Prospective techno-economic and environmental assessment of a national hydrogen production mix for road transport

(1)

Contents lists available atScienceDirect

Applied Energy

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

Prospective techno-economic and environmental assessment of a national hydrogen production mix for road transport

Zaira Navas-Anguita

^a,b

, Diego García-Gusano

^a,c

, Javier Dufour

^a,b

, Diego Iribarren

^a,⁎

aSystems Analysis Unit, IMDEA Energy, E-28935 Móstoles, Spain

bChemical and Environmental Engineering Group, Rey Juan Carlos University, E-28933 Móstoles, Spain

cTecnalia Research and Innovation, E-48160 Derio, Spain

H I G H L I G H T S

•

Techno-economic and environmental optimisation of the hydrogen production mix in Spain.

•

Analysis of scenarios using energy systems modelling and carbon footprint restrictions.

•

Steam reforming of natural gas as the main production technology in the short term.

•

Key role of water electrolysis and biomass gasiﬁcation in the medium-to-long term.

•

Savings > 50 Mt CO2eq in 2050 for the scenarios with carbon footprint restrictions.

A R T I C L E I N F O

Keywords:

Carbon footprint Energy systems modelling Fuel cell electric vehicle Hydrogen production mix Life cycle assessment Road transport

A B S T R A C T

Fuel cell electric vehicles arise as an alternative to conventional vehicles in the road transport sector. They could contribute to decarbonising the transport system because they have no direct CO2emissions during the use phase. In fact, the life-cycle environmental performance of hydrogen as a transportation fuel focuses on its production. In this sense, through the case study of Spain, this article prospectively assesses the techno-economic and environmental performance of a national hydrogen production mix by following a methodological framework based on energy systems modelling enriched with endogenous carbon footprint indicators. Taking into account the need for a hydrogen economy based on clean options, alternative scenarios characterised by carbon footprint restrictions with respect to a fossil-based scenario dominated by steam methane reforming are evaluated. In these scenarios, the steam reforming of natural gas still arises as the key hydrogen production technology in the short term, whereas water electrolysis is the main technology in the medium and long term.

Furthermore, in scenarios with very restrictive carbon footprint limits, biomass gasification also appears as a key hydrogen production technology in the long term. In the alternative scenarios assessed, the functional substitution of hydrogen for conventional fossil fuels in the road transport sector could lead to high greenhouse gas emission savings, ranging from 36 to 58 Mt CO2eq in 2050. Overall, thesefindings and the model structure and characterisation developed for the assessment of hydrogen energy scenarios are expected to be relevant not only to the specific case study of Spain but also to analysts and decision-makers in a large number of countries facing similar concerns.

1. Introduction

The current European transport sector annually releases above 1 Gt CO2 eq, which represents around 25% of the total greenhouse gas (GHG) emissions. Within this sector, road transport accounts for over 70% of the sectoral GHG emissions[1], thus heavily contributing to climate change [2]. This is linked to the fact that road transport accounts for 74% of the total energy consumption by the European

Environment Agency member countries[3]. In 2016, 55.6% and 41.2%

of all European passenger vehicles were fuelled by gasoline and diesel, respectively[4]. In this sense, the path towards a clean transport sector requires the implementation of alternative vehicles fuelled by biofuels, electricity, and hydrogen. In particular, the deployment of fuel cell electric vehicles (FCEV) could be a low-carbon alternative[5], especially suitable for long distances as well as for cases with air quality concerns [6]. FCEV employ compressed hydrogen gas as the

https://doi.org/10.1016/j.apenergy.2019.114121

Received 2 August 2019; Received in revised form 8 October 2019; Accepted 11 November 2019

⁎Corresponding author.

E-mail address:[email protected](D. Iribarren).

Applied Energy xxx (xxxx) xxxx

(2)

transportation fuel, which is electrochemically combined with oxygen from the air in the vehicle’s fuel cell with the aim of producing electricity to power the motor[7].

However, the penetration of FCEV in the current transport sector is still low. In 2017, the United States of America and Japan had a stock of several thousands of FCEV. They were distantly followed by Germany and France, with a stock of hundreds of FCEV [8]. In fact, many countries do not yet have a clear position on hydrogen for transport. In this sense, the development of comprehensive energy plans for the road transport sector is needed. Within this context, the use of energy systems modelling (ESM) to support energy planning is often recommended, providing a scientiﬁc basis for the prospective assessment of energy systems according to technical and economic conditions during a time frame[9]. Furthermore, ESM studies could be enhanced through the implementation of life-cycle sustainability indicators in the prospective assessment[10]. For instance, this type of implementation has been applied to case studies of power generation in Norway[11]

and Spain[12].

Since Spain arises as an illustrative country without a well-deﬁned hydrogen strategy, this article aims to develop and apply a national energy systems model focused on hydrogen production in Spain to satisfy its demand for road transport under a number of hypothetical scenarios. The case study of the Spanish road transport sector is especially relevant because the outcomes of the study could be extended to many other countries with a still underdeveloped hydrogen economy.

The Spanish stock of FCEV is almost negligible. In 2016, there were only 11 FCEV in Spain (3 cars and 8 vehicles classiﬁed as ‘other’), all of them linked to demonstration projects[13]. Moreover, there is a lack of prospective studies and roadmaps on FCEV penetration in Spain and the evolution of the national hydrogen production mix. In this regard, national plans such as the National Action Framework for Alternative Energy in Transport just include a rough environmental comparison between FCEV and fossil fuel-based vehicles, the current status of FCEV in Spain, and some information about hydrogen refuelling stations[13].

Regarding the draft National Energy and Climate Plan 2021–2030, even though there is no speciﬁc plan for FCEV penetration in Spain, the goal of 14% renewable energy for transport in 2030 and the purpose of abating direct pollutant emissions from fuel use make hydrogen a key alternative fuel that could eﬀectively contribute to achieving these targets[14].

Hence, this work addresses a prospective techno-economic and environmental assessment of the hydrogen production mix that could satisfy the hydrogen demand for road transport under alternative scenarios on the penetration of FCEV in Spain. After this introduction to the need for sound prospective studies on hydrogen energy,Section 2 focuses on the methodological framework followed, explaining the speciﬁc energy model concept, the deﬁnition of hydrogen production technologies, and their implementation in a national energy systems model enriched with harmonised carbon footprint indicators. Before concluding, Section 3 presents and discusses the results of the prospective study, focusing on the evolution of the hydrogen production technology mix and the corresponding carbon footprint under alternative scenarios.

