IMPROVING LOGISTICS FOR BIOMASS SUPPLY FROM ENERGY CROPS IN EUROPE: MAIN RESULTS FROM THE LOGIST'EC PROJECT Benoît Gabrielle1,*, Truls Flatberg2, Aurélie Perrin1, Julie Wohlfahrt3, Thor Bjørkwoll2, Inés Echevarría Goni4,
Raimo van der Linden5, Chantal Loyce6, Elise Pelzer6, Giorgio Ragaglini7, Ian Shield8, Nicola Yates8 1: AgroParisTech, INRA, UMR1402 EcoSys, Thiverval-Grignon, France. 2 : SINTEF Technology and Society, Department of Applied Economics, Trondheim, Norway. 3: INRA, SAD – ASTER, Mirecourt, France. 4: CENER,National Renewable Energy Centre, Biomass Department, Sarriguren, Navarra, Spain 5: Energy Research Centre of The Netherlands (ECN), Biomass & Energy Efficiency Unit, Bioenergy Group, Petten, NL. 6: AgroParisTech, INRA, Agronomy Joint Research Unit, Thiverval-Grignon, France. 7: Scuola Superiore Santa Anna, Institute of Life Sciences, Field Crops and Bioenergy, Pisa, Italy. 8: Rothamsted Research, Agro-Ecology, Harpenden, Herts., United Kingdom. * corresponding author: INRA, AgroParisTech UMR EcoSys, F-78850 Thiverval-Grignon, France. E-mail:
[email protected] Phone; (+33) 1 30 81 55 51 Fax: (+33) 1 30 81 55 63 ABSTRACT: Cost-efficient, environmental-friendly and socially sustainable biomass supply chains are urgently needed to achieve the 2020 targets of the Strategic Energy Technologies-Plan of the European Union, which are likely to be impeded by the potential scarcity of lignocellulosic biomass from agriculture. Innovative techniques for crop management, biomass harvesting and pre-treatment, storage and transport offer a prime avenue to increase biomass supply while keeping costs down and minimizing adverse environmental impacts. The LogistEC project aimed at developing new or improved technologies for all steps of the logistics chains, and to assess their sustainability at supply-area level for small to large-scale bio-based projects. It encompassed all types of lignocellulosic crops: annual and pluri-annual crops, perennial grasses, and short-rotation coppice, and included pilot- to industrial-scale demonstrations, in particular around 2 existing bioenergy and biomaterials value-chains in Europe (in Eastern France and Southern Spain). This paper reviews the main results obtained in the project on the main components of logistics chains, regarding feedstock production systems, harvesting and post-harvest handling, storage, densification and pretreatment of biomass. The information and tools delivered by the project provides a first step to guide in incremental improvements as well as systemic changes in biomass feedstock supply chains from energy crops. Keywords: biomass supply chains, energy crops, logistics, sustainability assessment, optimization, pretreatment, harvesting 1
INTRODUCTION
Cost-efficient, environmental-friendly and socially sustainable biomass supply chains are urgently needed to achieve the 2020 and 2030 targets of the SET-Plan, which are likely to be impeded by the high supply cost and potential scarcity of agricultural and forestry biomass in Europe. Innovative techniques for crop management, biomass harvesting, storage and transport offer a prime avenue to increase biomass supply while keeping costs down and minimizing adverse environmental impacts [1]. Significant challenges for the deployment of optimal feedstock supply chains include the scattered and bulky nature of biomass, its high moisture content, and potential for degradation during storage and transport. Avenues to improve feedstock logistics systems thus include [1]: An increased efficiency of harvesting, preprocessing, and transport equipments, The densification of biomass prior to transport, with or without thermal pre-treatment, to produce a carrier that can be handled in already existing transport, handling and storage equipments, An optimal integration of system components, Lower impacts on road transport by a reduction in transportation distances or a combination with other transportation means,
Adaptation of logistics operations (in particular storage) to high-moisture biomass, or enhancement of drying conditions, Feedstock production systems providing yearround supply with high productivity per unit area and consistent biomass properties.
