Techno-economic evaluation of biomass-to-end-use chains based on densified bioenergy carriers (dBECs)

Applied Energy 239 (2019) 715–724

Contents lists available at ScienceDirect

Applied Energy journal homepage: www.elsevier.com/locate/apenergy

Techno-economic evaluation of biomass-to-end-use chains based on densified bioenergy carriers (dBECs)

T

⁎

Fabian Schipfer , Lukas Kranzl Technische Universität Wien, Energy Economics Group, Gusshausstraße 25-29/370-3, Austria

H I GH L IG H T S

densification to pellets, torrefied pellets and pyrolysis oil are analysed. • Biomass technologies are economic feasible for residential space heating within the EU. • All energy contents and –densities result in lower marginal costs. • Enhanced to reduce costs of a developing European Bioeconomy is recommended. • Implementation • Biomass storage options with densified bioenergy carriers shall be investigated.

A R T I C LE I N FO

A B S T R A C T

Keywords: Pelletisation Torrefaction Fast pyrolysis Supply chain Market introduction Comparative economic assessment

The European Union plans to shift parts of its economy towards a biobased system commonly referred to as a bioeconomy in order to reduce emissions and fossil fuel dependence. Biomass exhibits lower carbon densities, higher moisture contents, and is more heterogeneous when compared to the feedstock basis of the current economy. In this paper, we simulate generic biomass-to-end-use chains to compare economic performances of the three technologically most advanced pre-treatment options for biogenic raw materials. Exemplary cellulosic biomass feedstocks are computed to be processed to pellets, torrefied pellets and pyrolysis oil based on current data from previous research and demonstration projects. Various distribution options are considered for the resulting densified bioenergy carriers to be finally converted to heat, electricity and liquid biofuels. We find that the discussed densified bioenergy carriers could compete in the existing residential heating market. Furthermore, large-scale conversion facilities like coal co-firing and gasification could profit from cost reductions for torrefied pellets when compared to conventional pellets. To reach commoditisation of these bioenergy carriers as well as full commercialisation of the respective technologies, upscaling would have to start now possibly by establishing a residential heating market based on torrefied pellets where framework conditions are most favourable.

1. Introduction Modern bioenergy plays a crucial role in securing the supply of renewable energy and thus in phasing out fossil fuels. In 2016, the energy mix of the EU28 was based on 13.2% renewables, more than half of which come from biomass and renewable wastes (8.6%) [1]. Biomass exhibits lower carbon densities, higher moisture contents, and is more heterogeneous than the feedstock basis of the current fossilbased economy. Pre-treatment of biogenic raw materials into homogenous solid, liquid and gaseous densified bioenergy carriers (dBECs) with high energy densities, therefore, holds the potential to improve overall supply costs [2] and environmental performance of the energy

⁎

system [3]. While dBECs played a limiting role just 15 years ago, consumption of wood pellets – currently the most important dBEC in terms of energy content – has increased, as a proportion of the EU28′s consumption of fuelwood, wood residues, and by-products, by up to 4% [4]. The development of a future bioeconomy and the subsequent conversion of biomass resources into bioenergy but also other bio-based products will rely on a series of tradable dBECs, such as wood pellets in order to meet demand in countries with resource deficits. To minimise competition with food production, second generation feedstocks, especially agricultural, forestry residues, organic waste products and dedicated energy crops, will be used in thermo-chemical conversion

Corresponding author. E-mail address: [email protected] (F. Schipfer).

https://doi.org/10.1016/j.apenergy.2019.01.219 Received 2 October 2018; Received in revised form 4 January 2019; Accepted 25 January 2019 0306-2619/ © 2019 Published by Elsevier Ltd.

Applied Energy 239 (2019) 715–724

F. Schipfer, L. Kranzl

2. Methodology

routes (combustion, gasification, pyrolysis, liquefaction) besides biochemical conversion routes (anaerobic digestion, microbial/enzymatic processes, enzymatic hydrolysis, microbial fermentation) and physiochemical conversion routes (mechanical extraction) [5]. In this paper, we focus on pre-treatment and densification of cellulosic biomass by pelletisation and the thermo-chemical processes torrefaction and fastpyrolysis because of their high technological readiness (TRL) (TRL 6-7 and TRL 9 in case of simple pelletisation) [6]. On the one hand, the augmented data availability relative to other thermo-chemical, chemical and biological processes due to recent interdisciplinary research and development projects increases the relevance of respective technoeconomic discussions for pyrolysis oil [7] and torrefied pellets [8]. On the other hand, besides technological advances, the incipient commoditisation of torrefied pellets and pyrolysis oil serves to justify our choice [9]. We discuss the importance of standardisation as well as of supply chain development and the development of market-related properties of the respective products for the example of wood pellets in [10]. Modelling and assessing biomass-to-end-use chains - including sourcing of the feedstock, supply to densification plants, pre-treatment and densification to dBECs, distribution to end users and conversion to heat, electricity, transport fuels or chemicals - attracted the attention of a diverse research community. In [11] for example, 124 references are discussed, including models optimising biomass supply chains based on fixed network structures, temporal and/or spatial granularity. The aim of these models is to provide strategic or tactical decision support with or without facility location decisions, but mainly as decision support tools for stakeholders from the industry. In [12] we find 54 papers reviewed, including the comparison of pre-treatment technologies and their possible contribution to increasing shares of biogenic carbon in the energy system. The supply chain literature relevant to our research can be found, to a certain extent, in [12] and can be categorised into four types. Thorough analysis was performed to optimise feedstock allocation, storage, pre-treatment and end-use locations in specific regions and countries [13–17], or to related topics and also based on Geographic Information System (GIS) methodologies on deriving biomass allocation potentials in specific regions and countries [18–20]. Without specific network structures and significant predetermined temporal- and/or spatial granularity, relevant publications discuss different densification technologies with the aim to optimise a specific end-use application [21–25]. More fittingly to our research in [26–33] we find their focus on the pre-treatment technologies themselves, however, discussing only one or two densification technologies (e.g. comparison between pellets and torrefied pellets). Still, only [34] compared pellets, torrefied pellets and pyrolysis oil and discussed possible market entry strategies of these densification technologies. This publication, however, was published ten years ago and cannot be considered up-to-date. With the presented paper we want to provide a framework that enables a transparent, scientific and sound comparison for more than two biomass densification technologies. The aim is to provide generic insights into the economic impact of different feedstocks and supply structures on the densification set-ups and densification in general as well as ideas on how the costs of the produced dBECs are reflected when forwarded in different distribution networks for the final energy conversion of various end-user types. On the one hand, these insights can highlight combinations of feedstocks, supply chains and end users and their current potentials to reduce costs by deploying respective bioenergy densification technologies. On the other hand, the findings allow us for the first time to derive statements about the general added-value of biomass densification technologies and how these technologies could play a role in the cost-efficient deployment of renewables based on densified bioenergy commodities.

