Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Zero-emission public transit could be a catalyst for decarbonization of the transportation and power sectors

Zero-emission public transit could be a catalyst for decarbonization of the transportation and... Despite small overall share of vehicles... transit bus Heavy-heavy Heavy duty Light/med duty CO ... public transit infrastructure investment can support the wider transition to sustainable, carbon-neutral economies. Keywords: carbon neutrality; electric vehicles; fuel-cell vehicles; hydrogen; infrastructure; net-zero emissions; public transit; transition Received: 13 May 2021; Accepted: 6 August 2021 © The Author(s) 2021. Published by Oxford University Press on behalf of National Institute of Clean-and-Low-Carbon Energy This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (http:// creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 Ku et al. | 493 The primary players are individual transit agencies, who Introduction bear responsibility for decisions on bus types and the in- Numerous countries, local jurisdictions and private com- frastructure investments to operate their fleets, original panies have made carbon-neutrality pledges, with time equipment manufacturers who have to ramp up produc- horizons ranging from 2025 to 2060 [1, 2]. The trans- tion to meet accelerating demand for vehicles and fuel portation sector currently generates about a quarter of suppliers (e.g. electric utilities, industrial gas companies) the global anthropogenic CO; a transition away from who operate capital-intensive businesses and make com- petroleum-based fuels will be an important part of efforts mercial decisions based on market signals. Cost signals to reach net-zero emission targets [3]. Public transit fleets from fuel providers influence transit-agency decisions, account for a relatively small fraction of the total vehicle and vice versa. Early in the transition, decisions by pri- and energy requirements for fuelling, but they can be an mary players can be made in a relatively independent important lever for governments to increase public accept- manner. As the transition progresses, interactions across ance of new technology, support the maturation of vehicle the supply chain lead to uncertainty in the planning land- supply chains, promote the deployment of a large-scale scape, increased risk in capital investment, constraints resilient refuelling infrastructure and experiment with arising from lock-in effects from earlier decisions, con- different policy frameworks [4–12]. Moreover, ambitious flicting objectives among stakeholders and possible schedules for a zero-emission bus (ZEB) fleet conversion bottlenecks in the adequacy and reliability of energy de- can help to uncover emergent challenges related to energy livery [10, 17, 24]. However, positive interactions across the supply and delivery, as these activities interact with con- landscape may also emerge. current efforts to reduce emissions from the power sector. This Perspective article argues that there exist pockets Electrification has emerged as a leading option for of ‘predictability’ that can be used to reduce investment decarbonizing ground transportation. The leading com- risk and therefore be leveraged to help close the gap be- mercial options are battery electric vehicles (BEVs) and tween public and private financing. For the sake of this dis- hydrogen fuel-cell vehicles (HFCVs). In light-duty pas- cussion, we define predictability as an operational attribute senger automobile markets, BEVs currently outnumber that is conducive to regular and sustained capital recovery. HFCVs; in medium- and heavy-duty applications, mixed The economic value of stable revenue streams is already fleets exist and vehicle selection is based on the relative well appreciated; our contribution is to point out an under - advantages and limitations of the technology for different appreciated and significant opportunity to leverage public use cases. Public transit agencies can opt for battery elec- funds invested for specific and local infrastructure needs tric buses (BEBs) or hydrogen fuel-cell buses (HFCBs). BEB to stimulate broader investment across the energy and fleets are relatively easy to pilot on small scales and will transportation sectors during the transition period. directly benefit as electric grids decarbonize. As fleets grow, To illustrate the point, we draw on our collective ex- the overnight demand from their fixed charging schedules periences in working to decarbonize public transit over might exacerbate energy-supply challenges for grids with the past decade. Using public transit in California as a high intermittent renewables generation (e.g. solar, wind) case study, we present a thought experiment exploring [13–18]. Specifically, as growth in BEVs accelerates, the how investments in public transit might catalyse broader energy-supply effects from charging vast numbers of these transformation across the transportation and power sec- vehicles will result in significant changes to the traditional tors. California currently operates >10  000 buses across ‘duck curve’, with resulting consequences still unknown. >200 transit agencies, and has committed to fully transi- A  critical, related issue and concern is the resiliency as- tion to a ZEB fleet by 2040 [25, 26]. In addition, the state sociated with heavy reliance on grid power as affected by has announced goals of 250  000 BEV fast chargers and natural disasters, such as earthquakes, wildfires and hurri- 200 hydrogen-refuelling stations by 2025, 5  million total canes. HFCB fleets offer greater range, lower weight, faster zero-emission vehicles on the roads and a 60% share of refuelling times and higher resiliency, but the high capital renewable electricity generation by 2030, a ban on sales of cost of refuelling stations and nascent supply chains for internal-combustion passenger vehicles starting in 2035 zero-carbon hydrogen pose significant barriers during the and carbon neutrality by 2045 [27]. The scale of public early stages of adoption [19, 20]. The supply issue might be transit and the variety in transit agencies across the state solved by using electrolysis powered by renewable energy; makes California an especially informative example. Our a ‘power-to-gas’ (P2G) approach may help to manage inter- analysis takes advantage of the relative clarity in problem mittency on the electric grid and provide long-term energy definition (i.e. articulated policy commitments and tar - storage [21, 22]. gets, well-defined segment of the transportation sector Progress towards zero-emission public transit requires with predictable refuelling profiles), availability of rele- cooperation across a range of stakeholders. Governments vant public data (i.e. transit-agency operating profiles and provide the overall policy framework and motivate action transition roadmaps, electricity supply and demand) and through regulations and incentives. They also exercise complementary studies of transition pathways in other convening power to gather other stakeholders and pro- sectors (i.e. light and heavy vehicles, electricity, industry) mote the sharing of information and experience [23]. to examine this issue in detail [28–30]. Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 494 | Clean Energy, 2021, Vol. 5, No. 3 The paper is organized around two questions. First, pm to 4 am) ranged from 90 to 120 GWh/night. For context, what will it take to deliver energy in the right form at the the successful introduction of 5 million light-duty BEVs by right times? Second, how can the necessary investments 2030 would increase the daily demand by an additional in energy delivery for public transit fleets be used to sup- 34–70 GWh/day (viz. 5M LDV x 0.22–0.45 kWh/mi/LDV x 31 port broader decarbonization efforts? We begin with a mi/day). California’s current grid has the ability to satisfy statewide energy balance to quantify the energy-delivery the incremental demand from an all-BEB fleet in 2040, but requirements at different stages of the transition; we also ~75–90% of night-time generation comes from natural gas, examine energy demand from ZEB use profiles at the level nuclear and imports that are expected to be phased out of individual transit agencies. We then provide a brief intro- over the same time horizon during which ZEB demand duction, for non-expert readers, of the investment needs emerges [32–34]. Renewable portfolio standards requiring for the refuelling infrastructure of two leading options for 60% renewable generation by 2030, en route to a net-zero BEB and HFCB fleets. This leads to a discussion on how the grid by 2045, mean investment in both zero-carbon night- predictable and long-term nature of the refuelling profile time generation and energy storage might be needed to might be used to support the scale-up of the fuelling in- ensure adequate electricity supply [26]. frastructure in a manner that can benefit the wider efforts California’s energy infrastructure will also need to to decarbonize the transportation sector and electric grid. evolve to deliver sufficient zero-carbon H for electrified Two examples are given to illustrate how this might be ac- transportation. Over 95% of H production in the USA cur - complished in California. The first involves the expansion rently comes from steam methane reforming (SMR). In a of overnight electricity supply in support of BEBs and the net-zero future, H production will need to shift towards second relates to establishing medium-scale low-carbon alternatives such as biogas reforming, SMR with CO cap- H supply for HFCB fleets. Although we focus on the spe- ture and storage (CCS) or electrolysis using electricity cific case of public transit in California, the possibility of from zero-carbon sources [21, 30]. Given its large share mobilizing capital using large, predictable revenue streams in California’s generation mix, the use of solar power in could be applied more generally; we conclude with some a P2G capacity could become part of the long-term solu- observations on how the lessons from California might be tion [21]. Assuming a daily H demand of 16–30 kg/HFCB/ extended to other locations. day, statewide supply would need to reach 160–290 tons per day (tpd) to support a fleet of 10  000 FCBs. Curtailed solar energy in 2019 ranged from 0.5 (August) to 7.3 GWh/ day (April and May), with an average of ~2.5 GWh/day. 1 State-level energy supply and delivery Assuming 55 kWh/kg for electrolysis and compression, requirements this corresponds to an H potential of ~45 tpd, but monthly Fig. 1a shows the size and share of the public transit bus variations in solar power generation means production fleet relative to the entire vehicle population in California. will vary between 8 and 133 tpd [32]. Curtailed solar power Despite rapid projected growth in the absolute number of is forecasted to grow to ~15 GWh in 2030 and 100 GWh in ZEBs, they would only account for <1% of all vehicles after 2045 [35]. In this scenario, P2G-derived H could meet the full adoption in 2040. For context, success in deploying forecasted demand from HFCBs while also supporting up 5 million total zero-emission vehicles (ZEVs) by 2030 would to 1 million light-duty HFCVs by 2045. For context, 43 retail result in an ~15% vehicle share. In this study, we perform a H -refuelling stations had a total capacity to deliver 12 tpd thought experiment around the two limiting cases of fully at the end of 2019 [36]. BEB or HFCB fleets by 2040. Mixed fleets favouring BEBs are Regardless of fleet composition, California has the re- more likely, but this approach allows us to investigate a sources to generate adequate energy for both types of ZEB full range of possibilities with respect to bus technology fleets throughout the entire transition and beyond 2040. [31]. Despite the limited numbers of vehicles in absolute The timely mobilization of capital investment in delivery terms, the regimented service schedules of ZEBs create a infrastructure is more likely to be the limiting factor. significant and predictable energy demand. Fig. 1b shows Beyond the investment in fleets themselves, BEBs will re- the energy footprint for different vehicle types. The height, quire energy storage or additional night-time generation, width and area of each box correspond to the energy in- while HFCBs will need a supply chain capable of delivering tensity per mile (with the range indicating the average ~100 tons per day of clean H along with the construction of and maximum values reported), average daily mileage and refuelling stations. Investment in electricity-transmission total daily energy demand per vehicle [25, 29]. and H -distribution networks will also be needed. Fig. 1c overlays the electricity demand for overnight charging of BEBs and the solar-power-based electrolysis 2 Local energy-delivery requirements production of H in support of HFCBs relative to current generation profiles. The aggregate electrical demand The actual path to full ZEB adoption involves decisions by during overnight hours is projected to be 2.6–4.8 GWh/ individual transit agencies concerning bus technology and night by 2040 (viz. 10k BEB x 2–3.6 kWh/mi/BEB x 132 mi/ schedule; these determine the nature and timing for infra- day). In 2020, the average overnight power generation (10 structure projects. Some agencies have already conducted Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 Ku et al. | 495 A C Vehicle Population (CA, 2019) Future cumulative ZEB energy use vs 2019 generation profile transit bus 1.6M 10K 30M BED overnight charging (2019 total, 10 pm to 4 am) max avg min Light and medium duty Heavy duty Heavy-heavy Incremental energy 2.0 demand (10k BEB) Estimated ZEV energy use (per vehicle) (3.6) 2.0 1.0 12:00 AM 6:00 AM 12:00 PM 6:00 PM 12:00 AM 3.0 (2.5) (1.9) 0.55 2.0 0.22 (0.74) (0.45) 1.0 H from solar P2G production 0.0 (2019 solar, 8 am to 4 pm)* Daily 31 33 74 186 132 mileage LDV HDV HDV HDV HDV (mi/d ) MDV (L) (M) (H) (BEB) 9.1 Incremental max energy (22) 10 (21) 6.7 demand avg 4.2 (10k HFCB) (9.1) (6.7) 0.8 min (1.0) 31 33 74 186 132 Daily 12:00 AM 6:00 AM 12:00 PM 6:00 PM 12:00 AM LDV HDV HDV HDV HDV mileage * P2G = 55 kWh/kg MDV (L) (M) (H) (FCEB) (mi/d ) Fig. 1: Energy impacts of a zero-emission California public transit fleet. (a) Size and share of transit bus fleet relative to other vehicles. (b) Estimated ZEV energy use per vehicle, with the width of each box on the x-axis showing the average daily mileage, and the average and maximum reported energy efficiencies shown on the y-axis. Ranges for energy efficiency are from refs [25] and [29]. The area of each box corresponds to the daily energy use per vehicle (kWh/d or kg H /day). (c) Comparison of overnight electricity demand for BEBs and H using solar-based P2G (at 55 kWh/kg for elec- 2 2 trolysis and compression) to current generating profiles (average, maximum and minimum monthly averages). The x-axis shows the time of day for each 24-h period and the y-axis shows the power generated; the curves show the average, monthly maximum and minimum, generation at 5-min intervals. The unshaded region in the upper plot indicates the amount of power available to support real-time overnight BEB charging. The unshaded region in the lower plot corresponds to the solar generation that can be used to support P2G H production; the curves were calculated using 2019 historical power data published by CAISO [32]. Overlaid boxes show the relative magnitude of future energy demand versus existing generating cap- acity. The boxes in the plots corresponds to the total energy needed to support 10 000 BEBs (upper plot) or produce sufficient H to support 10 000 HFCBs (lower plot). See the online Supplementary Information for details. field evaluation of ZEBs, published strategic roadmaps and current fleet size and average route mileage of transit fleets; begun transitioning their fleets [37]; others are still in the the marker size indicates the total mileage travelled within planning stage. As of December 2020, 10 transit agencies each county and is proportional to the energy demand. On had filed formal plans and another five had announced this basis, Los Angeles and Alameda counties stand out roadmaps for their ZEB transitions. An important feature owing to their larger fleet size and longer average route dis- that has emerged is the role of route distance and fleet tance, respectively. The shaded inset in Fig. 2a is enlarged size in the decision between BEBs and HFCBs. In an ideal in Fig. 2b to highlight use profiles for the next eight largest world without fuel supply or capital investment barriers, county fleets. Together, the top 10 counties account for