The Logist'EC project funded by the 7th Framework Programme of the European Commission. aimed at addressing these challenges. Its main tenet was that stepwise developments in the components of biomass logistics would lead to an optimization at supply-chain level, making the most of the developments carried out in the individual components. Chain integration via modelling allows for an optimization of the supply chains by combining the most appropriate technologies, as driven by the requirement of the biomass conversion processes or end-uses. The development of a comprehensive framework for sustainability assessment, encompassing economic, environmental and social criteria is intended to provide guidance in the chain optimization and to propose solutions tailored to various possible end-users (whether private stakeholders or policy makers) [2]. 2
MATERIALS AND METHODS
The projects combines incremental technological improvements in the supply chains and holistic assessment (Figure 1). As a basis for supply chain integration and full-scale demonstrations, Logist'EC involves 2 case-studies based on currently-operating value-chains in Europe. The first one involves a 16 MW bio-electricity plant running on agricultural and forest biomass in Miajadas (Southern Spain), and the second one a cooperative (Bourgogne Pellets) based in Eastern France which develops perennial grasses (miscanthus and switchgrass) on commercial farms, to make pellets or biomass chips for various local and national markets. The area currently managed by the cooperative is about 500 ha, and the Logist'EC project has been looking at ways to improve the logistics and scale up the production, which is currently limited by market demand. Regarding the individual chain components presented on Figure 1, the research within Logist'EC involved the benchmarking of currently commercial technologies and the development and testing of novel options such as legume intecrops for feedstock production systems, improved cutting heads for harvests or wet torrefaction of grasses to recover nutrients while densifying the biomass. A holistic framework is also being developed to integrate chain components and assess their sustainability in terms of environmental, economic and social impacts. It enables an economic optimization of the supply chains and an overall sustainability assessment of these solutions, making the most of the progresses achieved in the logistics components. The location of energy crops in relation to biomass conversion units is also crucial to the logistics, determining the transportation distances and the relevance of intermediate storage points [2]. A model with a high spatial resolution was developed to predict the most probable location of future miscanthus fields, in the vicinity of the cooperative, based on landscape features as well as yield potentials and farmers' survey data. It produces maps of miscanthus plots and yields for a given production target and maximum transportation radius from the plant. In terms of environmental sustainability, biomass logistics chains were evaluated using Life-Cycle Assessment (LCA), which is widely used for regulatory purposes for bioenergy chains (and liquid biofuels in particular). LCA is a multi-criteria and quantitative methodology which aims at comparing products over their entire life-cycle while avoiding burden shifting between impact categories or supply chain components [3]. The two main end-products supplied by Bourgogne Pellets (Miscanthus chips and Miscanthus bales) were evaluated for the year 2013. They were transported over an average distance of 10 km from the agricultural fields to the conversion plant.
Figure 1. Schematic of biomass supply chain components and main technological options tested during the Logist'EC project. 3
RESULTS
3.1 Feedstock production
Figure 2. Yield ratio for crop species directly compared to Miscanthus Giganteus as a reference crop. A meta-analysis was carried out to compare the yields of the major energy crops worldwide. A database was built based on published literature and the project partners' own data sets to compare the yields of energy crops using site-specific measurements. The data base included 865 yield data pertaining to 36 different crop species. Comparisons between crops were either direct (grown in the same site - Figure 2) or indirect (using other crops grown on the same site as intermediate benchmarks). Both comparisons yielded similar results. Miscanthus x giganteus appeared as one of the most productive crops, only superseded by Arundo donax and Pennisetum purpureum [4]. However the latter were only studied in a limited number of sites. Data related to the environmental impacts of energy crops were also searched but no meta-analysis could be conducted for lack of sufficient data. 3.2 Harvesting of energy crops Currently-commercial harvesting systems for willow SRC were first evaluated based on field observations of harvests in the UK. Forage harvesters producing wood chips directly, in the field, appear the most efficient
technology at the present time (Table I). This is subject to evaluating the drying options for wet wood chips. The whole rod harvester (loose rods) is also an efficient method for clearing the field, but account must be made for processing the material post-harvest. This is the subject of on-going work on storage and drying of whole rods. The Bio-baler is the least attractive option and this may be reflected in the decline in interest in the machine since 2011.
