Achieving carbon-neutral iron and steelmaking in Europe through the deployment of bioenergy with carbon capture and storage

The 30 integrated steel plants operating in the European Union (EU) are among the largest single-point CO2 emitters in the region. The deployment of bioenergy with carbon capture and storage (bio-CCS) could significantly reduce their emission intensities. In detail, the results demonstrate that CO2 emission reduction targets of up to 20% can be met entirely by biomass deployment. A slow CCS technology introduction on top of biomass deployment is expected, as the requirement for emission reduction increases further. Bio-CCS could then be a key technology, particularly in terms of meeting targets above 50%, with CO2 avoidance costs ranging between V60 and V100 tCO2 1 at full-scale deployment. The future of bio-CCS and its utilisation on a larger scale would therefore only be viable if such CO2 avoidance cost were to become economically appealing. Small and medium plants in particular, would economically benefit from sharing CO2 pipeline networks. CO2 transport, however, makes a relatively small contribution to the total CO2 avoidance cost. In the future, the role of bio-CCS in the European iron and steelmaking industry will also be influenced by non-economic conditions, such as regulations, public acceptance, realistic CO2 storage capacity, and the progress of other mitigation technologies. © 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).


42
The European iron and steel industry annually generates over 200 million tons of carbon 43 dioxide (Mt CO 2 ) (Borkent and Beer, 2016), which amounts to 5% of all CO 2 emissions 44 produced across EU-28 countries in 2016 (Eurostat, 2016). The majority of these emissions 45 come from the 30 integrated steel plants that produce 60% of the European steel output 46 Studying the potential of bio-CCS within a large system requires a modelling approach that 138 accounts for the biomass supply chain, the considered industry, and the CCS network. The 139 approach also has to be able to study the interaction between the three systems across the 140 studied time frame, and take into account the spatial distribution of elements as well as the 141 technical limitations that occur when they are applied within the same system. In our previous 142 work using the BeWhere-EU model (IIASA, 2015), we already linked biomass and iron and 143 steel plants in this way (Mandova et al., 2018). This work extends the BeWhere-EU iron & 144 steel model by adding a CCS framework for iron and steel, including CCS linkage to biomass, 145 which provides an opportunity to simultaneously study both the CCS and bio-CCS systems. 146 The section below gives a brief overview of the model, with further information provided in 147 the supplementary material. 148 The BeWhere-EU iron and steel model is written in the General Algebraic Modelling System 149 (GAMS), using Mixed Integer Linear Programming (MILP) and CPLEX as solver. The 150 concept of the model is to split the studied geographic region (EU-28) into equally sized grid-151 cells, each covering an area of 40 km × 40 km. Each grid-cell then contains area-specific 152 information that is important for modelling the system, including: 153 • types, amounts and costs of available feedstock; 154 • existing biomass demand; 155 • distance, mode of transport and biomass transport costs between different grid-cells; 156 • annual CO 2 emissions and energy demand of integrated steel plants; 157 • CO 2 storage potential, as well as CO 2 capture, transport and storage costs. 158 The cost of biomass upgrading, the types of fossil fuels used in an integrated steel plant, and 159 different CO 2 transport network possibilities are also included in the model.

168
As shown in Figure 2, the complexity of the modelled system requires the inclusion of a 169 variety of input data, constraints and internal data calculations. Specifically, the model is 170 composed of three modules, where the core module BeWhere-EU iron & steel is using the 171 outputs of the biomass module (labelled BeWhere-EU) and the CCS module (labelled CO 2 172 TranStorage). In particular, the biomass module is used to subtract the biomass requirement 173 of the existing industries from the total biomass potential. The CCS module has been 174 developed to obtain different CCS infrastructure configurations connecting the plants to 175 potential CO 2 storage sites using a minimum spanning tree algorithm (Hillier, 2012). The core 176 -iron and steel -module connects the two modules and provides outputs specific to the iron 177 and steel industry study. A mathematical description of each module can be found in the 178 supplementary material. Table 1