2. Materials and methods

The goal of this study is to develop a national energy systems model that supports the identiﬁcation of the technologies that could satisfy the hydrogen demand for road transport in Spain in the time frame 2020–2050. To that end, the combined use of two well-known meth- odologies –ESM [15,16]and Life Cycle Assessment (LCA) [17]– has recently been recommended when it comes to prospectively evaluating the techno-economic and environmental performance of energy systems[10]. In this study, a national energy systems model focused on hydrogen production was built using LEAP-OSeMOSYS [18]. This choice was motivated by the availability of a recent LEAP-OSeMOSYS

model for power generation in Spain[10]. As achieved in the Spanish power generation model[10], the endogenous integration of the carbon footprint of the production technologies involved in the new model was also pursued. It should be acknowledged that the stages after hydrogen production (e.g., hydrogen transport, storage, distribution and use) were not considered within the scope of the study. Future studies could include them in the analysis, which would probably require the sepa- rate or integrated use of other tools, e.g. for modelling and selecting speciﬁc features dependent on time and location (centralised/dis- tributed hydrogen production, location of refuelling stations, etc.).

2.1. Modelling hydrogen production for road transport

2.1.1. Modelling road transport

Several authors have used ESM with a focus on the transport sector.

For instance, Nieves et al.[19]analysed the long-term energy demand and GHG emissions associated with the transport sector in Colombia.

Similarly, Zhang et al.[20]analysed the long-term demand, energy consumption and CO2emissions of the transport sector in China and USA. Travesset-Baro et al.[21]developed a prospective model to study the carﬂeet energy consumption in Andorra until 2050. Rocco et al.

[22]illustrated the assessment of the life-cycle consequences of future technological scenarios through a case study of the German road transport sector in 2050. Hong et al.[23]examined the eﬀectiveness of the policies and targets set by the South Korean government in the transport sector. García-Olivares et al.[24]provided a review of current conditions and issues to attain a fully decarbonised state of the global transport system. Yeh et al. [25] compared frameworks and scenario projections from global transportation models, focusing on the fuel used, GHG emissions, and technology and fuel mixes under different scenarios.

In this article, the evolution of the Spanish hydrogen production mix for road transport was explored through a specific energy systems model conceived within a broader representation of the national road transport system. Such a general road transport model has several in- terconnected elements, as shown inFig. 1. A key part of the general model proposed inFig. 1 refers to the identification of, not only the different types of fuel currently used, but also those with future potential to be used (e.g., hydrogen). The demand in energy terms of these fuels is generally determined by the technical characterisation of the end-use technologies, i.e. the types of vehicle considered. In this regard, the demand for transport services is typically affected by socio-economic drivers and expressed per person-km for passenger transport (cars, buses, and motorcycles) and per tonne-km for freight transport (trucks and vans). Once the fuel demand is implemented, the focus is generally placed on the different production technologies involved and the energy resources that they employ. Hence, a good coverage of the technology options relevant to each fuel is essential. This work focuses on hydrogen as the transport fuel.

2.1.2. Modelling hydrogen production

The use of hydrogen for FCEV could contribute to decarbonising the road transport sector. FCEV have zero direct CO2emissions at the use phase, just releasing water from the tailpipe. In fact, from a life-cycle environmental perspective, the focus should be placed on the stages before hydrogen use, e.g. hydrogen production and vehicle infra- structure[26]. According to the goal and scope of the study, the focus of this work was placed on hydrogen production. As shown inFig. 2, hydrogen can be produced from diverse feedstock and energy sources through diﬀerent technologies. Fossil-based options such as the steam reforming of natural gas (steam methane reforming, SMR) and coal gasiﬁcation are mature technologies, but the implementation of CO2

capture systems would be required to make them low-carbon options.

Alternatively, hydrogen could be produced from renewable sources, e.g. through wind power water electrolysis, steam reforming of biofuels, and biomass gasiﬁcation.

(3)

The hydrogen production technologies considered in this ESM study are those shown inFig. 2. During the steam reforming of natural gas or renewable feedstock, steam and hydrocarbons are heated up to 800–1000 °C to produce syngas[27]. Afterwards, the syngas stream undergoes a water gas shift process to increase the hydrogen content [28]. In a subsequent step, hydrogen is separated and puriﬁed, e.g.

through a pressure swing adsorption unit[29]. Coal or biomass gasi- ﬁcation are also thermochemical processes[30]in which the feedstock

is processed at high temperature in a gasiﬁcation medium such as air, oxygen and/or steam to produce syngas[31], which is subsequently processed to hydrogen as mentioned above[32]. Finally, water electrolysis involves a process in which electricity is converted into che- mical energy in the form of hydrogen[33], with oxygen as a by-product [34]. The renewability of this technology option is determined by the electricity used as the energy source. In this sense, when using grid electricity, the country-speciﬁc electricity production mix becomes a Fig. 1. Framework of the general energy systems model for road transport.

Fig. 2. Energy systems model of hydrogen production for road transport.

Z. Navas-Anguita, et al. Applied Energy xxx (xxxx) xxxx

(4)

crucial aspect. There is growing interest in the possibility of using the electricity surplus from renewable power generation facilities such as wind farms and photovoltaic or hydropower plants to produce hydrogen through water electrolysis. For instance, in Spain the renewable electricity surplus means around 2% of the renewable power[35].

The implementation of the hydrogen production technologies in the specific energy systems model requires the quantification of economic data (e.g., investment costs), process efficiencies, and carbon footprints (Section 2.3). The optimisation procedure finds the economically optimal combination of technologies that could satisfy the hydrogen needs under a set of constraints. The hydrogen produced must supply the hydrogen demanded in the time frame. Hence, as detailed inSection 2.2, exploring the national hydrogen demand according to reasonable criteria is of paramount importance.