transit agencies with longer routes and larger fleets would >83% of the total miles travelled by public transit buses on tend towards HFCBs due to operating-cost advantages. The an average day (Fig. 2c)—indicative of the outsized impact exact threshold at which HFCBs become favourable varies of urban areas. The data also show use profiles in the other based on additional factors such as the frequency of stops, 39 counties with transit agencies that must also comply elevation changes, climate and vehicle speeds (see the on- with state targets. Distributions of average route distances line Supplementary Information) [37]. by county and bus population are shown in Fig. 2d and e. Fig. 2 compares use profiles across transit fleets at the Buses in ~80% of counties travel on average <100 mi/ county level. Fig. 2a and b show differences based on the day, but the skew associated with urban areas means that HFCV efficiency BEV efficiency (kg H /100 mi ) (kWh/mi ) Solar generation (GW) Generation (GW) Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 496 | Clean Energy, 2021, Vol. 5, No. 3 ~90% of the buses in California reach this average daily near larger fleets may be able to benefit from infrastruc- mileage. Counties with large fleets (>200 buses) may host ture investment and expertise from their neighbours. multiple transit agencies and the large fleet sizes allow the possibility of hybrid fleets with both BEBs and HFCBs. 3 Infrastructure investment for zero- Medium-sized fleets tend to support moderate population emission public transit centres or large service areas. These fleets are expected The transition to ZEB fleets will require capital investment to be a single bus-technology type due to the desire for across the entire energy supply chain. Fig. 3 summarizes transit agencies to simplify the fuelling logistics and main- the types of investment needed and the relevant actors tenance operations. Small fleets account for only 10% of for each area: transit authorities for bus- and refuelling- the total bus population, but operate in 26 counties; two- station decisions; and utilities and energy providers for thirds of these counties operate fleets with <50 buses. Most energy-delivery decisions. The top panel shows the supply have limited resources to support transition planning and chain for BEBs, where electricity from selected zero- fleet-conversion decisions for this group are likely to be carbon sources is delivered via bus-recharging stations. An constrained by logistics considerations. However, counties Total mileage CDF by county A C Daily mileage per bus vs fleet size 100% 75% 120 50% AC 25% OC 0% SD 0% 20%40% 60%80% 100% LA Fraction of counties Sac SF D Mileage CDF by county 0 1000 2000 3000 4000 5000 Fleet size (buses) Daily mileage per bus vs fleet size 0% 20%40% 60%80% 100% Fraction of counties SBr AC Mileage CDF by bus SCI SD OC SM Sac SF 40 AC LA OC SD 0 300 600 900 1200 SF Fleet size (buses) 0 0% 20%40% 60%80% 100% Fraction of buses Fig. 2: County-level segmentation of bus fleets and route profiles. (a) Daily mileage vs fleet size for 43 counties with transit agencies operating buses. For labelled counties, the areas of the circles are proportional to the total miles driven. (b) Expanded view of daily mileage vs fleet size, using the same shading scheme as in the first panel. (c) Cumulative distribution function (CDF) for share of total miles/day, segmented by county. The x-axis indicates the fraction of counties and the y-axis shows the cumulative share of daily miles driven across the entire state. The top 20% of counties account for ~75% of miles driven. (d) CDF for average daily mileage, by county. (e) CDF for average daily mileage versus number of buses across the entire state. Five counties are shown for illustration: San Francisco (SF), Orange (OC), San Diego (SD), Los Angeles (LA) and Alameda (AC). Data from ref. [25]. Average daily mileage (mi/day/bus) Average daily mileage (mi/day/bus) Cumulative share of Average daily mileage Average daily mileage total miles/day (%) (miles/day/bus) (miles/day/bus) Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 Ku et al. | 497 important feature in station costs is the non-linear invest- Among the options for central production, SMR facilities ment profile as the fleet size grows; a new transmission with CCS will take time to develop, leaving biogas-derived infrastructure (e.g. substations) may be required to deliver H as the leading option in the near term. Air Liquide is in adequate power for the fast charging of large fleets. The the process of commissioning a 30-tpd biogas-to-liquid H bottom panel shows the supply chain for HFCBs. There are facility in Las Vegas, NV [44]; by 2030, aggregate demand two primary paths for station design, based on whether from ZEVs in California could consume all of the supply the fuel is stored as a compressed gas or a cryogenic liquid, from multiple facilities of this scale. Depending on project and three representative options for production and distri- details and distance from transit-agency depots, electri- bution [38, 39]. city production for a BEB division (50 MWh/night) would Bus fleets are the largest capital expense: ~$100 million is require $5–20 million and zero-carbon H supply (2–3 tpd) needed for a single division, increasing proportionally with could range from $6 to $15 million. These values would in- larger numbers of divisions. Some of this required capital will crease by an order of magnitude for 10 divisions and an- have already been committed for the routine replacement other order of magnitude to support the entire ZEB fleet of existing assets. Since ZEBs cost ~$200–500  k more than in 2040. diesel buses, the incremental cost over the existing capital commitments would be ~$20–30  million for a division be- yond what would be spent under business-as-usual replace- 4 Public transit demand as a catalyst for ment schedules. ZEB fleets will also require the construction investment of BEB-charging or HFCB-refuelling stations. Capital invest- The central hypothesis of this study is that investments ment for stations varies depending on technology; we es- made to support energy delivery for public transit can timate a depot-based fast-charging facility for a full BEB also aid efforts to decarbonize the electricity grid and division could require $5–14 million, while an HFCB station broader transportation sector. The predictable and could be less expensive, ranging from $2 to $6  million [14, long-term energy demand from ZEB fleets is a valu- 39]. During a phased adoption in which buses are added over able attribute that can be used to facilitate long-term multiple years, HFCB fleets are more expensive to operate planning for grid evolution and also mobilize capital initially because stations must be constructed at their full- for infrastructure development. Increasing variability service capacity; however, they become more affordable as from growth in renewable-energy generation and expan- the number of buses in the fleet increases [31]. California has sion from electrified transportation will tend to amplify recognized this barrier and offers policy support to help fund the reliability challenges for the electricity grid. Whilst the construction of H stations for early adopters. Over time, time-of-use demand-response programmes can help to market mechanisms will need to provide the additional $20– spread charging loads throughout the day for passenger 100 million needed for 10 divisions and $200 million–$1 bil- vehicles or commercial fleets, there may be limits to lion needed for full conversion by 2040 [28]. their ability to fully rebalance energy requirements for Capital projects will also be needed to upgrade night-time charging [18, 24, 45]. With regard to capital California’s electricity-transmission and H -delivery net- mobilization, the predictability of demand from BEB works. Expansion of night-time electricity capacity to sup- fleets could allow electric utilities to negotiate commer - port BEB fleets could include combinations of generation, cial supply contracts (e.g. power purchase agreements) storage and transmission, as shown in Fig. 3. For electric designed around stable overnight-charging schedules. charging, we focus only on the depot option; interested Predictability in H-refuelling demand could be leveraged readers can consult