2.1. Techno-economic assessment and simulation of biomass-to-end-use chains In the assessment of bioenergy systems the entire biomass-to-enduse chain, from biomass sourcing to biomass conversion has to be considered, including possible pre-treatment steps and densification, its raw biomass supply, and dBECs distribution to the conversion site (see Fig. 1). Data and specifications on sourcing for two types of biomass, supply and densification in simple pelletisation, torrefaction combined with pelletisation as well as pyrolysis plants are mainly based on data compilation from several countries and technologies including practical tests from the BioBoost [7] and the Sector projects [8]. We calculate costs throughout the entire biomass-to-end-use chains. On the other hand, deriving and analysing prices, and including profit margins for the different supply chain stakeholders does not fall within the scope of this study. We set up a generic-biomass-to-end-use chain tool. The underlying algorithm calculates dBECs deployment costs by step-wise cumulating the costs of supply chain components. Firstly, it optimises densification plant sizes in a range of up to 500 * 106 g*year−1 input and respective feedstock supply distances for various feedstocks, supply modes and feedstock yield, availability and accessibility combinations. Then distribution costs for different, potentially relevant distances and transport mode combinations as well as storage are added to the resulting production costs. Finally, dBECs deployment for residential heating, gasification, and Fischer-Tropsch (FT) synthesis and coal co-firing are calculated and compared. Ensuring comparability of the different production technologies for each feedstock, supply, dBEC distribution and deployment combination defines the very core of the tool. This results in 72 possibly relevant biomass-to-end-use chains for each technology to be analysed and discussed. Input data was adjusted to the base year 2017 and all results are discussed in €2017 per energy content of the respective dBECs. One challenge of setting up a generic biomass-to-end-use chain tool capable of comparing different types of relevant transportation modes & distances, feedstocks and densification technologies as well as various end-user types actually lies in generating a homogenised and comprehensive dataset. At the same time, each calculation step has to be generalised well enough to enable the comparison of the biomass-toend-use chains down to a high resolution, optimally down to each chain link. The upcoming sub-sections describe the content of this database as well as the simulation algorithm, without going too much into details of the actual dataset to still ensure the papers readability and focus on its key messages. Sourcing and compilation of the database itself is described in an additional text document, which can be downloaded together with the database and the BioChainS-simulation tool in this Github-repository1 or with the data-in-brief via the permalink of this paper. 2.2. Feedstock supply In order to simplify the comparative biomass-to-end-use chain assessment we focus on (1) wood chips (beechwood) as wood forest residues, used wood or by-products from forest industries and on (2) wheat straw collected on the roadside next to agricultural land. Similar to the supply chain assessments in [35], these feedstock types serve as reference feedstock types that can be considered to be comparable, e.g. to miscanthus in the wheat straw case and other types of wood or woody crops in the wood chips case. Supply of feedstock takes place with either a tractor or a truck and an attached trailer. Parameters for 1 https://github.com/schipfer/scenarios/tree/master/BioChainS_Biomass_to_ end_use_chain_simulation.

716

Applied Energy 239 (2019) 715–724

F. Schipfer, L. Kranzl

Fig. 1. The biomass-to-end-use chain. Different feedstock supply- as well as dBEC distribution distances and modes have been calculated. Source; own illustration.

to value the higher technological readiness. Therefore, the previously discussed investment costs and capital recovery factors are used to calculate reference annualised capital costs (refCAPEX γ ) for reference densification plant output sizes (refSCLγ ) .

the techno-economic evaluation of the feedstock supply step includes handling costs, variable costs per distance including labour and fuel, the maximal cargo capacity in tonnes and the design ratio of the transportation mode in [kg*m−3] indicating the minimum density of the transported product. Lower densities lead to derating of the massspecific transport costs through not using the maximal cargo capacity. All values for feedstock supply are based on or directly taken from [36] who documented costs of straw bales and wood chips supply including the discussed derating effects.

CAPEX γ = refCAPEX γ ∗ oSCLγ s ∗ refSCLγ −s

(2)

Bioenergy carrier production costs (Cγ ) are now computed based on the objective function including annualised specific capital and supply costs as well as feedstock (Cα ) and plant-specific variable costs (varCγ ) .

Cγ = min {Cβ (oSCLγ ) + CAPEX γ (oSCLγ )} + varCγ + Cα ∗ Yγ −1

2.3. Production of densified bioenergy carriers

(3)

While increasing densification plant sizes decrease the specific biomass densification costs through economic unit scaling effects, supply distances and, therefore, specific biomass supply costs increase since the supply radius for biomass sourcing has to be extended [33]. The optimisation (3) is implemented to reflect these trade-offs and to calculate comparable bioenergy carrier production costs for similar biomass availability and supply settings. Densification plant sizes are thus sensitive to parameter variations including feedstock yield, availability, and accessibility, supply modus (truck or trailer), energy and mass density as well as scalability of the operational and capital expenditures of the operation.

Straw bales and wood chips are assumed to be processed with various densification technologies. Therefore techno-economic data is adapted from [37] and from [28] for the production of pellets, further denoted as white or traditional pellets, from BioBoost deliverables [35], from [38] for fast pyrolysis to biosyncrude and from relevant SECTOR project deliverables for torrefaction and pelletisation to torrefied pellets for costs [39] and energy and mass balances [40]. Torrefaction and pyrolysis, which are more cost-intensive technologies, thermo-chemically decompose the biomass in the absence of oxygen. Depending on the temperatures and residence times, the decomposition is either focused on solid fractions in the case of torrefaction or on the liquid fraction in case of fast pyrolysis. Ref. [38] define biosyncrude as “all-inone-slurry” including a mix of condensates and char from the pyrolysis process, “… reducing the complexity in terms of transport, as only one homogeneous energy carrier is dealt with” with no by-products. This is in contrast with catalytic pyrolysis, where higher quality bio-oil is produced alongside excess electricity, spent catalysts, ash and an unused aqueous phase [35]. In order to calculate potentially relevant dBEC production costs (Cγ ) , densification plant sizes are optimised depending on feedstock, supply and densification technology parameter, and costs. We use the simplified assumption that biomass feedstock is homogenously distributed around the theoretical densification plant. Therefore, it can be collected and supplied using an average supply radius (Dβ ). The average supply radius, necessary to calculate supply costs (Cβ ) , is based on the discussed feedstock yields (Yα ) as well as the feedstock input size (iSCLγ ) depending on the dBEC output size (oSCLγ ) of the densification plant and its dry mass yield (Yγ ) :

2.4. dBECs-distribution For transportation of dBECs to end users, referred to here as distribution, we discuss several options including road, rail, and, in the case of intercontinental trade, maritime transport. We also discuss bioenergy carrier storage, drawing primarily on [41]. While the actual prices paid for the distribution of dBECs depend on supply of and demand for transport and handling services and are sensitive both to economies of distances and to economies of scale [42], we pursue a more simplified approach to enable a transparent comparison and illustration of possible relevant transport costs that would have to be paid for delivering the discussed densified bioenergy carriers. Even though this paper aims for a generalisation of comparative analysis of dBECs, it is helpful to discuss exemplary distribution distances and constellations. We further use six representative distribution pathways (RDP1-6), short and long distances for road, rail and maritime transport respectively. In Table 1, distances for each mode of transport and the compilation of transport nodes are summarised and comparable pathway cases are outlined. While RDP1 could, for example, represent any regional supply chain, RDP2-4 could be used to describe transport to pellet importing countries like Germany, Austria, and the Netherlands from, for example, Ukraine and the Finnish Gulf via rail haulage or maritime transport. For short sea shipping, delivery

Dβ = iSCLγ 0.5∗ (Yα ∗100 ∗ π )−0.5 in [km] and with iSCLγ = oSCLγ ∗ Yγ −1 (1) Based on [33], an average scaling factor of s = 0.7 is assumed for pyrolysis and torrefaction to scale the annualised capital costs (CAPEX γ ) . For pelletisation we assume an average scaling factor of 0.8 Table 1 Representative distribution pathways (RDPs) and comparable pathway cases. Abbr.