2.2. Deﬁnition of scenarios

Hydrogen production technologies would satisfy a given demand of fuel to be used in FCEV. Such a demand projection was exogenously implemented in the model. Since hydrogen demand is generally un- certain, the scenarios considered in this study were deﬁned on the basis of alternative hydrogen demands. Reference features of several vehicle categories were considered in order to calculate the hydrogen demand caused by FCEV [36–39]. In this regard,Table 1 presents the share, energy consumption and average annual mileage associated with each FCEV category considered. These values were assumed to remain con- stant during the time frame under evaluation.

In order to estimate the FCEV demand for Spain, national targets for countries with FCEV penetration plans were taken into consideration (Table 2). These national FCEV stock targets were translated into per- centage of FCEV within the nationalfleet of each country (according to [40]for European countries,[41]for the USA, and[42]for Japan), defining the target area shown in Fig. 3. Considering the currently underdeveloped hydrogen economy in Spain, three alternative scenarios of FCEV penetration in Spain were defined (also shown inFig. 3):

LOW, MEDIUM, and HIGH, which correspond to a FCEV penetration of 10%, 15% and 20% of the Spanishﬂeet by 2050, respectively.

FCEV penetration in Spain would imply the hydrogen demand in 2050 presented inTable 3for each scenario. Since the use of hydrogen as a transportation fuel aims to decarbonise the transport system, production technologies should also be evaluated from an environmental point of view[43]. In this sense, the integration of carbon footprints into the prospective assessment paves the way towards sound decision- making processes based on evolved techno-economic and environmental indicators [10]. Moreover, in order to further explore the critical role of the life-cycle global warming performance of hydrogen, the scenario MEDIUM (taken herein as the most representative one) was divided into three sub-scenarios addressing diﬀerent carbon footprint restrictions: MEDIUM_0 (no carbon footprint restriction, i.e. evolved technology production mix based only on techno-economic criteria), MEDIUM_60 (environmental constraint of ≥60% carbon footprint saving with respect to the scenario MEDIUM_0 in 2050), and MEDIUM_80 (≥80% carbon footprint saving with respect to MEDIUM_0 in 2050). These carbon footprint restrictions are in the

range of the targets set for energy products in international directives [44]and other studies[45].

2.3. Techno-economic and environmental data of hydrogen production pathways

As shown inFig. 2, hydrogen can be produced through several pathways. A massive implementation of FCEV will imply changes in the development of hydrogen production technologies according to techno- economic and environmental features which must be implemented in the model. In particular,Tables 4 and 5present the investment cost and the process eﬃciency of the hydrogen production technologies involved in the study, respectively.

Regarding the implementation of water electrolysis in the model, further considerations are needed due to the critical link created between the power generation sector and the fuel production sector. In this sense, when using grid electricity as the energy source of the electrochemical process, the evolution of the national electricity production mix should be considered. In this regard,Table 6presents the evolution of the Spanish electricity production mix implemented in the model, which was computed using the national power generation model developed by García-Gusano et al.[10]. It should be noted that the use of this power generation model allows a consistent implementation of the hydrogen-electricity link due to the fact that the road transport model was built under the same modelling structure (e.g., in terms of the techno-economic parameters selected) and with common assumptions when needed (e.g., biomass availability).Table 6 shows that above 90% of the national electricity production mix would be based on renewable sources by 2050, which suggests that electricity storage will play a role in fulﬁlling the whole demand.

Finally, another singularity of the study is the endogenous integration of the carbon footprint of the hydrogen production technologies into the energy systems model. In this regard, the harmonised carbon footprints of hydrogen produced through the technologies in- cluded in the model were directly retrieved from Navas-Anguita et al.

[46]. It should be noted that the use of harmonised carbon footprints of hydrogen allows robust comparative life-cycle studies and is recommended when addressing alternative hydrogen production technologies[47]. Concerning the carbon footprint of hydrogen produced through grid electrolysis, the evolved carbon footprint of the electricity production mix was considered: 0.050 kg CO2 eq·MJH2−1 in 2020, 0.036 kg CO2eq·MJH2−1in 2030, 0.015 kg CO2eq·MJH2−1in 2040, and 0.014 kg CO2eq·MJH2−1in 2050.

3. Results and discussion

This section focuses on the application of the model detailed in Section 2, showing the key results obtained and thus proving its use- fulness to enhance decision-making processes in theﬁeld of hydrogen energy. Two main outcomes are reported in this section: the evolution of the hydrogen production mix (Section 3.1), and the corresponding prospective carbon footprint (Section 3.2).

Table 1

Values assumed by FCEV category.

Vehicle category

Distribution (%)^a Energy consumption (MJ·km⁻¹)

Annual mileage (km)^b

Cars 82 1.1[36] 10,000

Vans 5 2.1[37] 10,000

Trucks 12 6.7[38] 50,000

Buses 1 13.2[39] 70,000

a Based on[55].

b Based on[48].

Table 2

Expected stock of FCEV in diﬀerent countries.

Country Year 2030 Year 2050 Reference

France 773,000 7,300,000 [50,60]

Germany 1,870,000 – [60–61]

United Kingdom 1,270,000 – [60–61]

Scandinavia – 5,300,000 [61]

United States of America – 53,000,000 [8]

Japan 800,000 20,000,000 [8]

China 1,000,000 – [8]

Korea 630,000 – [8]

(5)

3.1. Evolution of the hydrogen production mix

Fig. 4shows the hydrogen energy demand from 2020 to 2050 under the scenarios LOW, MEDIUM, and HIGH. The demand in theﬁnal year varies from 193 PJ (LOW) to 386 PJ (HIGH). Without any environmental restriction (i.e., scenarios LOW, MEDIUM_0, and HIGH), all hydrogen demand would be completely satisﬁed through conventional, fossil-based SMR as the techno-economically optimal solution. How- ever, the international perspective refers to the establishment of a hydrogen economy based on clean hydrogen. Hence, relying on SMR (natural gas feedstock) as the main source of hydrogen would not be an actual solution in the medium and long term. Taking into account this situation, this section focuses the analysis of the evolution of the

hydrogen production mix on the scenarios MEDIUM_60 and MEDIUM_80, i.e. on the scenarios with a carbon footprint constraint in 2050.

The implementation of a carbon footprint limit gives novelty to the assessment, moving from purely techno-economic optimisation to techno-economic and environmental optimisation. In the combined ﬁeld of ESM and LCA[16], most of the studies available in the literature focus on the evolution of life-cycle indicators according to the techno- economically optimal evolution of a production mix[48], whereas few studies actually use life-cycle indicators within the optimisation problem (an example of this type of study can be found in[49]for the Fig. 3. Deﬁnition of hydrogen scenarios for road transport in Spain.

