECTRICITY HEAT ETHYLENE ETC. SURPLUS AGRICULTURAL LAND USE PER REGION POPLAR/EUCALYP. WOOD CHIPS OTHER CROPS Figure 3: Model structure for wood FLASH PYROLYSIS CHEMICAL CONVERSION GASIFICATION/ SYNTHESIS PYROLYSIS PYROLYSIS POWER PLANT TE-UNIT WOOD OVEN WOOD STOVE HTU M PLANT PLANT OIL LIGNINE HYDROTREATING BUTADIENE PUR ACETIC ACID CARBON BLACK CHARCOAL ELECTRICITY HEAT PROCESS HEAT RESIDENTIAL HEATING BIODIESEL M STRAW BOILER GREENHOUSE HEATING RESIDUAL STRAW STRAW HTU M PLANT OIL HYDROTREATING BIODIESEL M PLANT LIGNINE POPLAR/EUCALYPT. SURPLUS AGRICULTURAL LAND USE PER REGION MISCANTHUS SORGHUM SUGARBEET WHEAT ALGAE RAPE SUGARS RAPESEED OIL FERMENTATION FERMENTATION FERMENTATION PLANT HYDROTREATING ESTERIFICATION CONVERSION BUTANOL/ACETONE I-PROPANOL PHB/PHV BIODIESEL LUBRICANTS MARIGOLD SOLVENTS/PAINT Figure 4: Model structure for agricultural crops 4

5 4. Results The use of bioenergy is detailed in Figure 5. In the base case, only heat is produced. In the 5 ECU/t case, some ethanol production emerges. Ethanol production shows a particularly strong growth at higher emission penalties. In the ECU/t case, biomass is introduced for power production. Methanol is introduced on a large scale in the transportation market at a penalty level of 2 ECU/t. The HTU process is not applied in this case, but emerges on a large scale at penalty levels above 2 ECU/t CO 2. [MT BIOMASS/YEAR] M WOOD/STRAW ALGAE RAPESEED RESIDUES 2 BC ECU/T CO2 5 ECU/T CO2 2 ECU/T CO2 Figure 5: Bioenergy use, split into applications, 23 The use of biomaterials is specified in Figure 6. Biomass is introduced for production of petrochemical intermediates at a penalty level of 5 ECU/t CO 2. Flash pyrolysis of biomass for ethylene and butadiene production are introduced. The fermentation of I-propanol, phenol production from lignine, lignine for PUR production, carbon black and acetic acid (via synthesis gas) for petrochemicals are additionally introduced. Palm kernel oil is introduced at a penalty level of 2 ECU/t CO 2. Butadiene, natural rubber, the fermentation of butanol and acetone, Marigold flower resins and synthetic lubricants are not introduced on a large scale. Ethanol dehydrogenation for ethylene production is not introduced in any case. Regarding fiber applications, viscose for substitution of synthetic fibres is not introduced. Acetylated wood and PLATOnised wood are introduced from 5 ECU/t upward. Biopol and cellophane are not introduced in any of the emission reduction cases. At the 2 ECU/t penalty level, charcoal is introduced for iron production. The use of structural wood products for the building and construction market increases also at this high penalty level. However, its growth is comparatively small. A significant fraction of the energy content of biomaterials is recovered. A fraction (e.g. in flash pyrolysis) is converted into liquid and gaseous fuels during the production process. Residuals from wood sawing are applied for board materials and for energy purposes. The bulk of the petrochemical products, waste paper and used wood products are ultimately used for energy recovery. Note in Figure 6 the increasing pulp production in the 2 ECU/t CO 2 case: waste paper is used for energy recovery instead of being recycled. 5

6 [MT BIOMASS/YEAR] WOOD PRODUCTS PULP BIOPLASTICS LUBRICANTS PETROCHEMICALS CHARCOAL 2 BC ECU/T CO2 5 ECU/T CO2 2 ECU/T CO2 Figure 6: Biomaterial use, split into applications, 23 The discussion above focused on aggregated results. However, these results depend to a large extent on the model input parameters. The analysis below focuses on two parameters that are thought to be of key importance for the future of biomass: land availability ethanol production from linocellulose crops The following analysis shows the sensitivity of model results for variations in model input parameters that characterise these issues. From a methodological point of view, these case studies show how the relevance of uncertainty can be analysed with sensitivity analysis. Similar analyses have been done for other key data within the model, but will not be discussed in detail. Sensitivity analysis for land availability The availability of surplus agricultural land in Western Europe in the next three decades is highly uncertain. Two scenarios have been considered with a land availability of 2 million hectares and 22 million hectares, respectively. The 22 million hectare scenario has been used in the reference scenario model calculations, the 2 million hectare represents the sensitivity analysis. Higher land availability has not been analysed because the results for the 22 million hectare case show that land availability poses no constraint at penalty levels that are currently considered to be reasonable (up to ECU/t CO 2). The maximum land availability level is in both scenarios reached in the year 21. This implies a drastic change of European agricultural policies, especially for the 22 Mha case. Figure 7 shows the impact of land availability on the use of biomass for energy purposes. The impact proves to be very significant at penalty levels of more than 2 ECU/t CO 2. At lower penalty levels, around Mt biomass from wood residues and agricultural residues is used for energy purposes. The gap between both scenarios increases to around 25 Mt biomass at penalty levels of 2 ECU/t upward. The main difference is a significantly reduced production of ethanol from lignocellulose crops. Figure 8 shows the impact of land availability on the use of biomass for materials production. The impact is not as pronounced as for energy production. The difference between both land availability cases is approximately 5-75 Mt biomass. The difference can mainly be attributed to a reduced biomass use for production of petrochemical products like ethylene. 6