1st connection

2nd connection

3rd connection

Comparable pathway cases

RDP1 RDP2 RDP3 RDP4 RDP5 RDP6

50 km Road 1300 km Rail 200 km Rail 2000 km Rail 400 km Road 200 km Rail

no 50 km Road 2000 km Sea 50 km Road 50 km Road 15,000 km Sea

no no 50 km Road no no 50 km Road

Any regional supply Ukraine to Germany or Austria, Russia to ARA Russia to Germany or Austria From Austria to Italy US or CA West Coast, Africa East coast to ARA or Italy

717

Applied Energy 239 (2019) 715–724

F. Schipfer, L. Kranzl

based on straw and wood chips as well as their respective cost shares. Costs are specified on the net calorific value per energy carrier output. To provide comparability the same input parameters for feedstock harvesting and supply are used for the three technologies, leading to various optimal bioenergy carrier production plant sizes. We find an optimal annual production plant size of about 100 kt wood pellets. Upscaling of a pellets plant based on wheat straw to about 360 kt still leads to higher marginal cost reductions for the installation than supply costs increases. This is due, on the one hand, to the comparatively high combination of feedstock yield, accessibility and infrastructure surrounding the plant in case of wheat straw, which may be up to five times higher than that assumed for plants supplied with wood chip. On the other hand, a higher moisture content for wood chips leads to higher transportation costs and lower optimal plant sizes. Comparing the different densification technologies reveals that, with higher investment costs, scaling effects become more important over supply cost increments. For the delivery of raw materials we limited the optimisation algorithm to an annual 500 kt straw input, equating to transport activity of about three trucks per hour approaching the plant, which might raise issues with regard to public perception which can be “chiefly related to the impacts of truck movements” [51]. While torrefied straw pellets and straw and wood pyrolysis oil production upscaling are affected by this limit, production of torrefied wood chips pellets results in optimal plant sizes of about 315 kt input. For torrefied pellets based on straw and wood, WtT and StT respectively, production costs of 7.4 and 10.1 € GJdBEC−1 are derived. For pyrolysis oil (WtO and StO) costs of 13.7 and 15.9 €*GJdBEC−1 are calculated. For a comparison, [52] describe production costs of 8.8–11.4 €*GJdBEC−1 for torrefied pellets and 13.9–19.4 €*GJdBEC−1 for pyrolysis oil respectively. Based on the net calorific value of the produced dBEC, cost shares for feedstock, supply and densification can be compared. With increasing cost shares for capital and operational expenditures from about 30–40% for pelletisation to 50% for pyrolysis, shares for feedstock decrease. However, longer supply distances result in increasing supply costs (up to 1.9 €*GJdBEC−1 for WtO) while feedstock costs per energy content dBEC increase too, due to conversion losses in contrast to a 100% conversion efficiency assumed in the case of simple pelletisation. Operational expenditures include the drying process based on natural gas furnaces for all densification technologies. This results in feedstock costs for WtT and WtP at 3.3 €*GJdBEC−1 which is beneath the original feedstock costs (4.1 €*GJdBEC−1) when specified on the energy content

to ports such as St.Petersburg or Vyborg from the hinterland can be considered [43]. In the case of long-distance road transport, which can be assumed, for example, for imports to Italy from Austria, we assume an initial road haulage of 400 km to an intermediary storage from where freight could be delivered to small-scale end users [44]. Wood pellets imports from the west coast of the USA as well as from Canada to Italy and the ARA-ports (Amsterdam, Rotterdam and Antwerp) increased in recent years [4]. RDP6 is based on distances to be considered to represent similar supply chains, although supply chains from the west coast of Africa would also result in comparable distances. Exemplary storage options for densified bioenergy carrier with 20day and 80-day storage capacity are considered. Intermediary storage could be situated at the densification plant site in bioenergy carrier depots at transport hubs or in proximity to – or indeed within the vicinity of - the end user’s site [45]. 2.5. dBECs deployment Bioenergy carriers are considered to substitute fossil energy carriers for (1) small-scale heat production for residential consumers (2) largescale power production and (3) Fischer Tropsch (FT) diesel production for use in the transport sector. Energy conversion processes differentiate in specific conversion costs based on capital and variable costs of the conversion technologies as well as conversion efficiencies. Input data is mainly derived from the ENTRANZE database [46], from [47] for cofiring and from [48] for gasification and FT-synthesis. Reference prices for residential heating are calculated and a representative European cross-section (Spain, Germany, and Romania) [46]. For co-firing, Eurostat data on average 2017 electricity prices, without taxes and levies, for the largest consumer group are used as a reference (band IG, consumption 150.000 MWh or over) [49]. The gasification and FT-diesel production path is compared to average EU28 diesel prices, without taxes and levies, from the European Commission [50]. 3. Results 3.1. Bioenergy carrier production costs and respective optimal densification plant size We discuss all costs throughout the entire biomass-to-end-usechains in €2017 values. Fig. 2 illustrates bioenergy carrier production costs for traditional pellets, torrefied pellets and pyrolysis syncrude oil

Fig. 2. Bioenergy carrier production costs for three densification technologies and two biomass feedstock types [€2017*GJdBEC−1]. Forestry residues are abbreviated with “W” and straw with “S”, traditional pelletisation with “P”, torrefaction and pelletisation with “T” and fast pyrolysis with “O” resulting e.g. in “StO” as straw to pyrolysis oil.

718

Applied Energy 239 (2019) 715–724

F. Schipfer, L. Kranzl

delivery costs, including bioenergy carrier, transport, and storage costs for a set of representative distribution pathways reaching from regional supply to long distance, intercontinental pathways like production on the Westcoast of Canada or the United States to an end user in close proximity to the ARA ports or to an Italian port have been calculated (RDP6). All densified bioenergy carriers discussed here exceed the design ratios of the different modes of transport, thus fully utilising the cargo weight capacities. Cost reduction for transportation can only be achieved through increased energy contents. Based on the discussed energy contents, transportation of torrefied straw pellets and wood pellets are assumed to be 13% and 22% cheaper than their traditional pellet counterparts due to differences in net calorific value of the bioenergy carrier. Transport cost reductions for pyrolysis oil compared to conventional pellets are estimated to be in the same magnitude as for StT (13%). However, no differences in handling and transport of liquid and solid dBEC could be included in the calculation of this paper since no comparable data set could be found in the literature. Costs for transport ranges from 8.1 €*tdBEC−1 for RDP1 (regional supply chain) to 55.0 €*tdBEC−1 for RDP6 (inter-continental supply chain) for all densified bioenergy carriers. Our simplified assumptions for transportation result in marginal increases of 16.4 €*tdBEC−1 for 100 km for truck and 14.8 and 4.5 €*tdBEC−1 for 1000 km for rail and maritime transport excluding transhipment but including empty backhaul. These assumptions result in more expensive long-distance truck transport (RDP5), for example representing distribution from Austrian and German producers to Italian consumers than for medium rail transport and short-sea shipping (RDP2-4). To calculate storage costs, however, the energy density of the bioenergy carriers become decisive over energy content when storage providers are paid on the basis of storage volume. Energy densities of straw and wood pellets are about 3.2 and 2.9 higher than their original feedstocks, thus impacting on the storage costs with the same factor. Torrefaction leads to further energy density increases of a factor of 1.4 and 1.4 while pyrolysis oil is 1.7 and 1.4 times denser than torrefied straw and wood pellets respectively. This results in storage costs of about 0.31 €*GJdBEC−1 for straw, 0.10 €*GJdBEC−1 for StP, 0.07 €*GJdBEC−1 for StT while pyrolysis oil can be stored for 0.09 €*GJdBEC−1 for 20 days, also considering the higher costs for liquid than solid storage. For the sensitivity analysis in Table 4, we discuss storage up to 80 days. Lower bulk densities lead to the largest cost increases for wood and straw pellets at +4.8% and +4.5% for the shortest RDP (RDP1) and +3.5% and +3.3% for the long distance RDP6. When comparing torrefied and conventional pellets, the different cost increments result in theoretical break-even distances. For the considered distribution pathways torrefied straw pellets and torrefied wood pellets break-even in the RDP4 and in the RDP2 case respectively with their conventional pellets counterpart. Varying underlying distance dependent costs for the RDP1 has similar impacts on dBEC delivery costs to those that would be observed if underlying handling costs are varied. At ± 50% handling or distance costs, dBEC delivery costs increase/decrease by about ± 0.7–1.8%. With expanded distribution chain distance and

Table 2 Calculated supply costs of feedstock to various densification plants [€2017*GJdBEC−1]. The values in brackets indicate the supply distances resulting from the pellet plant size optimisation.