the crops. The main results are as follows: Herbaceous biomass requires lower dry torrefaction temperatures than woody biomass The Torwash process is capable of greatly reducing the ash content of herbaceous biomass Arundo donax Torwash fuel pellets comply with ENplus A1 standard (the most stringent standard for white wood pellets), while other Torwash fuels have the same potential
Table I: Summary of the performance of the selected SRC Willow harvesting machines.
Conventional densification routes (in the absence of thermal pre-treatments) were also intestigated for all energy crops. The emphasis was put on briquetting (and to some extent pelletizing) rather than the conventional baling technologies, which are already available, since briquetting as well as pelletizing are expected to result in denser materials more suitable for industrial use. Briquetting involves the pressing of biomass chips into a conical pipe (die) with an diameter of around 85 mm at the outlet. It produces cyclinder-shaped briquettes of biomass with a density similar to that of pellets (around 1000 kg DM m-3). The test on energy crops revealed an interesting trade off between briquette density and power consumption. In terms of briquette quality, lower die temperature generally created more friction and increased backpressure during densification, leading to a higher density briquette. On the other hand, higher temperature leads to lower power consumption.
Regarding grasses, different configurations were tested in relation to the timing of harvest and the status of the crops. In particular, a single-pass system to simultaneously cut, shred and bale the biomass was developed by a company based in Italy (Nobili) and compared with other systems involving up to three passes. The single-pass system proved interesting for fully senesced biomass and/or in dry environments (see [5] for more details). Different timings were also tested for grasses, such as double harvesting for Giant Reed in Italy (in early summer and autumn - [6]), or early harvest in autumn for Miscanthus and switchgrass (see [7] for more details). These options proved feasible and should provide benefits towards year-round supply schemes and reduced storage requirements, but are challenging in terms of nutrient exports and re-growth potential in the long-run for crops. Regarding the former point, a possible solution would consists in using a thermal pre-treatment that recovers nutrients and makes it possible to recycle them back to the crops (see below section). This option is currently tested in the project. 3.3 Pre-treatment of biomass Various types of biomass were tested in wet and dry torrefaction. Dry torrefaction involves heating biomass to a temperature in the 240 to 320°C range to increase its heating value and hence energy density, as well homogeneize its quality. Most of the work on this route has focused on woody material, and Logist'EC aimed at investigating the relevance of this process to grasses. For the latter, a potentially more suitable alternative was also explored which combined heating with washing in water. This wet route, branded 'Torwash', has the advantage of being removing alkali compounds which may be a problem for thermochemical application of herbaceous biomass, and makes it possible to recover nutrients from the solid biofuel end-product and to recycle them back to
3.4 Sustainability assessment at supply-area scale In the logistics of energy crops, the spatial location of the fields where the crops are grown with respect to the biomass processing unit is a key parameter. Information on this topic is helpful to design supply chains. A spatially-explicit model was developed to predict the location of potential new miscanthus plots based on georeferenced physical and economic information. The model uses a boosted regression tree approach and was trained on 118 parcels of land currently cropped to miscanthus around the BP cooperative in Burgundy, which is one of the case-study value-chains of Logist'EC. Figure 3 illustrates the application of the model in a scenario targeting an overall production of 30 kt/year, within a given radius around the plant (here set at 70 kms). The map shows that unfortunately the potential to establish miscanthus in the close vicinity of the pellet plant appears limited compared to more remote areas, especially to the South of the latter. This clearly impacts logistics for such higher demand scenarios and implies that bulk transport of chips can hardly be expanded with respect to the current situation.