Table 3

Description of the hydrogen scenarios considered in the study.

Scenario code H2demand in 2050

Remarks

LOW 193 PJ FCEV mean 10% of theﬂeet in 2050

MEDIUM 290 PJ FCEV mean 15% of theﬂeet in 2050

–MEDIUM_0 290 PJ No carbon footprint restriction

–MEDIUM_60 290 PJ ≥60% carbon footprint saving with respect to MEDIUM_0 in 2050

–MEDIUM_80 290 PJ ≥80% carbon footprint saving with respect to MEDIUM_0 in 2050

HIGH 386 PJ 20% of theﬂeet are FCEV in 2050

Table 4

Investment cost of hydrogen production technologies.

Technology Investment cost 2016

(€2017·GJ⁻¹·y)

Investment cost 2030 (€2017·GJ⁻¹·y)

Investment cost 2050 (€2017·GJ⁻¹·y)

References

Steam reforming of natural gas without CO2

capture

8.75 7.33 7.33 [27–29,62–63]

Steam reforming of natural gas with CO2capture 14.50 11.10 9.44 [27–28,63]

Steam reforming of biofuels without CO2capture 30.75 25.77 23.19 [30,64]

Coal gasiﬁcation without CO2capture 9.10 7.63 7.63 [65]

Coal gasiﬁcation with CO2capture 53.75 41.15 34.98 [31,66]

Biomass gasiﬁcation without CO2capture 48.27 21.1 17.94 [62–63,66]

Electrolysis 36.76 15.86 6.10 [33,63,67–68]

Table 5

Net energy eﬃciency of hydrogen production pathways (values in brackets correspond to 2030 estimates).

Technology Eﬃciency (%) References

Steam reforming of natural gas without CO2

capture

76 (85) [30,33,51,62,69]

Steam reforming of natural gas with CO2

capture

65 (70) [32,51]

Steam reforming of biofuels without CO2

capture

69 (72) [30,51]

Coal gasiﬁcation without CO2capture 60 (70) [31,51,66]

Coal gasiﬁcation with CO2capture 56 (60) [31,51]

Biomass gasiﬁcation without CO2capture 43 (46) [30,51,62,70]

Electrolysis 67 (85) [51,71–72]

(6)

implementation of a life cycle-based energy security index in the optimisation problem).

As shown in Figs. 5 and 6, when compared to the mid-demand scenario dominated by SMR (MEDIUM_0), the optimisation of the hydrogen production mix with carbon footprint restrictions (scenarios MEDIUM_60 and MEDIUM_80) significantly affects the results in terms of the evolution of such a mix. Regarding the scenario MEDIUM_60 (≥60% carbon footprint saving in 2050 with respect to MEDIUM_0), SMR and electrolysis were identified as the main technologies that satisfy the hydrogen demand (Fig. 5): the former in the short-to-medium term (i.e., until 2035), and the latter in the medium-to-long term (i.e., from 2035 to 2050). From the year 2034, the contribution of SMR to the hydrogen production mix rapidly falls, completely disappearing before

the time horizon. This extinction of hydrogen from conventional SMR is due to its high carbon footprint according to the limit implemented in the model. In contrast, grid electrolysis emerges as the main hydrogen production technology in the medium-to-long term. This ﬁnding is closely linked to the high share of renewable electricity expected in the national grid (> 90% in 2050), which leads to relatively low carbon footprints of hydrogen from grid electrolysis (Section 2.3).

The major role played by SMR and electrolysis is in agreement with international studies such as the International Energy Agency roadmap [5]. In this regard, it should be noted that key strengths of this article in comparison with other prospective studies such as[5]lie in the traceability of the model and the endogenous integration of a life-cycle in- dicator (carbon footprint).

Table 6

Evolution of the electricity production mix in Spain (GWh).

Electricity production technology Year 2020 Year 2025 Year 2030 Year 2035 Year 2040 Year 2045 Year 2050

Existing oil combustion engine 7011 3505 – – – – –

Existing natural gas combined cycle 36,000 49,956 39,735 – – – –

Existing cogeneration 27,405 22,267 10,277 5138 – – –

Existing nuclear—pressurised water reactor 48,057 – – – – – –

Existing nuclear—boiling water reactor 9566 – – – – – –

Existing hydropower—dam 44,588 44,588 44,588 44,588 44,588 44,588 44,588

Existing wind—onshore 43,481 20,405 4081 – – – –

Existing solar photovoltaics 7998 7035 6071 – – – –

Existing biomass power plant 5841 5667 5493 3793 – – –

Existing waste-to-energy plant 1239 1204 1169 – – – –

Existing biogas power plant 791 663 536 – – – –

New cogeneration 5555 9721 19,441 23,607 27,774 27,774 27,774

New hydropower—dam 432 432 432 432 432 432 432

New hydropower—run-of-river 149 149 149 2136 2136 1987 1987

New wind—onshore 10,407 41,943 73,479 106,713 138,762 168,995 189,746

New wind—oﬀshore 1051 1051 1051 22,083 23,382 24,681 24,666

New solar PV—plant 9106 36,700 55,188 55,188 55,188 56,502 57,816

New solar PV—roof 4914 27,252 28,188 44,676 44,676 45,990 47,304

New solar thermal with storage – – – – – – 6557

New biomass power plant – 4141 13,625 19,736 19,331 17,240 17,278

New waste-to-energy plant 4025 15,768 15,768 15,768 15,768 14,027 15,768

New biogas power plant 237 631 1025 1419 1577 1577 1577

New solid oxide fuel cells – – 86 397 395 429 687

New geothermal power plant – – 2298 6504 9059 12,079 16,378

Total 267,853 293,078 322,680 352,178 383,068 416,301 452,558

Fig. 4. Evolution of the hydrogen energy demand in the scenarios LOW, MEDIUM, and HIGH.

(7)

In the period 2034–2050, a relatively stable contribution of renewable electrolysis to the hydrogen production mix was also found, which corresponds to the assumed surplus of 2% of the annual

renewable electricity produced in Spain (the carbon footprint of hydrogen from electrolysis using surplus renewable electricity was assumed to be zero). Finally, Fig. 5shows that biomass gasiﬁcation is Fig. 5. Evolution of the hydrogen production mix in the scenario MEDIUM_60.