1 2 3 4

Feedstock [€/GJ]

Pelletisation

Torr_Pell

Pyrolysis

Wheat Straw via Truck Wood Chips via Truck Wheat Straw via Tractor Wood Chips via Tractor

0.5 (19 km) 0.9 (21 km) 0.6 (12 km) 1 (15 km)

0.6 (26 km) 1.2 (46 km) 1 (26 km) 1.3 (32 km)

0.9 1.9 1.4 2.1

(29 km) (60 km) (29 km) (39 km)

of the resulting densified bioenergy carrier, which is higher than the energy content of the original feedstock. Supply costs vary from 0.5 €*GJdBEC−1 for wheat straw pelletisation to 1.9 €*GJdBEC−1 for pyrolysis oil from wood chips production. Supplying the feedstock with tractors instead of trucks results in 10–14% higher costs for wood chips and 17–64% higher costs for wheat straw supply for the various densification facilities respectively. Increased transportation costs for tractors impact negatively on the plant size optimisation, resulting in shorter supply distances at higher or similar overall supply costs than when supplied via trucks (see Table 2). Based on dBEC production costs, less efficient supply transport modes result in the strongest effects for the torrefaction of wood pellets (WtT) at about +5.6%. This is generally higher for the supply of wood chips for pyrolysis oil (WtO with +5.1%) and simple pelletisation (WtP with +3.1%) than for torrefied straw pellets (StT with +3.8%), pyrolysis oil from straw (StO with +3.7%) and straw pellets (StP with +2.5%). Due to the exponential scaling function and an upper limit of 500 kt BEC-output, production costs are more sensitive to lower biomass yields than increased yields. With 80% lower annual yields of 0.62 t*ha−1 (instead of 3.10 t*ha−1) for straw and 0.13 t*ha−1 (instead of 0.65 t*ha−1) for residual wood chips, bioenergy carrier production costs increase by about +2.3% to +7.4% while, in contrast, maximum cost reductions of −2.2% can be reported for pyrolysis from wood chips with 80% higher yields. Generally, production costs are lower for bioenergy carriers based on chips than based on straw by about 16% for simple pelletisation, 27% for torrefaction and pelletisation and 14% for pyrolysis, mainly due to higher feedstock costs for wheat straw. Higher investment costs for StT and StP than for WtT and WtP are also included in the calculation, while no differences for the fast pyrolysis cases are assumed. Furthermore, varying investment costs by ± 10% leads to a maximum increase of +1.8% and −1.9% in pyrolysis oil production costs based on wood chips and effects are comparatively low for simple pelletisation with ± 0.7% and ± 0.4% for wood and straw pelletisation respectively. Increased investment costs result in increased plant sizes where possible (see Table 3). 3.2. Representative bioenergy carrier distribution paths and respective costs Distribution of bioenergy carriers from densification plants to end users has to be considered based on different modes of transport as well as transport distances and distribution chain configurations. Total

Table 3 Calculated sensitivity analysis for different delivery modes (Truck and Tractor), deviating feedstock yields ( ± 80%) and deviating investment costs ( ± 10%). Bioenergy carrier production costs in [€2017*GJdBEC−1] and optimized densification plant sizes in kt output annual. The first column (Truck delivery) represents the base case. Asterix indicate exhausted upscaling limit in the densification plant size optimization.

1 2 3 4 5 6

Tech.

Base [€/GJ]

Tractor

StP StT StO WtP WtT WtO

8 (337 kt/a) 10.3 (500 kt/a)* 15.7 (500 kt/a)* 6.7 (89 kt/a) 7.4 (315 kt/a) 13.8 (459 kt/a)

+2.5% +3.8% +3.7% +3.1% +5.6% +5.1%

(−64.0%) (+0.0%)* (+0.0%)* (−52.9%) (−52.9%) (−57.7%)

Yield-80%

Yield + 80%

Inv.Cost + 10%

Inv.Cost-10%

+2.3% +3.6% +3.5% +3.8% +7.4% +6.8%

−0.7% −0.7% −0.7% −1.1% −2.2% −2.0%

+0.4% +1.3% +1.6% +0.7% +1.3% +1.8%

−0.4% −1.3% −1.6% −0.7% −1.3% −1.9%

(−63.4%) (+0.0%)* (+0.0%)* (−63.4%) (−63.4%) (−68.3%)

719

(+44.4%) (0.0%)* (0.0%)* (+44.4%) (+44.4%) (+9.0%)*

(+12.7%) (0.0%)* (0.0%)* (+12.7%) (+12.7%) (+9.0%)*

(−12.3%) (+0.0%)* (0.0%)* (−12.3%) (−12.3%) (−14.0%)

Applied Energy 239 (2019) 715–724

F. Schipfer, L. Kranzl

Table 4 Calculated sensitivity analysis for delivery cost. The two values give the extreme cases (RDP1|RDP6), for the original values in [€2017*GJdBEC−1] and additional 80 storage days, variations in distance cost (Dist.Cost) and in handling cost (Hand.Cost) the deviations from the original values.

1 2 3 4 5 6

Tech.

Base [€/GJ]

Store + 80d

Dis.Cost ± 50%

Hand.Cost ± 50%

StP StT StO WtP WtT WtO

8.6|11.7 10.8|13.4 16.3|18.9 7.3|10.2 7.9|10.1 14.3|17

+4.5%|+3.3% +2.6%|+2.1% +2.3%|+2% +4.8%|+3.5% +3.1%|+2.4% +2.7%|+2.3%

± 1.5%| ± 11.2% ± 1.1%| ± 8.2% ± 0.7%| ± 5.8% ± 1.7%| ± 11.9% ± 1.2%| ± 9.3% ± 0.8%| ± 6.5%

± 1.6%| ± 4.2% ± 1.1%| ± 3.2% ± 0.7%| ± 2.2% ± 1.8%| ± 4.6% ± 1.3%| ± 3.6% ± 0.8%| ± 2.5%

Fig. 3. Calculated bioenergy deployment costs for different end users based on the six representative distribution pathways (each plus represent one RDP). Red horizontal lines indicate average energy deployment costs based on fossil resources (EU28 in 2014). Solid line indicates median, dotted lines lower and dashed line upper values.