Biomass supply chain 185
The biomass supply chain considers feedstock supply, transport and upgrading. The total 186 theoretical biomass potential within the EU in 2020 is estimated to be 8.5 EJ year -1 . This 187 potential includes stumps, stemwood and logging residues of coniferous and non-coniferous 188 trees, with costs ranging from €0.20 up to €8.30 GJ -1 (with price depending on the type of 189 wood and country of origin) (Dees et al., 2017). To incorporate biomass sustainability aspects 190 in the modelling, only 70% of the theoretical potential is considered. The model allows inter-191 European biomass trade, as well as biomass imports from non-EU countries to specific 192 harbour locations. The imported biomass from non-EU countries is assigned a cost 20% 193 higher than the average biomass cost in the country where a specific harbour is located, in 194 order to account for additional expenditure due to import taxes and long-distance transport. 195 Biomass harvested outside the EU is generally imported already pre-processed, for example, 196 in the form of pellets. However, as the current work assumes that biomass upgrading to the 197 final product is done on-site of the iron and steel plant, the modelling approach required raw 198 biomass import from outside of the EU. The cost of biomass imports from outside the EU 199 ranges from €3.56 to €6.01 GJ -1 (exact values are available in the supplementary material). 200 Transport of biomass from supply points to demand points is considered by truck, train and 201 ship, with the specific cost of each biomass type approximated on energy basis. Form of 202 transport and the corresponding distances are obtained from spatial data using the network 203 analysis tool in the ArcGIS software. The studied biomass demand includes the pulp and M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 8 paper industry (total of 1.4 EJ year -1 ) (CEPI, 2017), sawmills (1.6 EJ year -1 ) (FAO, 2016) and 205 heat and power plants (1.0 EJ year -1 ) (Platts, 2017). In total, 2.0 EJ year -1 of available 206 biomass potentially suitable for iron and steel production is identified from the biomass 207 module (BeWhere-EU) after meeting the existing demand. The distribution of the available 208 biomass in relation to the 30 integrated steel plants is shown in Figure 3. 209

Technologies for CO 2 emission reduction in integrated steel plants 225
In total, 30 integrated steel plants -the full number of currently operating plants using BF-226 BOF across EU-28 countries -are considered. In order to maintain transparency under limited 227 data availability and confidentiality, this work assumes that each plant has the same 228 technology and structure as a typical West European plant, as described in the IEA 229 Greenhouse Gas (GHG) report (IEAGHG, 2013). The energy demand of each plant is 230 estimated from the plants' annual hot rolled coil (HRC) production. This is obtained from 231 each plant's data on hot metal production in 2016 (VDEh data exchange, 2017), which is then further calibrated so that country specific crude steel production corresponds to data published 233 by the World Steel Association for the same year (World Steel Association, 2017). In 234 addition, it is assumed 1 t of hot metal produces 1.113 t of crude steel and 1.027 of hot rolled 235 coil, as presented in the IEAGHG report (IEAGHG, 2013  The key factors influencing the cost are the pipeline length and the specific CO 2 flow. The 304 CO 2 transport cost estimates also include the cost of compression up to supercritical pressure 305 (above 73.8 bar), investment, operational and maintenance costs, as well as whether it is an 306 onshore or offshore pipeline (IEAGHG, 2005). In addition, the calculation also takes into 307 account the extra CO 2 flow as a result of increasing the amount of CO 2 produced at a plant 308 due to the installation of CCS technology. A further description of the CO 2 pipeline cost 309 calculations can be found in the supplementary material. 310 As mentioned above, only offshore CO 2 storage in saline aquifers or depleted oil and gas 311 fields is considered, with locations around Europe shown in Figure 5

Scenario setting 322
To help answer our questions, we explore a range of scenarios that vary across two 323 dimensions: (1) the CO 2 emission reduction goal to be achieved, and (2) the configuration of 324 the physical CO 2 infrastructure. 325 To study the increasing importance of bio-CCS in the technology mix, we impose European 326 emission reduction targets ranging from 0 up to 100%, with a 5% step level. The analysis 327 focuses only on the CO 2 emissions occurring on-site for the integrated steel plants, in other 328 words, it does not consider the produced emissions during fuel transportation, upgrading or 329 production as such a study would require a detailed Life Cycle Analysis (LCA). The follow 330 up discussion takes place on both plant and country level, in order to evaluate whether any 331 country has an outstanding opportunity for bio-CCS deployment that would be able to 332 significantly reduce CO 2 emissions on its own. 333 To account for the possibility of several plants sharing a CO 2 pipeline system, two CO 2 334 networks, classified as individual or collaborative, are considered ( Figure 6). In both cases, 335 the costs are calculated for a "plateau flow" of CO 2 (a CO 2 pipeline network where all plants 336 start delivering their maximum CO 2 volumes from day one). It is important to note that