an extensive literature on alternate in a similar way. In both cases, long-term contracts offer charging options [6, 11, 13–17]. H supply chains can be con- secured revenue streams that can be used for capital re- figured for local production at refuelling stations or central pro- covery in the financing of infrastructure projects, as il- duction with distribution. The first option occurs entirely at lustrated schematically in Fig. 4T . able 1 shows estimates the station and involves electrolysis, small-scale SMR, gas of how much capital might be mobilized by such supply compression, high-pressure ground storage and, possibly, agreements; assuming simple amortization schedules, refrigeration. Electrolysis is currently more expensive than a single 100-bus division could support in the order of alternatives; cost-reduction roadmaps have been devel- $10–50  million in capital investment. Applied to 10  000 oped, but it could take a decade to fully realize the savings ZEBs at 100 bus divisions across California, this has the [40]. For production scales and transport distances relevant potential to mobilize $1–5 billion for infrastructure pro- to California, supply chains using centralized production jects; on a national basis across the USA, the sums could with liquid-H (LH ) distribution is increasingly favoured by 2 2 approach an order of magnitude larger. industry and offers competitive economics. The costs from liquefaction are more than offset by savings from the re- duced cost of distribution and simplification of station de- 4.1 Example 1: overnight charging for BEB fleets sign (i.e. liquid pumping in lieu of gas compression, direct filling to reduce ground storage, heat integration with li- Historical patterns in California wind generation show quid vaporization to reduce refrigeration) [39, 41–43]. peak power in the evening through to the early morning Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 498 | Clean Energy, 2021, Vol. 5, No. 3 Transit agencies Energy providers Bus fleet Recharging station Electricity transmission Electricity production Solar: $2000/kW Storage: $250/kWh (NREL, 2020) $4-10M $25-60M/100 km (150 or 350 kW fast charger) NGCC + CCS $1-10M (MISO, 2019) $2400/kW 100 bus division (upgrade) (650 MW) HFCB Capital cost factors: CO (NETL, 2016) 2020: $0.8-1.2M/bus $5-30M (new) - rated voltage 2030: $0.5-0.6M/bus (MISO, 2019) - structures (Transit agencies, 2020; Wind: $1400/kW - land/rights costs BEB DOE, 2020; BNEF, 2018) Storage: $250/kWh (NREL, 2020) Fleet size (buses) Bus fleet H deliveryH supply Refueling station 2 2 Gaseous H 2 $0.9-3.4M/tpd (0.4 to 50 tpd) (see SI) $4-8M (3 tpd station) (ANL HDRSAM, 2020) $60-80M/100 km $0.5-0.6M/tpd (US DOE EERE, 2018) CH 4 (380 tpd) 100 bus division (NREL, 2018) CO 2020: $0.9-1.4M/bus Liquid H $1-3M/tpd (10-30 tpd ) 2030: $0.6M/bus $1.7-2.5M/tpd (50 tpd) $1.1M (4-5 tH trailer) (Transit agencies, 2020; $2-5M (3 tpd station) $3-11M/tpd (7.5 tpd) (US DOE-EERE, 2016) DOE, 2019; 2012) (ANL HDRSAM, 2020) (CEC, 2020) CO BEB or HFCB Electrical Transmission Solar + storage NGCC + CCS Wind + storage Fast charging station 100 bus division Substation lines CH CO Gas compressor Liquid pump Liquefier Pipeline Biogas Electrolyzer + cascade storage + cascade storage + tanker SMR + CCS + reformer + refrigeration + refrigeration Fig. 3: Estimated ranges for capital investment in BEB and HFCB supply chains. Investment options for transit agencies and energy providers in BEB and HFCB supply chains. The options illustrated here are indicative and not exhaustive. See the online Supplementary Information for details. hours [46]. Wind generation in 2019 was ~14 TWh from an require ~1.6 GWh/night. (A detailed discussion of aggre- installed base of ~6 GW, with generation during the winter gated wind generation and BEB demand is included as on- months about half that of summer production. For con- line Supplementary Information.) While the overbuilding text, the California Energy Commission (CEC) expects that of wind to ensure adequate supply year-round is not in- 3 GW of wind capacity will be added between 2020 and consistent with CEC forecasts, a more robust approach 2030. This begs the question of whether this resource can is to invest in energy storage under the assumption that help meet the overnight-charging demand, initially for the daily energy generation is adequate to meet transit public transit and in the longer term for electrified trans- needs. Fig. 5b shows the cumulative distribution function portation generally. for the estimated incremental overnight wind generation Fig. 5a shows the seasonal variation in hourly gen- in 2030, assuming a profile similar to historical generation eration at the four most productive wind power sites in (2019). Assuming an hourly generation profile consistent California, which accounted for 90% of the wind gener - with 2019, <1.6 GWh/night of added generation would be ation of the state in 2019 [47, 48]. A  fully electrified ZEB expected during 17% of the year (63 nights). Fig. 5c and d fleet in the Bay Area and greater Los Angeles would shows the effect of storage on overnight reliability and Legend HFCB BCB Capital cost Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 Ku et al. | 499 Transit agency Total cost of operation (transit agency, $/mi) TA operating revenues Incentives Operating Transit agency Fuel cost costs capital recovery Predictable demand allows long-term contracts Fuel price Capital for financing energy supply Operations Supplier infrastructure (incl profits) capital recovery Capital markets Energy supplier Fig. 4: Use of predictable fuelling schedules to support capital mobilization for energy infrastructure. From the transit-agency perspective, the total cost of operation (top bar) is the sum of the operating revenues and government incentives (second bar) as well as the sum of the operating costs (third bar). Operating costs include contributions from fuel, other transit-agency operating costs and capital-recovery costs for transit-agency in- vestments. Fuel purchased from external providers (fourth bar), in turn, supports the operating costs for energy suppliers and capital recovery for infrastructure projects (fifth bar). Predictable demand allows the negotiations of long-term contracts that can be used by energy providers to mo- bilize larger sums for capital financing. Table 1: Estimates of capital that could be supported by energy contracts Financing terms 8% 15% 8% 15% (rate of return [%], period [yrs]) 10 yrs 10 yrs 20 yrs 20 yrs BEB 100-bus division $15 million $11 million $22 million $14 million Short routes: 200 MWh/night Energy sales value: $30/MWh 100-bus division $37 million $27 million $54 million $34 million Long routes: 500 MWh/night Energy sales value: $30/MWh HFCB 100-bus division $15 million $11 million $22 million $14 million 3 tpd; energy value: $2/kg the investment required for different levels of reliability This simplified analysis uses 2019 data. As such, it does not (number of days in a year in which wind generation can account for annual variations in wind generation. Moreover, cover the incremental charging demand) from different the productivity of added wind capacity may be lower, as the levels of energy storage. Fig. 5c shows the effect of energy most productive sites are developed first. These effects are ex- storage on reliability. This use profile indicates that energy pected to impact the exact number of days on which gener - storage in support of BEB demand would be needed for ation is inadequate, but the general range is expected to be ~17% of the year, primarily during the winter months and consistent with the results obtained from our initial estimate. occasionally on days with below-average wind generation. Further work will be needed to identify specific projects and During the remaining days, the facilities would be avail- fully quantify their benefits for renewable power integration. able as a general capability to support grid stability. Fig. 5d shows the value of investment in energy storage on re- 4.2 Example 2: Medium-scale zero-carbon H liability. Revenue from bus energy-supply contracts could supply for fuel-cell vehicle fleets support investment at a level of $90–170 million (cf. T able 1, A key difference between BEB and HFCB supply chains is the scaled to 1.6 GWh/night); spending this on energy storage timing of infrastructure deployment. Whereas BEB-charging could increase reliability to ~93%. Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 500 | Clean Energy, 2021, Vol. 5, No. 3 AB Variability in wind generation Incremental wind generation by 2030 (2019 wind, 10 pm to 4 am) due to forecasted growth Q3 Q2 Avg Q1 Q4 0 0 0 20 40 60 80 12:00 AM 6:00 AM 12:00 PM 6:00 PM 12:00 AM Ranked day C D Energy storage capacity needed Impact of energy storage to satisfy overnight demand investment on reliability 100% 95% 300 90% 85% 80% 0 400 800 1200 1600 80% 85% 90% 95% 100% Installed ES capacity (MWh) Reliability (% days/yr demand met) Fig. 5: Prospects for meeting overnight BEB demand using wind and energy storage. (a) Quarterly variability in daily wind-generation profiles in California, plotted at 5-min resolution. (b) CDF for incremental wind generation forecasted by 2030. The x-axis shows days, in ascending order. The y-axis shows the incremental generation potentially available from a 3-GW increase in wind capacity. The dotted line indicates the threshold needed to meet the 1.6-GWh/night incremental demand from all-BEB fleets in the Bay Area and Southern California; a gap exists for 63 nights of the year. (c) Impact of increasing energy-storage capacity (x-axis) on reliability (y-axis), as indicated by the number of days on which the overnight-charging demand would be satisfied. (d) Investment (y-axis) needed to deploy sufficient energy storage to satisfy different reliability thresholds (x-axis), at an assumed energy-storage cost of $250/kWh. The shaded band indicates the level of investment ($90–170 million) potentially available from transit- agency-backed contracts. facilities can be built in a stepwise manner, H supply chains Contributions from the Low Carbon Fuel Standard (LCFS) require significant upfront investment. Since full conversion are considered for each case. For the on-site electrolysis to ZEB fleets by 2040 requires 100% of bus purchases and de- case shown in Fig. 6b, electricity from the grid still includes liveries by 2028, a medium-scale H supply chain is necessary a significant contribution from fossil energy; we assume an to preserve the option of HFCB adoption in the long term. LCFS credit of ~$2/kg assuming a basis of 33% renewable H Fig. 6a compares two options for clean H supply in the and $199/tCO (2020 average) [49, 50]. Assuming financing 2 2 intermediate term. The first involves on-site electrolysis to scenarios from Table 1, ~$3/kg of the revenue would be meet the needs of individual stations. While this is pos- needed to support capital recovery for the refuelling station sible for small-scale pilots involving <20 buses, full-scale and the electrolyser. In this case, the transit agency is re- deployment for a 1000-bus division will be challenging due sponsible for the infrastructure and is the main beneficiary. to footprint and electricity-delivery constraints in many The biogas-H with LH distribution case uses a dif- 2 2 urban areas. The second option uses biogas-to-H with LH ferent arrangement. Here, H is produced at a central 2 2 2 distribution. This path requires central production facil- facility, liquefied and delivered to the station. Fig. 6c dif- ities of ≥9 tpd (enough to supply three bus divisions) for ferentiates between the revenues available to the transit economics, but these facilities can also support broader agency and the H supplier. As with on-site electrolysis, demand from the transportation sector. The Air Liquide the transit agency allocates revenue and incentives to Las Vegas project suggests that this path is being taken operating costs and capital recovery; one difference is that seriously by industry [44]. the transit agency buys fuel from the H supplier, which Fig. 6b and c illustrates how capital recovery can be sup- then allocates sales revenue to its own operations and ported by the value streams from refuelling operations. capital recovery for the production and liquefaction in- The top bar in each case shows the revenue accrued by the frastructure costs. A second difference is the higher LCFS transit agency from operations and government incentives. credit; biogas-derived H is 100% renewable and qualifies Reliability (% of days/yr demand met) Wind generation (GW) Energy storage Incremental wind Investment ($M) generation by 2030 (GWh/night) Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 Ku et al. | 501 AC Medium-scale zero-carbon H supply 2 Option 2. Biogas-H + Liquid delivery TA Operating revenues LCFS credit ($4/kg) Option 1. Option 2. On-site electrolysis Biogas-H + Liquid delivery 2 Station Fuel cost costs H supply primarily Expandable supply supports single TA 10 tpd central facility For a 100-bus division Financing potential 3 tpd onsite facility (support 3-4 divisions) (3 tpd station), (serves 1 division) TA covers station costs $1/kg (allocated) TA pays for station H supplier(s) pay for and electrolyzer biogas-H production LH refueling 10-20 yrs; 8 to 15% and liquifier/tankers station $3-6 M $6-11 M B Option 1. On-site electrolysis Long-term fuel contract LCFS credit TA Operating revenues ($2/kg) Fuel price Operations Supplier Station Electrolyzer + station operating costs capital recovery (incl profits) capital recovery Supply for 3 bus divisions Financing potential Financing potential For a 100-bus division (3 tpd station), (prorated 3 tpd from 9 tpd facility) $1/kg + $2/kg $1/kg + $4/kg (allocated) (LCFS) (allocated) (LCFS) Onsite cgH Refueling 10 yrs ; 8 to 15% 10-20 yrs ; 8 to 15% station electrolyzers Liquefier + tanker Biogas-H $4-8 M+ $3-10 M $17-33 M $4-10 M+ $9-33 M $28-54 M Fig. 6: Financing pathways for medium-scale zero-carbon H supply chains. (a) Overview of two options for supplying clean H . (b) Financing poten- 2 2 tial for on-site electrolysis. The top bar represents the revenue from transit-agency operations and government (LCFS) credits; the second bar shows allocations among station-operating costs and capital recovery. The capital-recovery potential is calculated using scenarios frT om able 1 and the panel shows that ~$3/kg allocated to capital recovery could cover the capital costs associated with the approach. (c) Financing potential for biogas-H with liquid delivery. The top two bars indicate the revenue and cost allocations from the perspective of the transit agency. In the near term, the higher renewable content of biogas-derived H corresponds to a larger LCFS credit. A fraction of the total value is used by the transit agency to sup- port capital recovery for the refuelling station. The lower two bars show the revenue available to H suppliers. The revenue available is simply the fuel price, and it is allocated to operations (including profits) and capital recovery among the suppliers. The fuel price is a negotiated value between the transit agency and energy supplier. Predictability can facilitate long-term arrangements, which can mobilize larger sums of capital; the larger LCFS credit is accrued by the transit agency (as the user of the fuel) but affords flexibility in setting the price. This figure shows a situation in which the value from the LCFS is effectively passed along to H suppliers to help offset their capital requirements. for a credit of $4/kg (2020 average), and this value can be competitive alternatives as cost and regulatory hurdles are effectively allocated between the transit agency and H addressed. As with the BEB example, predictable demand, supplier through supply contract negotiations. Although sustained over decades, from transit agencies is a key en- the value of LCFS credits is expected to decline over time, abler for investments in infrastructure that can simultan- their current value provides an avenue for intermediate- eously meet immediate needs and act as a bridge to larger, term development. Predictability of the fuelling profile is system-wide changes. important in that it allows the longer-term agreements that can extend time horizons and access to the longer 5 Broader implications and conclusion capital-recovery scenarios. The numbers in Fig. 6c reflect an allocation of revenues that can support near-term de- Although we have focused on the situation in California, velopment of this type of supply chain. In contrast to the the predictability of energy demand from large ZEV fleets distributed-electrolysis approach, this