distribution pathways, small-scale heating based on WtT is about +0.8 €*GJ−1 more expensive when based on RDP1, while a −0.1 €*GJ−1 cost advantage can be highlighted for RDP6. Similar results are calculated for StP and StT for gasification and FT-synthesis with +0.7 €*GJ−1 higher prices and −0.4 €*GJ−1 cost advantages respectively. Differences between deployment costs based on torrefied pellets and pyrolysis oil are smallest for straw biomass in the residential heating case. For RDP1 and small-scale combustion, a minimum of 1.0 €*GJ−1 difference between StO and StT can be identified, with pyrolysis oil being more expensive. Bioenergy deployment costs are calculated to be more expensive when based on wheat straw (8.6–18.8 €*GJdBEC−1) than residual wood chips (7.3–16.6 €*GJdBEC−1). Due to the relatively low conversation rates, electricity production is affected most strongly with −3.3 €*GJdBEC−1 and up to −7.3 €*GJdBEC−1 cost advantages for the RDP1 and white wood pellets and torrefied wood pellets compared to their straw based counterparts. For the RDP6 pathway, the differences spread to −3.9 €*GJdBEC−1 and −8.2 €*GJdBEC−1 respectively. WtO is about −4.9 €*GJdBEC−1 cheaper than StO in the RDP1. In comparison to estimated reference energy deployment costs, we find that residential heat production based on wood pellets and torrefied wood pellets for lower RDPs can result in costs (33.6–42.9 €*GJ−1) comparable to the upper range of EU28 heating costs in 2017 (36.9 €*GJ−1). However, average heating costs are calculated with 25.2 €*GJ−1 and minimum heating costs with 19.6 €*GJ−1. The calculations are based on heating oil prices [50] (11.4–13.9 €*GJ−1 with an average of 12.8 €*GJ−1), natural gas prices for small consumer (4.7–18.0 €*GJ−1 with an average of 10.9 €*GJ−1) and electricity prices (10.6–102.2 €*GJ−1 with an average of 25.4 €*GJ−1) [49]. FT-synthesis based on WtT can be discussed as the most feasible

additional handling steps, the variations can impact delivery costs by up to ± 11.9% when production costs are low and when distribution makes up a considerable share of the delivery costs. 3.3. Bioenergy deployment costs In Fig. 3, bioenergy deployment costs are illustrated for three enduser types as well as for dBEC-delivery based on the values from the previous chapter to facilitate comparability. Conversion efficiencies inflate the delivered fuel costs while utility costs including capital and operational expenditures increase the levelized costs of useful energy derived depending on the dBEC type used as input. We compare costs across the six dBEC types and between the bioenergy conversion routes and reference price ranges for EU consumers in 2017. Delivery to the end-users is most cost-effective for wood pellets at 7.3–10.2 €*GJ dBEC−1. Small cost-advantages can be highlighted for the longest transport distances for torrefied wood pellets with 10.1 €*GJ −1 . Costs for wood pellets calculated in the model about 1.0 €*GJ dBEC −1 higher than average pellet prices shipped to the ARA-ports in dBEC 2017 with (6.2–8.9 €*GJ−1). In the period under consideration however, industrial pellet prices exhibited a record low in the current decade [53]. Simulated energy deployment costs are lowest for gasification and FT-Synthesis followed by residential heating and electricity production in coal co-fired power plants. For large-scale combustion (FT and cofiring) torrefied wood pellets are simulated to be about −3.0 €*GJ−1 to −3.3 €*GJ−1 and −2.8 €*GJ−1 to −3.1 €*GJ−1 cheaper than their conventional counterparts for the two extreme RDP cases respectively. For residential heating we derive comparable deployment costs for WtP and WtT pathways. However, with small variations for various 720

Applied Energy 239 (2019) 715–724

F. Schipfer, L. Kranzl

Table 5 Calculated sensitivity analysis for Bioenergy deployment costs (row 1–3) [€2017*GJdBEC−1] and deployment cost differences (row 4–6) [€2017*GJdBEC−1] for FTsynthesis (FT), electricity production (El.) and residential heating (RH). The value pairs represent the RDP1 and RDP6 values respectively.

1 2 3 4 5 6

Case

Base [€/GJ]

BM.cost-20%

Yield + 80%

Yield-80%

dist.cost + 50%

Store + 80d

WtT FT WtT El. WtT RH WtT vs WtP FT StT vs StP FT StO vs StT RH

24.9|30.1 33.9|41 33.6|37.5 1.9|3.3 −0.7|0.4 −1|−1.1

−4.8%|−4.0% −4.9%|−4.0% −2.7%|−2.4% 1.8|3.2 −0.6|0.5 −0.4|−0.5

−0.6%|−0.5% −0.6%|−0.5% −0.3%|−0.3% 2.1|3.4 −0.7|0.5 −0.9|−1.1

+1.9%|+1.6% +1.9%|+1.6% +1.0%|+0.9% 1.4|2.8 −1|0.1 −1.2|−1.4

+0.9%|+7.3% +0.9%|+7.4% +0.5%|+4.3% 2|3.8 −0.7|0.9 −1|−1.2

+2.6%|+2.1% +2.6%|+2.2% +1.4%|+1.3% 2.1|3.5 −0.5|0.6 −1.1|−1.3

4. Discussion

option with respect to large-scale bioenergy deployment in our biomass-to-end-use chain selection with 23.0–26.8 €*GJ−1. However, average diesel prices (not production costs!) are reported at about 14.0 €*GJ−1 in 2017. The minimum and maximum prices ranged from 12.7−14.9 €*GJ−1 [50]. Current liquid biofuels production costs are reported at 14.0–25.0 €*GJ−1 for hydrotreated vegetable oil (HVO) and 25.0–35.0 €*GJ−1 for FT-diesel from wood [54]. In the co-firing case, the combustion of WtT is also simulated to be most cost-effective (32.3–37.9 €*GJ−1), albeit within another cost range than the estimated EU28 electricity production costs in 2017. For the comparison we discuss average reported electricity prices of the largest consumers in the EU of about 11.5 €*GJ−1 in 2017 with a range from 7.7 to 19.9 €*GJ−1 [49]. Differences between WtT and WtP cofiring are between 1.6 and 3.2 €*GJ−1 for RDP1 and RDP6 respectively. Ref. [8] describe a difference of about 0.9 €*GJ−1. While the FTsynthesis pathway costs already include a remedy for produced electricity, a thorough assessment of possible differences in CO2 credits remedies was not within the scope for this work. A sensitivity analysis of selected bioenergy deployment pathways, as well as selected dBECs comparisons, is illustrated in rows 1–3 of Table 5. The base values are illustrated in the first column through a value pair representing the RDP1 and RDP6 values respectively. The most cost-effective bioenergy deployment option, torrefied wood pellets to FT-diesel via gasification exhibits the most substantial impacts of 20% reduced biomass costs with up to −4.8% total cost reductions. Reducing feedstock yields by 80% increases bioenergy deployment costs by up to about +1.9%. Storing the torrefied wood pellets for 80 days instead of 20 days results in +2.6% increased costs. Increasing the distance-dependent costs by 50% affects the discussed option by increased total deployment costs of up to +7.4% when considering the RDP6 case. For the less cost-effective energy deployment cases examined, sensitivities in the production and distribution chain are depressed due lower total deployment cost shares. We furthermore compare bioenergy deployment based on different bioenergy carriers (in rows 4–6 in Table 5) by discussing deployment cost differences and deviations of these differences due to changing parameters in the biomass-to-end-use chain. Positive differences such as those between the RDP1 and RDP6 option of WtT and WtP or those between the RDP6 option of StT and their conventional counterpart represent increased cost-efficiency for the more investment-intensive densification technologies. Both torrefied pelletisation and pyrolysis oil production benefit from increasing wood and straw yields when compared to less investment-intensive pre-treatment technologies. This is illustrated by slightly increased positive and negative cost differences, although these differences equate only to about 0.2 €*GJ−1 (for WtT vs. WtP in the FP case) when compared to the base case. Increased distance dependent costs and higher storage time positively affect the torrefied cases when compared to the conventional pelletisation cases, while the comparison between straw based pyrolysis oil and torrefied pellets for residential heating is more favourable for the less investment-intensive densification technology. This is mainly due to higher storage costs assumed for liquid versus solid bioenergy carriers.