The importance of bio-CCS for various CO 2 reduction targets 348
The optimal technology mix to meet different CO 2 emission reduction targets is shown in 349 The range of the CO 2 avoidance costs of bio-CCS is due to different economics behind the 385 deployment of biomass and CCS in each plant. For example, avoiding CO 2 emissions using 386 biomass costs on average €61 t CO 2 -1 at the maximum technically-feasible substitution. For the 387 plant in Romania however, the CO 2 is avoided using biomass at costs as low as €40 t CO 2 The economics of CCS on the other hand, are influenced by the distance of the plants to the 394 storage locations, the amount of CO 2 transported annually, the type of CO 2 storage reservoir, 395 as well as country-specific electricity prices. The resulting average CO 2 emission reduction 396 cost using CCS technology is estimated at €92 t CO 2 -1 avoided. This cost includes the 397 technology investment, as well as the operational cost related to CO 2 capture, transport and its 398 injection into the reservoirs. In general, CCS deployment is the most expensive for plants in 399 Germany and the UK, as the biggest expense related to CCS deployment is the CO 2 capture 400 cost (around 76% of the overall CO 2 avoidance cost), which is heavily influenced by the cost 401 of electricity in the country. 402 Initial biomass substitution is cheaper than the deployment of CCS, as the CO 2 avoidance cost 403 for CCS technology exceeds the CO 2 avoidance cost for initial biomass substitution, as 404 presented in Figure 8. However, plants in the Netherlands and Belgium have CO 2 avoidance 405 costs by bio-CCS that exceed the costs of CCS on its own (€67 t CO 2 -1 and €64 t CO 2 -1 for the 406 Netherlands, and €81 t CO 2 -1 and €71 t CO 2 -1 for Belgium , for bio-CSS and CCS, respectively). 407 In these cases, biomass is economically preferable to CCS for only very low emission 408 reduction levels, and the introduction of CCS on top of biomass is expected even at lower 409 emission targets, before the maximum technically feasible substitution by biomass is  The modelling results demonstrate that bio-CCS can achieve a 100% CO 2 emission reduction 451 across European integrated steel plants. However, these results are related to the emissions 452 occurring only on-site, and rely heavily on the assumption of carbon neutrality of biomass. As 453 emissions of the bio-CCS system are also produced off-site due to land use change, biomass 454 harvesting, transport and upgrading, as well as due to CO 2 capture, transport and storage, iron 455 and steelmaking in Europe would not be carbon-neutral from the whole system perspective. 456 Our findings show that bio-CCS can play a role in achieving carbon-neutrality across these 523 plants when considering only emissions produced on-site. However, bio-CCS would not be an 524 economically favourable option when aiming to reach specific CO 2 emission reduction targets 525 below 20% for which an autonomous deployment of biomass over full bio-CCS is more 526 favourable. Therefore, biomass can be considered a strategic solution for an initial 527 decarbonisation, of which the CO 2 emission reduction potential could be enhanced through 528 the additional deployment of CCS (resulting in bio-CCS), if required. 529 In this study, an average CO 2 avoidance cost using bio-CCS in European iron and steel plants 530 is calculated to €80 t CO 2 -1 . This is indeed a large additional expenditure that would 531 significantly increase the steel production cost of the plants, even for the most suitable ones. 532 The work shows that an initial biomass substitution is cheaper than CCS deployment, but then 533 costs related to the high level of biomass utilisation are similar to the deployment cost of 534 CCS. Despite CO 2 capture accounting for the biggest share of CO 2 avoidance cost by CCS, 535 the opportunities in cost reduction actually emerge in CO 2 transport as plants start sharing 536 CO 2 pipeline networks. Especially for small integrated steel plants, the CO 2 transport cost 537 could be reduced by up to 90%. Opportunities for the reduction of CO 2 capture costs could 538 also occur in the future. Cost of a first-of-a-kind capture plant is usually significantly greater 539 than the cost of a mature nth-of-a-kind (Rubin et al., 2015). This has been demonstrated at, for 540 example, the Shand power plant, based on lessons learnt from the Boundary Dam, or 541 discussed in a work by van den Broek et al. (2009). Hence, there is a high likelihood that the 542 CO 2 avoidance cost of using bio-CCS could be even lower than €80 t CO 2 -1 in the future. 543 However, in the present, a significant cost reduction of bio-CCS is difficult, and the EU has to 544 propose stronger economic incentives that would ensure a competitive iron and steel industry 545 in the EU, if carbon-neutrality using bio-CCS is defined as the way to go. 546 M A N U S C R I P T

22
From specifically a geographical viewpoint, no country presents an outstanding opportunity 547 for bio-CCS. In general, the technology is most likely to be developed in France, the 548 Netherlands, Belgium and in one of the plants in Sweden, since these plants achieve the 549 lowest bio-CCS deployment costs. On the other hand, the least favourable countries are 550 Germany and the UK due to the comparably high costs of CO 2 capture. 551 It is important to mention that if we want bio-CCS to be developed at a large scale in Europe, 552 non-economic barriers of a regulatory-social-environmental nature must also be resolved, or 553 at least accounted for in the policy agenda. Further study is necessary to identify the most 554 essential problems that the EU or specific countries and regions are facing. It is recommended 555 that a sensitivity analysis of the impact of overcoming barriers on the CO 2 avoidance cost for 556 each plant shown in this work be included in such a study.