7 BIOMASS FOR ENERGY [MT/YEAR] ECU/T 5 ECU/T ECU/T 2 ECU/T 5 ECU/T PENALTY [ECU/T CO2] 22 Mha 2 Mha Figure 7: Biomass use for energy, depending on land availability, increasing GHG penalties, 23 BIOMASS FOR MATERIALS [MT/YEAR] ECU/T 5 ECU/T ECU/T 2 ECU/T 5 ECU/T PENALTY [ECU/T CO2] 22 Mha 2 Mha Figure 8: Biomass use for materials, depending on land availability, increasing GHG penalties, 23 The difference between both scenarios increases to approximately MT CO 2 equivalents at higher emission penalties (from ECU/t CO 2 upward). This difference represents 5-1% of the total GHG emissions at these penalty levels or approximately 3% of the GHG emissions in the base case. This sensitivity analysis shows that land availability is of key importance for bioenergy production, but to a much lower extent relevant for biomaterials production. Whether or not large amounts of bioethanol are produced depends to a high extent on the availability of large areas of 7

8 surplus land. The selection of crops remains largely the same for the high and low land availability cases, with a dominance of high yield crops. Sensitivity analysis for ethanol production efficiency The results for the reference case show that the ethanol production represents a very significant part of bioenergy production if GHG emission penalties are introduced. In a second sensitivity analysis, the efficiency of ethanol production has been analysed in more detail. This parameter has been selected because the model assumptions regarding conversion efficiencies are thought to be fairly optimistic. The reference case assumes an energy efficiency (carbon in biomass to ethanol, excluding additional steam inputs for distillation etc. and excluding consideration of lignine inputs and by-products) of 74% for sugar and starch, 74% for cellulose and 67% for hemicellulose. This includes pretreatment, hydrolysis, and fermentation losses, and is quite close to the theoretical yield of 79%. Given that the process is not yet proven on a commercial scale and given the complex process route, efficiencies still are uncertain. The results in Figure 9 show that total biomass use for energy production is not affected by the lower ethanol production efficiency. However, the fraction of biomass use for ethanol production is significantly reduced. Instead, methanol production increases and the use of biomass for electricity production increases. This result shows that the production of transportation fuels from biomass (both quantity and fuel type) depends critically on the efficiency of the production processes. The parameters of this process should be evaluated critically. [MT BIOMASS/YEAR] M WOOD/STRAW ALGAE RAPESEED RESIDUES 5 REFERENCE EFFICIENCY 25%< Figure 9: The sensitivity of bioethanol production for the conversion efficiency of crops to ethanol, Conclusions A significant fraction of energy and materials consumption can theoretically be covered with biomass. However economic considerations pose a constraint for the full exhaustion of this potential. The study shows that GHG emission reduction policies can increase the use of renewable biomass resources. MARKAL modelling results show that Western European biomass availability is no constraint at emission penalty levels up to 5 ECU/t CO 2, if a land availability of 22 Mha is assumed. As a 8

9 consequence, no competition occurs between bioenergy and biomaterial applications. On the contrary: the production of biomaterials results in an increased availability of process waste and post consumer waste that can be used for energy recovery. Only at emission penalty levels from ECU/t CO 2 upwards, a trade-off between both applications will occur. At penalty levels up to ECU/t, materials applications dominate energy applications. At higher penalty levels, energy applications dominate. This can be attributed to the combination of higher energy market volumes and the features of competing emission abatement strategies in energy and material markets. The conclusion for biomass strategy analysis is that materials applications must also be considered for the future assessment of bioenergy. Future land availability is a key parameter with high uncertainty. The sensitivity analysis for individual model parameters showed that bioethanol production from lignocellulose crops is a key technology in the analysis. The parameters for this technology determine to a large extent how the biomass should be applied. However the total amount of biomass that is applied is relatively independent of these assumptions. The BRED case study for biomass shows how techno-economic systems engineering models can be used for the ex-ante analysis of sustainability strategies. The consideration of technological change, resource availability and market volumes proves to be crucial for the proper analysis of future sustainability potentials. If financial aspects and policy goals are integrated into the analysis, the resulting tool can provide insights that are relevant for policy makers. This analysis is a pilot study for the BRED project (Biomass for greenhouse gas emission REDuction) within the framework of the Environment and Climate programme of the European Union. This project is a joint effort of the National Technological University in Athens, the Bundesforschungsanstalt für Holzwirtschaft in Hamburg and ECN. The results will be used for R&D strategies and future European research programming. More detailed results will become available in References [1] D.J. Gielen, T. Gerlagh, A.J.M. Bos: MATTER 1.. A MARKAL Energy and Materials System Model Characterisation. ECN-C Petten, September [2] Input data available via Internet: [3] D.J. Gielen, T. Gerlagh, A.J.M. Bos: Biomass for Energy or Materials? A Western European MATTER 1. MARKAL model Characterisation. ECN-C Petten, forthcoming. 9