A comparative biomass-to-end-use chain assessment tool was applied to simulate and compare bioenergy deployment costs for a set of potentially relevant supply chains with dBECs based on pelletisation, torrefaction plus pelletisation and fast pyrolysis. The input data, adjusted to the base year 2017, is based on the most recent research and development projects. The techno-economic results that are presented can, therefore, be seen as relevant to current discussions concerning possible market entry strategies for densification technologies and their dBECs. In order to compare the economic performance of the discussed densification technologies under possible relevant framework conditions, the same chain set-ups are calculated for the three technologies, while densification plant size costs are optimised against the respective feedstock supply costs. This results in generally lower overall dBECs deployment costs than discussed and compared in current literature calculating deployment costs for pyrolysis oil [35] and torrefied pellets [8] or reviewing available estimates for these technologies [52]. At the same time, pellet delivery costs via the ARA-ports is estimated in a comparable but slightly higher range than 2017 average ARA-port prices. On the one hand reported prices (average 2017 from ARGUS media) include below-average outliers and represent prices payed for takeover from the buyer at the ARA-ports and not delivered to end-user plant gate as assumed in the model. On the other hand, the model does not include any profit margins and delivery costs should thus be lower than delivery prices. Supply chains based on wood tend to exhibit cost advantages when compared to straw-based chains and energy from solid dBECs can be deployed with greater economic efficiency than those in liquid form. Cost differences between torrefied and conventional wood pellets decrease with increasing transport distance and can be partly compensated and reverted for intercontinental maritime transport as well as road haulage within Europe. Due to capacity effect costs in large-scale conversion facilities like coal co-firing and gasification, cost differences can result in reductions of up to about 3 €*GJ−1 for torrefied pellets compared to conventional pellets. While no capacity effect costs can be expected for small-scale users, cost advantages for the more investment-intensive technology can be highlighted for longer distance dBEC supply. We find in general that pyrolysis oil, torrefied and traditional pellets based on wood and straw can compete in the current European residential heating market. Bioenergy already plays an important role in the residential European heating and cooling market with 17% of primary energy input in 2012 for the EU28 [55]. Depending on preferences and existing assets across different European countries, replacement of fossil-based heating devices with pellets, torrefied pellets, or pyrolysis oil boilers and stoves or the replacement or blending of heating oil could be considered. The production of densified BECs and the subjacent supply of biogenic raw material have to be considered jointly; with increasing investment costs of pre-treatment technologies optimal densification plant sizes also increase, depending on the marginal feedstock supply costs. Owing to higher straw yields (compared to residual forestry wood chips yields) and the lower moisture content of straw bales, densification plants based on agricultural residues should be dimensioned up to 721

Applied Energy 239 (2019) 715–724

F. Schipfer, L. Kranzl

the technologies discussed here have been tested and verified over recent years, uncertainties about investment costs, scaling effects, and technical performance are relatively high. However, for torrefaction and pelletisation we have drawn on data from several technology providers acquired during the FP7 SECTOR project thus reducing the uncertainties affecting data reported in previous publications. For the downstream of densified bioenergy, data allowing the comparison of dBECs is more limited. Currently, data on handling, transport, storage, and conversion costs of wood pellets is sparse and comparative assessments for other dBECs present clear research gaps. Approximate functions concerning durability, material safety, and biological activity under various framework conditions would be necessary to enable a more accurate comparison of distribution and conversion costs. Further logistical improvements such as removing the need for shelter during the transport and storage of hydrophobic torrefied pellets, as well as the potential for pipeline distribution of pyrolysis oil could increase significantly the competitiveness of respective supply chains. The research is further limited by our focus on end-users situated within the EU and on the current, early stage of technology commercialisation and dBEC commoditisation. We also do not address theoretical learning curves nor do we include feedstock potentials and feedstock price scenarios, both of which are important for later technology and dBEC-diffusion stages. Further comparative assessments based on empirical data with regards to downstream of dBECs are necessary to improve the findings of this research. In particular, comparative data on changes in dBEC quality throughout the supply chains under various weather conditions as well as for different handling, transport and storage options are needed. Future research has to include parallel calculation of emission and socio-economic parameters, as shown in [29]. The comparative biomass-to-end-use chain calculation tool utilised in this paper could be further deployed as an easily applicable open access tool for researchers and market actors to support decision making, further research and to cloud-map interests. Knowledge gaps on social acceptance and participation should also be addressed and the preconditions and development for market-related properties including competitiveness and liquidity should be analysed. Additionally, other feedstock types need to be discussed, for example the use of pre-treatment to enhance tradability as well as the technical properties for various conversion routes as recommended, for example, in [57].

three times larger than those based on forestry residues. Effects of yields and moisture content on optimal plant sizes have also been discussed in [33]. They still result in lower supply distances but higher production costs for straw-based dBECs. Varying yields reveal a more stringent sensitivity to decreasing than to increasing yields due to indirect exponential densification plant unit scaling function and possible relevant scaling limitations, i.e., the maximum number of truck deliveries to the plants per hour. Energy content differences of dBECs may be the single most decisive factor for transport cost reductions since wood pellets are already mostly unaffected by transport mode design factors. However, storage of bioenergy carriers at the production site, during the distribution chain, or at the end user’s site should be considered, for which energy density differences of the carrier become crucial. We find that, with increased storage periods of up to 80 days, even with short-distance delivery, torrefied wood pellets can be more cost effective than traditional wood pellets. For the timeframe 2016–2030 the International Energy Agency (IEA) discusses based on its World Energy Model (WEM) energy price trajectories in their Policy Scenario which can be seen as a middle way of their scenario family in terms of price developments [56]. Most significant price increases are discussed for oil (135%) before natural gas (73%) and coal (30%). The discussed densification technologies are to a certain extent coupled with fossil fuel prices due to transportation in supply and distribution but also due to cropping and harvesting practices, fertiliser costs as well process heat and -electricity. Therefore, decreasing fossil energy consumption in the respective supply chains should be envisaged to decouple and harness the chances of possible fossil fuel price increments. For the extension of biogenic carbon demand of the developing European bioeconomy, we can show that broadening the feedstock and intermediates portfolio is already an economically-viable option for small-scale use such as residential heating. Given particular consideration to the iLUC Directive, it seems that FT-Diesel produced in (very) large-scale gasification facilities will also rely on a set of fungible cellulosic based dBECs eventually sourced abroad. With the expansion of intermittent electricity production from solar and wind, co-firing of biomass but also bioenergy storage will gain in importance until electricity storage becomes feasible to bridge the production gaps. For the deployment in large-scale facilities, however, a level playing field with fossil energy carriers has to be established, for example through CO2 pricing mechanisms. This occurs when larger-scale densification technologies (such as torrefaction and pyrolysis of forestry and agricultural residues) lead to significantly reduced supply costs for European biogenic carbon deployment. To achieve commoditisation of bioenergy carriers as well as full commercialisation of their respective technologies, upscaling would have to start now by, for example, establishing a residential heating market based on torrefied pellets and pyrolysis oil where framework conditions are most favourable. The Italian wood pellet market, for example, is based on imports via long-distance shipping but also longer distance road haulage from Germany and Austria, thereby enabling torrefied pellets to be delivered at lower costs when compared to their conventional counterparts. Torrefied wood and straw pellets could further broaden bioenergy carrier portfolio and reduce costs for European FT-diesel production as well as co-firing for electricity production mainly due to reduced capacity effect and comparable bioenergy carrier delivery costs in comparison to traditional pellets. However, seasonal consumption patterns especially in the small-scale market demand effective storage solutions. Regulatory issues will also need to be addressed in order to deal with health and safety requirements as well as end-user perception would have to be improved when it comes to handling of “coal-like” pellets. Relatively high taxation on bioenergy carrier for residential consumers, however, would inflate cost-advantages of torrefied wood pellets or pyrolysis oil in comparison to conventional pellets. The presented calculations are limited in several ways. Even though