Fig. 6. Evolution of the hydrogen production mix in the scenario MEDIUM_80.

(8)

slightly introduced as an additional hydrogen production technology in thefinal years (2048–2050) in order to fulfil the final carbon footprint target.

In this study, technologies provided with CO2capture did not arise as an effective option for hydrogen production due to their high economic values (Section 2.3). Other prospective studies such as the technology roadmap of the International Energy Agency[5], the French national mobility plan[50]or the updates to parameters of hydrogen production pathways in GREET[51]did identify these variants as selected options for hydrogen production. However, the techno-economic definition and the assumptions behind the models used in these references are not readily available. In this respect, the traceability of the model proposed in this article is expected to benefit other analysts when it comes to developing alternative versions of the model or alternative scenarios according to the specific objectives of a different study.

Finally,Fig. 6shows the evolution of the hydrogen production mix in the scenario MEDIUM_80 (≥80% carbon footprint saving in 2050 with respect to MEDIUM_0). When compared to the results found for SCENARIO_60, the contribution of the hydrogen production technologies in the long term (2040–2050) changes significantly due to the more restrictive carbon footprint limitation. Under very stringent carbon footprint limits (scenario MEDIUM_80), the total disappearance of SMR was found to be hastened (2040 as thefinal year with SMR contribution) and grid electrolysis was found to be partly substituted by biomass gasification in the long term. In this regard, despite the high renewability of the national electricity grid in the medium-to-long term, the achievement of the ambitious target of 80% carbon footprint reduction was found to be conditioned by a growing implementation of biomass gasification in the hydrogen production mix. This relevant role of biomass gasification –also considered in international studies such as[5]–

is closely linked to the very low harmonised carbon footprint of hydrogen produced through this technology[52].

3.2. Prospective harmonised carbon footprint

Fig. 7shows the evolution of the harmonised carbon footprint of the hydrogen production mix in the three mid-demand scenarios. The scenario MEDIUM_0, with no carbon footprint restriction, involves the most unfavourable evolution, accounting for the release of more than 25 Mt CO2eq in 2050. Nevertheless, the same carbon footprint evolution was found in the three scenarios for the period 2020–2032 since the whole hydrogen demand is satisfied through SMR in the three cases during this period. The scenarios MEDIUM_60 and MEDIUM_80 also show a similar carbon footprint evolution in the period 2033–2040, with a relatively stable annual release of ca. 7 Mt CO2eq. This similar performance is linked to the similar hydrogen production mix (SMR and electrolysis) in both scenarios until 2040 (Figs. 5 and 6), whereas a significant worse carbon footprint evolution was found for the scenario MEDIUM_0 as it is completely dominated by SMR during the whole time frame. Finally, in the period 2041–2050, a highly different carbon footprint evolution was found in the three scenarios, which is in line with the significantly different evolution of the hydrogen production mix. In this final period, the scenario MEDIUM_60 (dominated by electrolysis) starts with a 55% carbon footprint reduction with respect to the scenario MEDIUM_0 and reaches a 60% reduction in 2050, while the scenario MEDIUM_80 (dominated by both electrolysis and biomass gasification) starts with a 61% reduction and reaches an 83% reduction in 2050.

A preliminary estimation of the carbon footprint savings associated with the replacement of conventional fuels with hydrogen was also performed in order to further support potential decision- and policy- making processes on the suitability of FCEV deployment in Spain. Thus, Fig. 8shows the quantiﬁcation of the net potential carbon footprint beneﬁts associated with the use of hydrogen for FCEV instead of conventional fossil fuels in the Spanish road transport sector, following the same computation approach as in [48]. Both production [53] and combustion[54]of the avoided fossil fuels were taken into account. It was considered that 57% of the vehicles are fuelled by diesel and 43%

Fig. 7. Evolution of the carbon footprint of the hydrogen production mix in the mid-demand scenarios.

(9)

by gasoline [55], taking into account the diﬀerent types of vehicles involved (cars, vans, trucks, and buses). The average emissions associated are 3.74 kg CO2eq per kg fossil fuel. On average, 1 kg H2in FCEV provides the same transport function as the production and combustion of 7 kg of fossil fuels[56].

As shown inFig. 8for the three mid-demand scenarios, net potential carbon footprint savings grow rapidly when promoting the replacement of diesel and gasoline vehicles with FCEV, reaching high savings in the three scenarios. Nevertheless, even though the scenario MEDIUM_0 involves net savings of ca. 35 Mt CO2eq in 2050, significantly higher savings (above 50 Mt CO2 eq in 2050) could be achieved when im- plementing carbon footprint restrictions (scenarios MEDIUM_60 and MEDIUM_80). In these scenarios with carbon footprint restrictions, the net savings associated with FCEV penetration in Spain are especially high in the medium-to-long term, when a higher hydrogen demand is expected and mainly fulfilled by low-carbon technology options (electrolysis and biomass gasification). Key aspects behind this finding include the replacement of both passenger and freight vehicles and the high renewability of the prospective Spanish electricity production mix.

In fact, the portfolio of vehicles substituted by FCEV potentially leads to a higher avoidance of GHG emissions than that associated with the penetration of hybrid and battery electric vehicles in Spain (ca. 20 Mt CO2eq in 2050) despite the lower number of vehicles involved[48].

This preliminary estimation suggests the suitability of FCEV penetration to decarbonise the transport system. Actually, as a rule of thumb, FCEV would remain suitable provided that the carbon footprint of the substituted fuels is above 1 kg CO2eq per kg of conventional fuel mix. This means a high margin for the suitability of FCEV since dramatic de- creases in the kg CO2eq per kg of avoided fuel mix and/or in the kg of avoided fuel mix per kg of hydrogen are not expected (even considering a potential substitution of a fuel mix dominated by natural gas).

3.3. Final remarks

This is theﬁrst study addressing the prospective techno-economic

and carbon footprint assessment of the hydrogen production mix within the Spanish road transport system. Despite the specific geographical scope of the study (Spain), thefindings of this analysis could be ex- trapolated to other countries facing similar concerns (e.g., the lack of a national plan for the deployment of hydrogen energy systems, including FCEV). Moreover, beyond thefindings of the analysis, the structure and data behind the model proposed (costs, efficiencies, carbon footprints, etc.) could be used by other analysts to develop similar models and explore further scenarios, even when addressing countries with a different situation.