option is compat- can be a useful lever for promoting transitions to carbon ible with future growth of the H  supply chain. Once an ini- neutrality in other locales. There are two complementary tial supply chain has been established, market forces can principles: the leveraging of large, predictable, long-term augment it in response to the growing HFCV population aggregate demand from vehicle fleets to support capital and improving technology. Over time, electrolysis with recovery in the financing of infrastructure projects; and on-site storage for P2G and SMR with CCS may become the alignment of projects to support broader system-level Transit agency value map H suppliers(s) Transit agency value map Value map Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 502 | Clean Energy, 2021, Vol. 5, No. 3 evolution. This approach could be particularly relevant for Third, government support can take different forms. urban areas in North America, Europe and East Asia where California is aggressively pursuing carbon neutrality across commitments to carbon neutrality intersect with suffi- multiple sectors using a combination of mechanisms. ciently large public transit systems or corporate fleets to Within this context, some counties have taken concrete support meaningful numbers of projects. We note four nu- steps to pilot ZEBs and develop detailed transition plans ances that should be considered in applying these prin- and roll-out schedules coordinated across multiple public ciples to other situations. agencies (see the online Supplementary Information for First, scale matters. Transportation and energy infra- details); others are still in the early stages of defining their structures typically involve large projects; the associated specific roadmaps. An even wider range of postures exists capital investments are generally tens of millions of dol- across the jurisdictions that can apply the principles iden- lars or more (cf. Fig. 3). Large fleets will be needed to gen- tified in this study. For example, Shenzhen, China, com- erate sufficient revenue to support capital recovery for this pleted the transition of its entire bus fleet of >16  000 magnitude of investment. In the California example, we vehicles to electric buses in 2018 [4]. This was accom- focused on bus divisions with 100 ZEBs. This is a natural plished through significant government support across the scale relevant to transit-agency operation that also appears entire supply chain, leverage of the unique regulatory en- capable of generating appropriate levels of revenue; many vironment in China [5, 52]. At the other extreme, there are larger projects will require cooperation between multiple jurisdictions that have announced commitments but are bus divisions. In 2018, the UN identified 42 metropolises still in the early stages of defining their path forward. For with populations of >5 million (along with 264 areas with municipalities in this latter group, calibrating across the populations of >1  million) across North America, Europe examples from California and other leading locales could and East Asia [51]. The size of the public bus fleets in these provide guidance for their planning activities. cities is more than sufficient to participate in the financing Finally, commercial fleets might also be able to con- model suggested here, and most of these cities have made tribute. Revenue streams from commercial fleets would direct commitments to climate action or are covered by be backed by business operations, so the financing terms national-level pledges. and time horizons would depend on the details of each Second, predictability can help in multiple ways. In the situation and could be more aggressive than for public California example, predictability offered benefits on two transit. Assuming financing terms and vehicle-use profiles levels. On one level, well-defined energy-demand profiles similar to the California situation (cf. Fig. 1b and Table 1), in the form of stable, daily charging schedules for BEBs or fleet sizes in the order of 3400–3900 light-duty (passenger known growth rates in cumulative H demand over time cars), 340–360 medium/heavy-duty trucks and 70–100 can be used by energy providers to plan for operations and Class 8 heavy-duty trucks would have similar energy de- growth. Coordination across the supply chain is a natural mand and revenue potential. In reality, commercial fleets consequence due to the mutual benefits. This can be as operate under a wide range of conditions, so these thresh- simple as sharing information between transit agencies, olds are indicative; separate analysis will be needed for energy suppliers and government about strategic inten- each situation. tions in a market-oriented framework or involve active Reaching carbon neutrality will require signifi- participation by government through public incentives cant changes across multiple sectors of the economy. for private capital decisions. Specific projects can then be Interactions, within and across sectors, can create feed- selected to preserve or even increase strategic flexibility back loops that introduce complexity and make the tran- across the energy system. On a second level, the long-term sition more difficult. However, these interactions are not nature of public transit demand unlocks the ability to fi- always negative and this Perspective article has attempted nance capital over extended periods. Government agen- to show situations in which interactions can be used to aid cies play a critical role in establishing markets and have a the transition to a more sustainable future. vested interest in the creation of mechanisms that engage private capital markets. The public sector plays a key role in setting transformational targets and providing the seed Supplementary data funding to nucleate action; mechanisms that can de-risk Supplementary data is available at Clean Energy online. investments, such as the one presented in this paper, can help to mobilize private capital to supplement public funds. While transit agencies are subject to the budgetary prior - Data availability ities of government and their capital expenses are often Data and analysis for figures and tables are included as online supported by subsidies (e.g. federal grants, sales taxes, Supplementary Information. Further information and requests for parcel taxes, and road and bridge tolls), public transit is resources and reagents should be directed to and will be fulfilled by seen as a public good that provides stability to its revenue the Lead Contact, Anthony Ku (anthonyku@nicenergy.com). Data streams. Although the specific details related to these two on vehicle populations were obtained from the California Transit facets will vary, the general principles should translate Authority and verified at individual transit-agency websites. Data across jurisdictions around the world. on energy production, including capacity and generation, were Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 Ku et al. | 503 obtained from the California Energy Commission and California com/sites/default/files/2020-07/Electrification%20%26%20 ISO (CAISO) websites. Other data were obtained from references Infrastructure%20Guidebook-Final_06.29.20.pdf (23 December as cited in the text. 2020, date last accessed). [14] Chen  Z, Yin  Y, Song  Z. A cost-competitiveness analysis of charging infrastructure for electric bus operations. Transportation Research, Part C: Emerging Technologies, 2018, Conflict of interest statement 93:351–366. [15] Lajunen A. Lifecycle costs and charging requirements of elec- None declared. tric buses with different charging methods. Journal of Cleaner Production, 2018, 172:56–67. References [16] He  Y, Song  Z, Liu  Z. Fast-charging station deployment for battery electric bus systems considering electricity demand [1] National, Regional, and Local Government Climate Pledges. charges. Sustainable Cities and Society, 2019, 48:101530. Paris Agreement Nationally Defined Contributions. https://unfccc. [17] An  K. Battery electric bus infrastructure planning under de- int/process-and-meetings/the-paris-agreement/nationally- mand uncertainty. Transportation Research, Part C: Emerging determined-contributions-ndcs/nationally-determined- Technologies, 2020, 111:572–587. contributions-ndcsClimate . Ambition Alliance: Net Zero by 2050. [18] van Triel F, Lipman TE. Modeling the future California electri- https://unfccc.int/climate-action/race-to-zero-campaign (23 city grid and renewable energy integration with electric ve- December 2020, date last accessed). hicles. Energies, 2020, 13:5277. [2] Organizational Climate Pledges. Net Zero Carbon by 2040. [19] Reddi  K, Mintz  M, Elgowainy  A, et  al. Building a hydrogen https://www.theclimatepledge.com/. Pledge to Net Zero. infrastructure in the United States. In: Kundu P, Dutta K. https://www.pledgetonetzero.org/. Climate Neutral Now. (eds.) Compendium of Hydrogen Energy, Volume 4: Hydrogen Use, https://unfccc.int/climate-action/climate-neutral-now/i-am- Safety and the Hydrogen Economy. Cambridge, UK: Woodhead a-company/organization/climate-neutral-now-signatories Publishing, 2016, 293–319. (23 December 2020, date last accessed). [20] Kurtz J, Sprik S, Bradley TH. Review of transportation hydrogen [3] Comello  S, Glenk  G, Reichelstein  S. Cost-efficient Transition infrastructure performance and reliability. International Journal to Clean Energy Transportation Services. 