5. Conclusion This paper provides a novel framework that enables a transparent, scientific, and sound comparison focusing on the economic performance of three biomass densification technologies in relevant supply chains. A generic biomass-to-end-use chain tool is presented that is capable of comparing different types of relevant transportation modes and distances, feedstocks and densification technologies, as well as various end-user types using a homogenised and comprehensive dataset based on current research projects that increased the TRLs of the technologies discussed. Based on our extensive modelling efforts, we find that enhanced energy contents and densities of dBECs lead to smaller cost ranges for different supply chain set ups, for example, for maritime transport but also longer road haulage and especially for increased storage periods. Cost reductions can be expected for the deployment of torrefied wood pellets when compared to simple pellets for most of the biomass-to-enduse chains discussed in this paper. For torrefied straw pellets, similar results can only be highlighted for longer-distance distribution and for the utilisation in FT-synthesis. The deployment of pyrolysis oil is expected to be most cost-intensive in all cases. For the development of a future bioeconomy and the subsequent conversion of biomass resources into bioenergy but also other bio-based products, increasing volumes of tradeable dBECs with reasonable marginal costs will be necessary. We stress that limitations regarding 722

Applied Energy 239 (2019) 715–724

F. Schipfer, L. Kranzl

the expansion of conventional wood pellets should be overcome by commercialising the torrefaction and fast pyrolysis technologies and commoditisation of their respective dBECs where framework conditions are most favourable. For residential heating favourable conditions include long storage periods, limited local supply and high bioenergy carrier taxes. For FT-diesel and electricity production, investment into strategic retrofitting to harness capacity effect costs would result in cost-advantages for torrefied pellets and pyrolysis oil. Further research could focus on the utilisation of more specific feedstock and feedstock flexibility of densification and end-user technologies. In addition, possible effects of decoupling dBEC production from fossil energy consumption and respective technical solutions have to be assessed based on energy price scenarios. The commoditisation process of wood pellets in recent decades could further deliver insights into market-related properties of novel densified bioenergy carriers. Further empirical analysis is required to help explore and optimise the handling, transport and storage of dBECs also with respect to possible pipeline distribution and more cost-effective transport and storage options for hydrophobic torrefied pellets than for conventional pellets while lower public perception of handling of more “coal-like” pellets has to be tackled.

[16] De Meyer A, Snoeck M, Cattrysse D, Van Orshoven J. A reference data model to support biomass supply chain modelling and optimisation. Environ Model Softw 2016;83:1–11. https://doi.org/10.1016/j.envsoft.2016.05.007. [17] Mabee WE, Mirck J. A regional evaluation of potential bioenergy production pathways in Eastern Ontario, Canada. Ann Assoc Am Geogr 2011;101:897–906. https://doi.org/10.1080/00045608.2011.568878. [18] Hoefnagels R, Resch G, Junginger M, Faaij A. International and domestic uses of solid biofuels under different renewable energy support scenarios in the European Union. Appl Energy 2014;131:139–57. https://doi.org/10.1016/j.apenergy.2014. 05.065. [19] De Laporte AV, Weersink AJ, McKenney DW. Effects of supply chain structure and biomass prices on bioenergy feedstock supply. Appl Energy 2016;183:1053–64. https://doi.org/10.1016/j.apenergy.2016.09.049. [20] Li Y, Tittmann P, Parker N, Jenkins B. Economic impact of combined torrefaction and pelletization processes on forestry biomass supply. GCB Bioenergy 2016;n/a-n/ a. https://doi.org/10.1111/gcbb.12375. [21] Miedema JH, Benders RMJ, Moll HC, Pierie F. Renew, reduce or become more efficient? The climate contribution of biomass co-combustion in a coal-fired power plant. Appl Energy 2017;187:873–85. https://doi.org/10.1016/j.apenergy.2016. 11.033. [22] Zimmer T, Rudi A, Müller A-K, Fröhling M, Schultmann F. Modeling the impact of competing utilization paths on biomass-to-liquid (BtL) supply chains. Appl Energy 2017;208:954–71. https://doi.org/10.1016/j.apenergy.2017.09.056. [23] Mauro C, Rentizelas AA, Chinese D. International vs. domestic bioenergy supply chains for co-firing plants: the role of pre-treatment technologies. Renew Energy 2018;119:712–30. https://doi.org/10.1016/j.renene.2017.12.034. [24] Rentizelas AA, Li J. Techno-economic and carbon emissions analysis of biomass torrefaction downstream in international bioenergy supply chains for co-firing. Energy 2016;114:129–42. https://doi.org/10.1016/j.energy.2016.07.159. [25] Haarlemmer G, Boissonnet G, Imbach J, Setier P-A, Peduzzi E. Second generation BtL type biofuels – a production cost analysis. Energy Environ Sci 2012;5:8445. https://doi.org/10.1039/c2ee21750c. [26] Hamelinck CN, Suurs RAA, Faaij APC. International bioenergy transport costs and energy balance. Biomass Bioenergy 2005;29:114–34. [27] Agar DA. A comparative economic analysis of torrefied pellet production based on state-of-the-art pellets. Biomass Bioenergy 2017;97:155–61. https://doi.org/10. 1016/j.biombioe.2016.12.019. [28] Mobini M, Sowlati T, Sokhansanj S. A simulation model for the design and analysis of wood pellet supply chains. Appl Energy 2013;111:1239–49. https://doi.org/10. 1016/j.apenergy.2013.06.026. [29] Adams PWR, Shirley JEJ, McManus MC. Comparative cradle-to-gate life cycle assessment of wood pellet production with torrefaction. Appl Energy 2015;138:367–80. [30] Pirraglia A, Gonzalez R, Denig J, Saloni D. Technical and economic modeling for the production of torrefied lignocellulosic biomass for the U.S. Densified Fuel Industry. BioEnergy Res 2013;6:263–75. https://doi.org/10.1007/s12155-012-9255-6. [31] Sultana A, Kumar A, Harfield D. Development of agri-pellet production cost and optimum size. Bioresour Technol 2010;101:5609–21. https://doi.org/10.1016/j. biortech.2010.02.011. [32] Chen W-H, Peng J, Bi XT. A state-of-the-art review of biomass torrefaction, densification and applications. Renew Sustain Energy Rev 2015;44:847–66. https://doi. org/10.1016/j.rser.2014.12.039. [33] Svanberg M, Olofsson I, Flodén J, Nordin A. Analysing biomass torrefaction supply chain costs. Bioresour Technol 2013;142:287–96. https://doi.org/10.1016/j. biortech.2013.05.048. [34] Uslu A, Faaij APC, Bergman PCA. Pre-treatment technologies, and their effect on international bioenergy supply chain logistics. techno-economic evaluation of torrefaction, fast pyrolysis and pelletisation. Energy 2008;33:1206–23. https://doi. org/10.1016/j.energy.2008.03.007. [35] Mireles IH, van Horssen A, van Harmelen T, Hagen E. BioBoost D6.4 Energy Carrier Chain LCA. Sustainability assessment of energy carriers; 2015. [36] Rotter S, Rohrhofer C. BioBoost Deliverable 4.1, logistic concept; 2014. [37] Obernberger I, Thek G. The Pellet Handbook; 2010. [38] Nicoleit T, Niebel A, Funke A, Lappas A, Kalogiannis K, Michaeilof C, et al.. BioBoost Deliverable 2.6. Final Report of WP2 summarizing the optimization of conversion technology processes: Pyrolysis, catalytic pyrolysis, hydrothermal carbonization; 2016. [39] Arpiainen V, Fagernäs L, de Wild P, Carbo M, Pommer L. SECTOR working paper D3.6; 2014. [40] Koppejan J, Schaubach K, Witt J, Thrän D. SECTOR working paper D10.2; 2015. [41] Hoefnagels R, Junginger M, Resch G, Matzenberger J, Panzer C, Pelkmans L. Development of a tool to model European biomass trade IEA Task 40; 2013. [42] Kim NS, Van Wee B. The relative importance of factors that influence the breakeven distance of intermodal freight transport systems. J Transp Geogr 2011;19:859–75. https://doi.org/10.1016/j.jtrangeo.2010.11.001. [43] Proskurina S, Heinimö J, Mikkilä M, Vakkilainen E. The wood pellet business in Russia with the role of North-West Russian regions: present trends and future challenges. Renew Sustain Energy Rev 2015;51:730–40. https://doi.org/10.1016/j. rser.2015.06.051. [44] Schipfer F, Kranzl L. IEA Bioenergy - Task 40. Country Report for Austria 2014. Vienna University of Technology; 2015. [45] Lamers P, Tan ECD, Searcy EM, Scarlata CJ, Cafferty KG, Jacobson JJ. Strategic supply system design – a holistic evaluation of operational and production cost for a biorefinery supply chain. Biofuels Bioprod Biorefining 2015;9:648–60. https://doi. org/10.1002/bbb.1575. [46] ENTRANZE. ENTRANZE database 2014. http://www.entranze.eu/ (accessed May