On the other hand, it should be noted that–as in any ESM study– the study would benefit from the use of refined values that could be available in the future (e.g., in terms of current and projected costs and efficiencies of the different technologies), as well as from further dis- aggregation of the technology portfolio (e.g., further distinction between electrolysis technologies such as proton exchange membrane electrolysis[57]and solid oxide electrolysis[58], among others[59]).

Nevertheless, in comparison with other prospective studies on FCEV with a life-cycle perspective[22]and–in general– other ESM studies on transport systems (focused on the transport energy demand[21]or the GHG emission reduction of the transport sector [23]), the focus on hydrogen production and the varied technology portfolio are singular aspects of this novel study.

Overall, FCEV could be one of the solutions to eﬀectively mitigate the impacts associated with the high fossil dependence of the transport sector in countries such as Spain. However, with the aim of thoroughly supporting energy planning and decision-making processes, further studies are needed that enlarge the scope of the analysis to include not only hydrogen production but also other stages such as storage, distribution, and use. Furthermore, future studies could consider the implementation of other assumptions, (life-cycle) indicators, and scenarios.

Fig. 8. Net potential carbon footprint savings in the mid-demand scenarios.

(10)

4. Conclusions

Under merely techno-economic criteria, the hydrogen demand associated with the future penetration of FCEV in the Spanish road transport system would be completely satisfied through conventional steam reforming of natural gas due to the competitive prices of natural gas and the maturity of the technology. However, the international perspective on hydrogen energy systems refers to a hydrogen economy based on clean options. In this regard, in hydrogen energy scenarios with carbon footprint restrictions using harmonised life-cycle impacts, the role of conventional SMR would be limited to the short-to-medium term, facilitating the penetration of hydrogen as an alternative fuel in Spain. In the medium-to-long term, electrolysis would arise as a key technology to fulfil the hydrogen demand. Furthermore, under stringent carbon footprint restrictions, biomass gasification would also emerge as a key hydrogen production technology in the long term.

While the life-cycle GHG emissions of producing hydrogen only via SMR in Spain would exceed 25 Mt CO2eq in 2050 under a mid-demand scenario, the application of carbon footprint restrictions would lead to signiﬁcantly lower emissions (below 11 Mt CO2 eq in 2050 for the scenarios assessed). Furthermore, the functional replacement of conventional fossil fuels with hydrogen in the road transport sector was estimated to involve high net GHG emission savings, above 50 Mt CO2

eq in 2050 for the scenarios with carbon footprint restrictions. Finally, beyond these ﬁndings, the model structure and characterisation developed herein for the prospective techno-economic and carbon footprint assessment of hydrogen energy scenarios is expected to be relevant not only to Spanish actors but also to analysts and decision- makers in many other countries, contributing to the overall goal of sustainable energy systems, climate change mitigation, and environmental pollution reduction.

Declaration of Competing Interest

The authors declare no conﬂict of interest.

Acknowledgements

This research has been partly supported by the Spanish Ministry of Economy, Industry and Competitiveness (ENE2015-74607-JIN AEI/

FEDER/UE).

References

[1] European Commission. A European strategy for low-emission mobility,https://ec.

europa.eu/clima/policies/transport_en; 2017 [accessed April 17, 2019].

[2] European Environment Agency. Greenhouse gas emissions from transport,https://

www.eea.europa.eu/data-and-maps/indicators/transport-emissions-of-greenhouse- gases/transport-emissions-of-greenhouse-gases-11; 2018 [accessed May 6, 2019].

[3] European Environment Agency. Final energy consumption by mode of transport.

EEA: Luxembourg; 2018.

[4] European Automobile Manufacturers Association. Passenger carﬂeet by fuel type, https://www.acea.be/statistics/tag/category/passenger-car-ﬂeet-by-fuel-type;

2016 [accessed April 29, 2019].

[5] International Energy Agency. Technology Roadmap. Hydrogen Fuel Cells. Paris:

IEA/ETSAP; 2015.

[6] Roland Berger. Development of Business Cases for Fuel Cells and Hydrogen Applications for Regions and Cities - Consolidated technology introduction dossiers.

Brussels: FCH JU; 2017.

[7] International Energy Agency. Hydrogen. Tracking Clean Energy Progress.https://

www.iea.org/tcep/energyintegration/hydrogen/; 2018 [accessed October 10, 2018].

[8] International Energy Agency (IEA). Hydrogen. Tracking Clean Energy Progress 2018.https://www.iea.org/tcep/energyintegration/hydrogen/; 2018 [accessed October 10, 2018].

[9] Bhattacharyya SC, Timilsina GR. A review of energy system models. Int J Energy Sect Manage 2010;4:494–518.https://doi.org/10.1108/17506221011092742.

[10] García-Gusano D, Martín-Gamboa M, Iribarren D, Dufour J. Prospective analysis of life-cycle indicators through endogenous integration into a national power generation model. Resources 2016;5:39.https://doi.org/10.3390/resources5040039.

[11] García-Gusano D, Iribarren D, Martín-Gamboa M, Dufour J, Espegren K, Lind A.

Integration of life-cycle indicators into energy optimisation models: the case study

of power generation in Norway. J Clean Prod 2016;112:2693–6.https://doi.org/10.

1016/j.jclepro.2015.10.075.

[12] García-Gusano D, Iribarren D, Dufour J. Is coal extension a sensible option for energy planning? A combined energy systems modelling and life cycle assessment approach. Energy Policy 2018;114:413–21.https://doi.org/10.1016/j.enpol.2017.

12.038.

[13] Interministerial Group for Coordination of the National Action Framework for Alternative Energy in Transport. National action framework for alternative energy in transport. Market development and deployment of alternative fuels infra- structure. Madrid: Government of Spain; 2016.

[14] Interministerial Group for Ecological Transition. Draft national plan of energy and climate 2021–2030. Madrid: Government of Spain; 2019.