2020. https://madoc. of Hydrogen Energy, 2019, 44:12010–12023. bib.uni-mannheim.de/55376/ (23 December 2020, date last [21] Schoenung SM, Keller JO. Commercial potential for renewable accessed). hydrogen in California. International Journal of Hydrogen Energy, [4] Li M, Ye H, Liao X, et al. How Shenzhen, China pioneered the 2017, 42:13321–13328. widespread adoption of electric vehicles in a major city: im- [22] Wang  D, Muratori  M, Eichman  J, et  al. Quantifying the flexi- plications for global implementation. Wiley Interdisciplinary bility of hydrogen production systems to support large-scale Reviews: Energy and Environment, 2020, 9:e373. renewable energy integration. Journal of Power Sources, 2018, [5] Li Y, Zhan C, de Jong M, et al. Business innovation and govern- 399:383–391. ment regulation for the promotion of electric vehicle use: les- [23] US Department of Transportation. Federal Transit Administration. sons from Shenzhen, China. Journal of Cleaner Production, 2016, https://www.transit.dot.gov/. California Air Resources Board. 134:371–383. Innovative Clean Transit. https://ww2.arb.ca.gov/our-work/ [6] Kunith  A, Mendelevitch  R, Goehlich  D. Electrification of a programs/innovative-clean-transit (23 December 2020, date city bus network—an optimization model for cost-effective last accessed). placing of charging infrastructure and battery sizing of [24] Sovacool  BK, Kester  J, Noel  L, et  al. Actors, business models, fast-charging electric bus systems. International Journal of and innovation activity systems for vehicle-to-grid (V2G) tech- Sustainable Transportation, 2017, 11:707–720. nology: a comprehensive review. Renewable and Sustainable [7] Rogge M, van der Hurk E, Larsen A, et al. Electric bus fleet size Energy Reviews, 2020, 131:109963. and mix problem with optimization of charging infrastruc- [25] California Transit Authority. https://caltransit.org/about/ ture. Applied Energy, 2018, 211:282–295. transit-data/ (23 December 2020, date last accessed). [8] Gallet  M, Massier  T, Hamacher  T. Estimation of the en- [26] California State Legislature. Senate Bill 1. Transportation Funding. ergy demand of electric buses based on real-world data for 2018; Senate Bill 100. California Renewables Portfolio Standard large-scale public transport networks. Applied Energy, 2018, Program: Emissions of Greenhouse Gases. 2018. California Air 230:344–356. Resources Board, Innovation Clean Transit regulation. 14 [9] Bakker  S, Konings  R. The transition to zero-emission buses December 2018. in public transport–the need for institutional innovation. [27] State of California. Senate Bill-100 California Renewables Transportation Research Part D: Transport and Environment, 2018, Portfolio Standard Program: Emissions of Greenhouse Gases (2018); 64:204–215. Executive Order-B-48–2018; Executive Order-55–2018; Executive [10] Pelletier  S, Jabali  O, Mendoza  JE, et  al. The electric bus fleet Order-N-79–2020. transition problem. Transportation Research, Part C: Emerging [28] California Air Resources Board. 2020 Annual Evaluation of Fuel Technologies, 2019, 109:174–193. Cell Electric Vehicle Deployment & Hydrogen Fuel Station Network [11] Lin  Y, Zhang  K, Shen  ZJM, et  al. Charging network planning Development (Report Pursuant to AB-8; Perea, Chapter 401, for electric bus cities: a case study of Shenzhen, China. Statutes of 2013). 2020. https://ww2.arb.ca.gov/sites/default/ Sustainability, 2019, 11:4713. files/2020-09/ab8_report_2020.pdf (23 December 2020, date [12] Logan  KG, Nelson  JD, Hastings  A. Electric and hydrogen last accessed). buses: shifting from conventionally fueled cars in the UK. [29] Forrest K, Mac Kinnon MM, Tarroja B,et  al. Estimating the Transportation Research Part D: Transport and Environment, 2020, technical feasibility of fuel cell and battery electric vehicles 85:102350. for the medium and heavy duty sectors in California. A pplied [13] Southern California Edison. Take Charge: A  Guidebook to Energy, 2020, 276:115439. Fleet Electrification and Infrastructure. 2020. https://www.sce. Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 504 | Clean Energy, 2021, Vol. 5, No. 3 [30] Energy Futures Initiative and Stanford University. An Action [42] Mayer  T, Semmel  M, Morales  MAG, et  al. Techno-economic Plan for Carbon Capture and Storage in California: Opportunities, evaluation of hydrogen refueling stations with liquid or gas- Challenges, and Solutions. 2020. https://sccs.stanford. eous stored hydrogen. International Journal of Hydrogen Energy, edu/sites/g/files/sbiybj7741/f/efi-stanford-ca-ccs-sfpm- 2019, 44:25809–25833. rev1-10.25.20_0.pdf (23 December 2020, date last accessed). [43] Reed J, Dailey E, Shaffer B, et al. Roadmap for the Deployment and [31] California Fuel Cell Partnership. Fuel Cell Electric Buses: Enable Buildout of Renewable Hydrogen Production Plants in California. 100% Zero-emission Bus Procurement by 2029. 2019. https://cafcp. CEC. Publication Number: CEC-600-2020-002. 2020. https:// org/sites/default/files/2019-CaFCP-FCEB-Road-Map.pdf (23 cafcp.org/sites/default/files/Roadmap-for-Deployment-and- December 2020, date last accessed). Buildout-of-RH2-UCI-CEC-June-2020.pdf (27 July 2021, date [32] California ISO. http://www.caiso.com/todaysoutlook/pages/ last accessed). supply.html and http://www.caiso.com/informed/Pages/ [44] Air Liquide. Air Liquide committed to producing renewable ManagingOversupply.aspx (23 December 2020, date last hydrogen for the West Coast mobility market with new li- accessed). quid hydrogen plant. Press release. 8 October 2019. https:// [33] California Energy Commission (CEC). Final 2019 Integrated www.airliquide.com/united-states-america/air-liquide- Energy Policy Report. Report19-IEPR-01. pp. 70–113, 227–245. 2020. committed-producing-renewable-hydrogen-west-coast- https://www.energy.ca.gov/data-reports/reports/integrated- mobility-market (23 December 2020, date last accessed). energy-policy-r e port/2019-inte gr ated-energy-policy- [45] Cheng AJ, Tarroja B, Shaffer B, et al. Comparing the emissions report/2019-iepr (23 December 2020, date last accessed). benefits of centralized vs. decentralized electric vehicle smart [34] California Public Utilities Commission. Decision Approving charging approaches: a case study of the year 2030 California Retirement of Diablo Canyon Nuclear Power Plant. 11 January 2018. electric grid. J Power Sources, 2018, 401:175–185. https://docs.cpuc.ca.gov/PublishedDocs/Published/G000/M205/ [46] Hingtgen J, Le D, Davis B, et  al. Productivity of Wind Generation K090/205090240.PDF (23 December 2020, date last accessed). in California 2014 through 2016. California Energy Commission [35] Gerke B, Gallo G, Smith S, et al. The California Demand Response staff report CEC-200-2019-002. 2019. https://ww2.energy. Potential Study, Phase 3: Final Report on the Shift Resource through ca.gov/2019publications/CEC-200-2019-002/CEC-200-2019- 2030. 2020. https://escholarship.org/uc/item/7bx121k6 (23 002.pdf (23 December 2020, date last accessed). December 2020, date last accessed). [47] Wang  M, Ullrich  P, Millstein  D. The future of wind energy in [36] California Energy Commission and California Air Resources California: future projections with the Variable-Resolution Board. Joint Agency Report on Assembly Bill 8: 2019 Annual CESM. Renewable Energy, 2018, 127:242–257. Assessment of Time and Cost Needed to £attain 100 Hydrogen [48] Tarroja  B, Mueller  F, Eichman  JD, et  al. Spatial and temporal Refueling Stations in California. CEC-600-2019-039. 