Although the maps of Figure 3 may be used in an 'upscaling' exercise, the current locations of miscanthus field were first used to develop and test a mathematical model of the logistic chain on the Burgundy case-study [9]. The model represents the chains as a spatial network of nodes producing, storing or transforming the biomass, corresponding to miscanthus fields, intermediate storage or pick-up points, and the biomass conversion plant. Compared to previous logistic models, this one can handle uncertainty in the demand for end-products, with a monthly granularity to account for seasonal variations in biomass production, storage capacity, or demand for endproducts. The model can also be run in a deterministic setup. Simulations of the BP case highlighted the importance of uncertainties in the demand for end-products on the
Figure 3. Predicted locations of potential future miscanthus fields to produce 30 kt of biomass within a 70-km radius from the BP cooperative.
Figure 4. Results on the economic optimization of logistics for miscanthus supply in Burgundy, with or without a stochastic component in the demand of pellets from miscanthus. Miscanthus fields are harvested with different technologies, delivering either bales, chips or baled chips. The left chart shows the optimal breakdown among the technologies, while the right-hand chart shows the expected profits with the optimal solutions under different demand scenarios. optimal logistical paths (Figure 4). Depending on the seasonality of the demand for pellets from miscanthus, the most profitable technologies of harvesting and conditioning (as bales or chips) was variable. Anticipating these seasonal variations also made it possible to increase the profits of the cooperative, by adjusting the production ratio between the chips and bales, and the miscanthus area harvested. Figure 4 also illustrates for the effect of uncertainty in pellet demand on the quantities of miscanthus harvested with different technologies, producing chips, bales with long strands (which may only be pelletized) or bales with short strands (which may be sold as mulching or animal bedding material, or pelletized).
4
Figure 5. Contribution analysis for 1MJ eq. of feedstock for the two main existing logistics scenarios in 2013, and for 6 LCA impact categories from cradle-to-plant gate. From the environmental perspective, biomass logistics chains were evaluated using Life-Cycle Assessment (LCA) widely used for bio-based chains. Figure 5 presents the contribution analysis for the two main products delivered at the transformation plant in Burgundy in 2013: miscanthus chips and miscanthus bales, assuming a fixed transportation distance of 10 km between the fields and the plant. Bales correspond to a densified product directly on field, which reduces the contribution of the transport stage to climate change, acidification and water resource depletion per MJ of Miscanthus delivered at the plant gate (Figure 5). However, the high contribution of harvest and handling stages to Freshwater eutrophication and Marine eutrophication tends to reverse the ranking of products, notably due to the impact of heavier machinery requirements (a baling machine, high-power tractors and telescopic-handlers). To complete the sustainability analysis, social impacts were assessed with both quantitative and qualitative methodologies. The former aspect relied on an economic input/output (I/O) analysis of the impacts of developing biomass in both case-study areas (in Spain and France). This method has been commonly used to estimate macroeconomic impacts of industries within the national or regional economy. It was developed by the economist Leontief and describes, through symmetrical tables, the interdependencies between activity sectors within an economy [10]. Among others, it makes it possible to estimate the net effect of employment at regional level of biomass development. Preliminary results for the Burgundy case-study evidenced a net benefit in terms of job creations of establishing 400 ha of miscanthus (see [10] for more details). The qualitative approach to social impacts relied on methods from the sociology of organizations and innovation, based on interviews with local stakeholders (biomass suppliers, end-users, local authorities and agencies) in the Spanish case-study. Interviews evidenced shared concerns about the unpredictability of regulations (in particular the support scheme for power production from energy crops, which was drastically altered in 2014). The advantages of bioenergy were perceived as not sufficiently disseminated, in particular by the scientific community. Overall, there was also a consensus on the positive impact of bioenergy development (in terms of employment, added economic value, and the creation of new market for crops).