[15] Lund H, Mathiesen BV, Christensen P, Schmidt JH. Energy system analysis of marginal electricity supply in consequential LCA. Int J Life Cycle Assess 2010;15:260–71.https://doi.org/10.1007/s11367-010-0164-7.

[16] Vandepaer L, Gibon T. The integration of energy scenarios into LCA: LCM2017 Conference Workshop, Luxembourg, September 5, 2017. Int J Life Cycle Assess 2018;23:970–7.https://doi.org/10.1007/s11367-017-1435-3.

[17] Ekvall T. Cleaner production tools: LCA and beyond. J Clean Prod 2002;10:403–6.

https://doi.org/10.1016/S0959-6526(02)00026-4.

[18] Heaps CG. Long-range Energy Alternatives Planning (LEAP) System (Software Version 2017.0.11.0) n.d.https://www.energycommunity.org[accessed May 8, 2019].

[19] Nieves JA, Aristizábal AJ, Dyner I, Báez O, Ospina DH. Energy demand and greenhouse gas emissions analysis in Colombia: a LEAP model application. Energy 2019;169:380–97.https://doi.org/10.1016/j.energy.2018.12.051.

[20] Zhang H, Chen W, Huang W. TIMES modelling of transport sector in China and USA: Comparisons from a decarbonization perspective. Appl Energy 2016;162:1505–14.https://doi.org/10.1016/j.apenergy.2015.08.124.

[21] Travesset-Baro O, Gallachóir BPÓ, Jover E, Rosas-Casals M. Transport energy demand in Andorra. Assessing private car futures through sensitivity and scenario analysis. Energy Policy 2016;96:78–92.https://doi.org/10.1016/j.enpol.2016.05.

041.

[22] Rocco MV, Casalegno A, Colombo E. Modelling road transport technologies in future scenarios: Theoretical comparison and application of Well-to-Wheels and Input-Output analyses. Appl Energy 2018;232:583–97.https://doi.org/10.1016/j.

apenergy.2018.09.222.

[23] Hong S, Chung Y, Kim J, Chun D. Analysis on the level of contribution to the national greenhouse gas reduction target in Korean transportation sector using LEAP model. Renew Sustain Energy Rev 2016;60:549–59.https://doi.org/10.1016/j.rser.

2015.12.164.

[24] García-Olivares A, Solé J, Osychenko O. Transportation in a 100% renewable energy system. Energy Convers Manage 2018;158:266–85.https://doi.org/10.1016/j.

enconman.2017.12.053.

[25] Yeh S, Mishra GS, Fulton L, Kyle P, McCollum DL, Miller J, et al. Detailed assessment of global transport-energy models’ structures and projections. Transp Res Part D Transp Environ 2017;55:294–309.https://doi.org/10.1016/j.trd.2016.11.001.

[26] Valente A, Iribarren D, Candelaresi D, Spazzafumo G, Dufour J. Using harmonised life-cycle indicators to explore the role of hydrogen in the environmental performance of fuel cell electric vehicles. Int J Hydrogen Energy; in press.http://doi.org/

10.1016/j.ijhydene.2019.09.059.

[27] Keipi T, Tolvanen H, Konttinen J. Economic analysis of hydrogen production by methane thermal decomposition: Comparison to competing technologies. Energy Convers Manage 2018;159:264–73.https://doi.org/10.1016/j.enconman.2017.12.

063.

[28] Collodi G, Azzaro G, Ferrari N, Santos S. Techno-economic Evaluation of Deploying CCS in SMR Based Merchant H2Production with NG as Feedstock and Fuel. Energy Procedia 2017;114:2690–712.https://doi.org/10.1016/j.egypro.2017.03.1533.

[29] Mondal KC, Ramesh Chandran S. Evaluation of the economic impact of hydrogen production by methane decomposition with steam reforming of methane process.

Int J Hydrogen Energy 2014;39:9670–4.https://doi.org/10.1016/j.ijhydene.2014.

04.087.

[30] United States Department of Energy. Hydrogen production. Fuel cell technologies oﬃce – multi-year research, development, and demonstration plan. Washington:

FCTO; 2015.

[31] Hart D, Howes J, Lehner F. Scenarios for deployment of hydrogen in contributing to meeting carbon budgets and the 2050 target. Lausanne: E4tech; 2015.

[32] International Energy Agency. Hydrogen production distribution. Paris: OECD/ IEA;

2007.

[33] United States Department of Energy. Hydrogen production: electrolysis.https://

www.energy.gov/eere/fuelcells/hydrogen-production-electrolysis; 2017 [accessed April 13, 2018].

[34] United States Department of Energy. Hydrogen production: electrolysis 2017.

https://www.energy.gov/eere/fuelcells/hydrogen-production-electrolysis(accessed April 13, 2018).

[35] WindEurope. WindEurope views on curtailment of wind power and its links to priority dispatch. Brussels: WindEurope; 2016.

[36] Toyota Mirai.https://newsroom.toyota.eu/download/540469/toyota-mirai-dpl-es.

pdf; 2015 [accessed December 4, 2018].

[37] Roland Berger. Development of business cases for fuel cells and hydrogen applications for regions and cities: FCH delivery vans. Brussels: FCH JU; 2017.

[38] Roland Berger. Development of business cases for fuel cells and hydrogen applications for regions and cities: FCH heavy-duty trucks. Brussels: FCH JU; 2017.

[39] Pocard N, Reid C. Fuel Cell Electric Buses : an attractive value proposition for zero- emission buses in the United Kingdom. London: Ballard; 2016.

[40] European Automobile Manufacturers Association Vehicles in use Europe 2017.

(11)

Brussels: ACEA; 2017.

[41] Statista. Number of motor vehicles in the United States from 1990 to 2017.https://

www.statista.com/statistics/183505/number-of-vehicles-in-the-united-states-since- 1990/; 2019 [accessed January 14, 2019].

[42] Statista. Total number of motor vehicles in use in Japan from 2008 to 2017.https://

www.statista.com/statistics/674574/japan-motor-vehicle-in-use-numbers/; 2019 [accessed January 14, 2019].

[43] Fernández-Dacosta C, Shen L, Schakel W, Ramirez A, Kramer GJ. Potential and challenges of low-carbon energy options: Comparative assessment of alternative fuels for the transport sector. Appl Energy 2019;236:590–606.https://doi.org/10.

1016/j.apenergy.2018.11.055.