Conflicts of interest There are no conflicts of interest to declare. Acknowledgements The presented work was performed under the framework of the FP7 SECTOR Project under grant agreement n°282826 References [1] EUROSTAT. Supply, transformation and consumption of renewable energies - annual data; 2016. http://ec.europa.eu/eurostat/web/energy/data/database (accessed June 17, 2016). [2] Richard TL. Challenges in scaling up biofuels infrastructure. Science 2010;329:793–6. [3] Forsberg G. Biomass energy transport: analysis of bioenergy transport chains using life cycle inventory method. Biomass Bioenergy 2000;19:17–30. https://doi.org/10. 1016/S0961-9534(00)00020-9. [4] Junginger M, Goh CS, Faaij A. International Bioenergy Trade. vol. 17. Dordrecht: Springer Netherlands; 2014. [5] Srirangan K, Akawi L, Moo-Young M, Chou CP. Towards sustainable production of clean energy carriers from biomass resources. Appl Energy 2012;100:172–86. https://doi.org/10.1016/j.apenergy.2012.05.012. [6] Lamers P. Developing the global bioeconomy: technical, market, and environmental lessons from bioenergy. 1st ed. Cambridge, MA: Elsevier; 2016. [7] KIT. Biomass based energy intermediates bossting biofuel production - A European Research project on renewable energies; 2014. [8] Thrän D, Witt J, Schaubach K, Kiel J, Carbo M, Maier J, et al. Moving torrefaction towards market introduction – Technical improvements and economic-environmental assessment along the overall torrefaction supply chain through the SECTOR project. Biomass Bioenergy 2016. https://doi.org/10.1016/j.biombioe.2016.03. 004. [9] Bhutto AW, Qureshi K, Harijan K, Abro R, Abbas T, Bazmi AA, et al. Insight into progress in pre-treatment of lignocellulosic biomass. Energy 2017. https://doi.org/ 10.1016/j.energy.2017.01.005. [10] Olsson O, Lamers P, Schipfer F, Wild M. Chapter 7 - Commoditization of Biomass Markets. Dev. Glob. Bioeconomy, Academic Press; 2016. p. 139–63. [11] Ba BH, Prins C, Prodhon C. Models for optimization and performance evaluation of biomass supply chains: an Operations Research perspective. Renew Energy 2016;87:977–89. https://doi.org/10.1016/j.renene.2015.07.045. [12] Gold S, Seuring S. Supply chain and logistics issues of bio-energy production. J Clean Prod 2011;19:32–42. https://doi.org/10.1016/j.jclepro.2010.08.009. [13] Hu H, Lin T, Wang S, Rodriguez LF. A cyberGIS approach to uncertainty and sensitivity analysis in biomass supply chain optimization. Appl Energy 2017;203:26–40. https://doi.org/10.1016/j.apenergy.2017.03.107. [14] Vukašinović V, Gordić D. Optimization and GIS-based combined approach for the determination of the most cost-effective investments in biomass sector. Appl Energy 2016;178:250–9. https://doi.org/10.1016/j.apenergy.2016.06.037. [15] Mupondwa E, Li X, Tabil L, Phani A, Sokhansanj S, Stumborg M, et al. Technoeconomic analysis of wheat straw densification in the Canadian Prairie Province of Manitoba. Bioresour Technol 2012;110:355–63. https://doi.org/10. 1016/j.biortech.2012.01.100.

723

Applied Energy 239 (2019) 715–724

F. Schipfer, L. Kranzl

19, 2014). [47] Arpiainen V, Alakangas E, Kroon P, Carbo M. SECTOR working paper D3.7; 2015. [48] Meerman JC, Ramírez A, Turkenburg WC, Faaij APC. Performance of simulated flexible integrated gasification polygeneration facilities, Part B: economic evaluation. Renew Sustain Energy Rev 2012;16:6083–102. https://doi.org/10.1016/j. rser.2012.06.030. [49] EUROSTAT. Energy prices - Eurostat 2018. https://ec.europa.eu/eurostat/news/ themes-in-the-spotlight/energy-prices (accessed December 7, 2018). [50] EC. Weekly Oil Bulletin - Energy - European Commission. Energy 2018. /energy/ en/data-analysis/weekly-oil-bulletin (accessed December 7, 2018). [51] Devine-Wright P. Reconsidering public attitudes and public acceptance of renewable energy technologies: a critical review. Manch Sch Environ Dev Univ Manch; 2007. Available http://www Sed Manch Ac Ukresearchbeyondnimbyism. [52] Alakangas E, Asikainen A, Grammelis P, Hämäläinen J, Haslinger W, Janssen R,

[53] [54] [55]

[56] [57]

724

et al. Biomass Technology Roadmap. European Technology Platform on Renewable Heating and Cooling; 2014. Tovey-Fall L. Exploring the changing global supply and demand dynamics of industrial pellets 2018:45. Landälv I. Biofuels potential, production costs and GHG reduction capability Results from SGAB work 2016; 2018. Fleiter T, Steinbach J, Dengler J, Ragwitz M, Köhler B, Reitze F, et al. Mapping and analyses of the current and future (2020-2030) heating/cooling fuel deployment (fossil/renewables). Project for the European Commission; 2016. IEA. World Energy Outlook 2018; 2018. Chiang K-Y, Chien K-L, Lu C-H. Characterization and comparison of biomass produced from various sources: Suggestions for selection of pretreatment technologies in biomass-to-energy. Appl Energy 2012;100:164–71. https://doi.org/10.1016/j. apenergy.2012.06.063.