CONCLUSION
Nearly 3 years into the project, Logist'EC provided benchmarks for existing technologies along the biomass supply chain, and explored innovative options and their potential to improve logistics. Some of these options are already commercial (such as the harvesting systems), and could be demonstrated at pilot to industrial scales, while others are more medium to long-term but also have a significant potential for reducing costs and improving sustainability. Overall, there is clearly room for improving logistic chains from energy crops based on the above developments, but piecing together the different parts of the chains in the context of actual value-chains is definitively a challenge. The structure of these chains, which are relatively recent and still in search of a viable business model makes their optimization a moving target, undermined by a lack of data on the supply chains and plant facilities. Logist'EC provided a first step down this road, and its transfer to practitioners and project developers will be facilitated by a contribution in the form of data bases and simulation models to the tool box put together by the ongoing FP7 project S2Biom. 5
REFERENCES
[1] Gold, S., Seuring, S. (2011). Supply chain and logistics issues of bio-energy production. Journal of Cleaner Production, 19, 32-42 [2] Gabrielle, B.; Bamière, L.; Caldes, N.; DeCara, S.; Decocq, G.; Ferchaud, F.; Loyce, C.; Pelzer, E.; Perez, Y.; Wohlfahrt, J. & Richard, G. (2014). Paving the way for sustainable bioenergy in Europe: technological options and research avenues for largescale biomass feedstock supply Renewable and Sustainable Energy Reviews, 33, 11-25 [3] von Blottnitz, H. & Curran, M.A. (2007). A review of assessments conducted on bio-ethanol as a transportation fuel from a net energy, greenhouse gas, and environmental life cycle perspective. Journal of Cleaner Production, 15, 607-619. [4] Laurent A, Pelzer E, Loyce C, Makowski D. (2015). Ranking yields of energy crops: A meta-analysis using direct and indirect comparisons. Renewable and Sustainable Energy Reviews, 46, 41-50. [5] Ragaglini, G., Shield, I., Yates, N.E., Tozzini, C., Villani, R., Sanchez, D. (2015). Baling of Miscanthus: the effect of harvest time on productivity, consumption and biomass losses. Proc. 23Rd EU BCE, Vienna, 2015. [6] Dragoni, F, Nassi o Di Nasso, N, Tozzini, C, Bonari, E, Ragaglini, G. (2015). Aboveground Yield and Biomass Quality of Giant Reed (Arundo donax L.) as Affected by Harvest Time and Frequency. BioEnergy Research, Available online. [7] Yates, N.E., Riche, A.B., Shield, I., Zapater, M., Ferchaud, F., Ragaglini, G., Ronuci, N. (2015). Investigating the longterm biomass yield of Miscanthus Giganteus and Switchgrass when harvested as a green energy feedstock. Proc. 23Rd EU BCE, Vienna, 2015. [8] Rizzo, D., Martin, L., Wohlfahrt, J. (2014). . Miscanthus spatial location as seen by farmers:
A machine learning approach to model real criteria. Biomass and Bioenergy, 66, 348-363. [9] Kaut, M., Egging, R., Flatberg, T., Tolstad Uggen, C. BLOMST – An Optimization Model for the Bioenergy Supply Chain (2015). In Eksikoglu, S.D., Rebennack, S.D., Pardalos, P. (Eds.), 'Handbook of Bioenergy', Springer, forthcoming. [10] De la Rua, C., C. de la Rúa, Lechón, Y., Perrin, A., Morandi, F., Bjørkvoll, T., Gabrielle, B., Østergård, H., Flatberg, T., Wohlfahrt, J., Damman, S. (2015). Socio-economic effects of biomass supply chain: case studies from Logist’EC Project. Proc. 23Rd EU BCE, Vienna, 2015.
6
ACKNOWLEDGEMENTS
The Logist'EC project is funded under the 7th Framework Programme of the European Union. Contributions from C. De la Rua (CIEMAT, Spain) and Sigrid Damman (SINTEF, Norway) are acknowledged, as well as from Philippe Béjot (BP, France) who provided data and field expertise for the Burgundy case-study.
For further information and update on the Logist'EC project, readers may visit its web site: www.logistecproject.eu.