[44] Union European. Directive (EU) 2018/2001 of the European Parliament and of the Council of 11 December 2018 on the promotion of the use of energy from renewable sources. Brussels: European Union; 2018.

[45] Ogden J, Jaﬀe AM, Scheitrum D, McDonald Z, Miller M. Natural gas as a bridge to hydrogen transportation fuel: Insights from the literature. Energy Policy 2018;115:317–29.https://doi.org/10.1016/j.enpol.2017.12.049.

[46] Navas-Anguita Z, García-Gusano D, Iribarren D. A review of techno-economic data for road transportation fuels. Renew Sustain Energy Rev 2019;112:11–26.https://

doi.org/10.1016/j.rser.2019.05.041.

[47] Valente A, Iribarren D, Dufour J. Harmonised life-cycle global warming impact of renewable hydrogen. J Clean Prod 2017;149:762–72.https://doi.org/10.1016/j.

jclepro.2017.02.163.

[48] Navas-Anguita Z, García-Gusano D, Iribarren D. Prospective life cycle assessment of the increased electricity demand associated with the penetration of electric vehicles in Spain. Energies 2018;11:1185.https://doi.org/10.3390/en11051185.

[49] García-Gusano D, Iribarren D. Prospective energy security scenarios in Spain: The future role of renewable power generation technologies and climate change im- plications. Renew Energy 2018;126:202–9.https://doi.org/10.1016/j.renene.2018.

03.044.

[50] Mobilité Hydrogène France. H2 Mobilité Hydrogène France. Study for a Fuel Cell Electric Vehicle national deployment plan. Paris: Mobilité Hydrogène France; 2016.

[51] Elgowainy A, Han J, Zhu H. Updates to parameters of hydrogen production pathways in GREET^TM. Chicago: Argonne National Laboratory; 2013.

[52] Valente A, Iribarren D, Dufour J. Harmonising methodological choices in life cycle assessment of hydrogen: A focus on acidiﬁcation and renewable hydrogen. Int J Hydrogen Energy 2019;44:19426–33.https://doi.org/10.1016/j.ijhydene.2018.03.

101.

[53] Weidema BP, Bauer C, Hischier R, Mutel C, Nemecek T, Reinhard J, et al. Overview and methodology– data quality guideline for the ecoinvent database version 3. St.

Gallen: The ecoinvent Centre; 2013.

[54] European Environment Agency. Air pollutant emission inventory guidebook.

Luxembourg: EMEP/EEA; 2016.

[55] Historical Records—Vehicle Fleet.http://www.dgt.es/es/seguridad-vial/

estadisticas-e-indicadores/parque-vehiculos/; 2019 [accessed January 15, 2018].

[56] Simons A. Road transport: new life cycle inventories for fossil-fuelled passenger cars and non-exhaust emissions in ecoinvent v3. Int J Life Cycle Assess

2016;21:1299–313.https://doi.org/10.1007/s11367-013-0642-9.

[57] Bareiβ K, de la Rua C, Möckl M, Hamacher T. Life cycle assessment of hydrogen from proton exchange membrane water electrolysis in future energy systems. Appl Energy 2019;237:862–72.https://doi.org/10.1016/j.apenergy.2019.01.001.

[58] AlZahrani AA, Dincer I. Modeling and performance optimization of a solid oxide electrolysis system for hydrogen production. Appl Energy 2018;225:471–85.

https://doi.org/10.1016/j.apenergy.2018.04.124.

[59] Bhandari R, Trudewind CA, Zapp P. Life cycle assessment of hydrogen production via electrolysis– a review. J Clean Prod 2014;85:151–63.https://doi.org/10.1016/

j.jclepro.2013.07.048.

[60] Adolf J, Balzer CH, Louis J, Schabla U, Fischedick M, Arnold K, et al. Energy of the future? Sustainable Mobility through Fuel Cells and H2. La Haya: Shell; 2018.

[61] CertifHy. Overview of the market segmentation for hydrogen across potential cus- tomer groups, based on key application areas. Brussels: CertifHy; 2015.

[62] Susmozas AI. Analysis of energy systems based on biomass gasiﬁcation. Móstoles:

Rey Juan Carlos University; 2015.

[63] Hydrogen Europe. Hydrogen, enabling a zero emission Europe. Technology Roadmaps Full Pack. Brussels: Hydrogen Europe; 2018.

[64] Song H, Ozkan US. Economic analysis of hydrogen production through a bio- ethanol steam reforming process: Sensitivity analyses and cost estimations. Int J Hydrogen Energy 2010;35:127–34.https://doi.org/10.1016/j.ijhydene.2009.10.

043.

[65] Sciazko M, Chmielniak T. Cost estimates of coal gasiﬁcation for chemicals and motor fuels. Gasif Pract Appl 2012:247–76.https://doi.org/10.5772/48556.

[66] Dodds PE, Mcdowall W. A review of hydrogen production technologies for energy system models. London: UCL Energy Institute; 2012.

[67] Schmidt O, Gambhir A, Staﬀell I, Hawkes A, Nelson J. Future cost and performance of water electrolysis: an expert elicitation study. Int J Hydrogen Energy 2017;42:30470–92.https://doi.org/10.1016/j.ijhydene.2017.10.045.

[68] National Renewable Energy Laboratory. Hydrogen Production Cost Analysis.

https://www.nrel.gov/hydrogen/production-cost-analysis.html#fn5; 2011 [accessed January 16, 2019].

[69] Nikolaidis P, Poullikkas A. A comparative overview of hydrogen production processes. Renew Sustain Energy Rev 2017;67:597–611.https://doi.org/10.1016/j.

rser.2016.09.044.

[70] National Energy Technology Laboratory. 2014 Technology Readiness Assessment - A Chekpoint along a Challenging Journey. Albany: DOE/NETL; 2015.

[71] United States Department of Energy. Hydrogen: a clean,ﬂexible energy carrier.

https://www.energy.gov/eere/articles/hydrogen-clean-ﬂexible-energy-carrier;

2017 [accessed April 10, 2018].

[72] Ursúa A, Gandía LM, Sanchis P. Hydrogen production from water electrolysis:

current status and future trends. Proc IEEE 2012;100:410–26.https://doi.org/10.

1109/JPROC.2011.2156750.