2019. https:// analysis of electric wind generation intermittency and dy- ww2.energy.ca.gov/2019publications/CEC-600-2019-039/CEC- namics. Renewable Energy, 2011, 36:3424–3432. 600-2019-039.pdf (23 December 2020, date last accessed). [49] California Air Resources Board. LCFS Electricity and Hydrogen [37] California Air Resources Board. Innov ative Clean Transit—Rollout Provisions. https://ww2.arb.ca.gov/resources/documents/lcfs- Plans. https://ww2.arb.ca.gov/our-work/programs/innovative- electricity-and-hydrogen-provisions. Appendix H: Low Carbon clean-transit/ict-rollout-plans (23 December 2020, date last Fuel Standard (LCFS) Program and Examples. 2020. https://ww3. accessed). arb.ca.gov/regact/2018/ict2018/apph.pdf (23 December 2020, [38] Reddi  K, Elgowainy  A, Sutherland  E. Hydrogen refueling date last accessed). station compression and storage optimization with tube- [50] California Air Resources Board. Weekly LCFS Credit Transfer trailer deliveries. International Journal of Hydrogen Energy, 2014, Activity Reports. 2021. https://ww3.arb.ca.gov/fuels/lcfs/ 39:19169–19181. credit/lrtweeklycreditreports.htm (6 January 2021, date last [39] Li XJ, Allen JD, Stager JA, et  al. Paths to low-cost hydrogen accessed). energy at a scale for transportation applications in the USA [51] United Nations. World Urbanization Prospects: The 2018 and China via liquid-hydrogen distribution networks. Clean Revision. File 12: Population of Urban Agglomerations with 300,000 Energy, 2020, 4:26–47. Inhabitants or More in 2018, by Country, 1950–2035 (Thousands) [40] Schmidt  O, Gambhir  A, Staffell  I, et  al. Future cost and per - POP/DB/WUP/Rev.2018/1/F12. https://population.un.org/wup/ formance of water electrolysis: an expert elicitation study. Download/ (23 December 2020, date last accessed) (29 January International Journal of Hydrogen Energy, 2017, 42:30470–30492. 2021, date last accessed). [41] Petitpas  G, Aceves  SM. Liquid hydrogen pump performance [52] Dong  D, Duan  H, Mao  R, et  al. Towards a low carbon transi- and durability testing through repeated cryogenic vessel tion of urban public transport in megacities: a case study of filling to 700 bar. International Journal of Hydrogen Energy, 2018, Shenzhen, China. Resources, Conservation & Recycling, 2018, 43:18403–18420. 134:149–155. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Clean Energy Oxford University Press

Zero-emission public transit could be a catalyst for decarbonization of the transportation and power sectors

Loading next page...
 
/lp/oxford-university-press/zero-emission-public-transit-could-be-a-catalyst-for-decarbonization-mdGvYB8cv9
Publisher
Oxford University Press
Copyright
Copyright © 2021 National Institute of Clean-and-Low-Carbon Energy
ISSN
2515-4230
eISSN
2515-396X
DOI
10.1093/ce/zkab029
Publisher site
See Article on Publisher Site

Abstract

Despite small overall share of vehicles... transit bus Heavy-heavy Heavy duty Light/med duty CO ... public transit infrastructure investment can support the wider transition to sustainable, carbon-neutral economies. Keywords: carbon neutrality; electric vehicles; fuel-cell vehicles; hydrogen; infrastructure; net-zero emissions; public transit; transition Received: 13 May 2021; Accepted: 6 August 2021 © The Author(s) 2021. Published by Oxford University Press on behalf of National Institute of Clean-and-Low-Carbon Energy This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (http:// creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 Ku et al. | 493 The primary players are individual transit agencies, who Introduction bear responsibility for decisions on bus types and the in- Numerous countries, local jurisdictions and private com- frastructure investments to operate their fleets, original panies have made carbon-neutrality pledges, with time equipment manufacturers who have to ramp up produc- horizons ranging from 2025 to 2060 [1, 2]. The trans- tion to meet accelerating demand for vehicles and fuel portation sector currently generates about a quarter of suppliers (e.g. electric utilities, industrial gas companies) the global anthropogenic CO; a transition away from who operate capital-intensive businesses and make com- petroleum-based fuels will be an important part of efforts mercial decisions based on market signals. Cost signals to reach net-zero emission targets [3]. Public transit fleets from fuel providers influence transit-agency decisions, account for a relatively small fraction of the total vehicle and vice versa. Early in the transition, decisions by pri- and energy requirements for fuelling, but they can be an mary players can be made in a relatively independent important lever for governments to increase public accept- manner. As the transition progresses, interactions across ance of new technology, support the maturation of vehicle the supply chain lead to uncertainty in the planning land- supply chains, promote the deployment of a large-scale scape, increased risk in capital investment, constraints resilient refuelling infrastructure and experiment with arising from lock-in effects from earlier decisions, con- different policy frameworks [4–12]. Moreover, ambitious flicting objectives among stakeholders and possible schedules for a zero-emission bus (ZEB) fleet conversion bottlenecks in the adequacy and reliability of energy de- can help to uncover emergent challenges related to energy livery [10, 17, 24]. However, positive interactions across the supply and delivery, as these activities interact with con- landscape may also emerge. current efforts to reduce emissions from the power sector. This Perspective article argues that there exist pockets Electrification has emerged as a leading option for of ‘predictability’ that can be used to reduce investment decarbonizing ground transportation. The leading com- risk and therefore be leveraged to help close the gap be- mercial options are battery electric vehicles (BEVs) and tween public and private financing. For the sake of this dis- hydrogen fuel-cell vehicles (HFCVs). In light-duty pas- cussion, we define predictability as an operational attribute senger automobile markets, BEVs currently outnumber that is conducive to regular and sustained capital recovery. HFCVs; in medium- and heavy-duty applications, mixed The economic value of stable revenue streams is already fleets exist and vehicle selection is based on the relative well appreciated; our contribution is to point out an under - advantages and limitations of the technology for different appreciated and significant opportunity to leverage public use cases. Public transit agencies can opt for battery elec- funds invested for specific and local infrastructure needs tric buses (BEBs) or hydrogen fuel-cell buses (HFCBs). BEB to stimulate broader investment across the energy and fleets are relatively easy to pilot on small scales and will transportation sectors during the transition period. directly benefit as electric grids decarbonize. As fleets grow, To illustrate the point, we draw on our collective ex- the overnight demand from their fixed charging schedules periences in working to decarbonize public transit over might exacerbate energy-supply challenges for grids with the past decade. Using public transit in California as a high intermittent renewables generation (e.g. solar, wind) case study, we present a thought experiment exploring [13–18]. Specifically, as growth in BEVs accelerates, the how investments in public transit might catalyse broader energy-supply effects from charging vast numbers of these transformation across the transportation and power sec- vehicles will result in significant changes to the traditional tors. California currently operates >10  000 buses across ‘duck curve’, with resulting consequences still unknown. >200 transit agencies, and has committed to fully transi- A  critical, related issue and concern is the resiliency as- tion to a ZEB fleet by 2040 [25, 26]. In addition, the state sociated with heavy reliance on grid power as affected by has announced goals of 250  000 BEV fast chargers and natural disasters, such as earthquakes, wildfires and hurri- 200 hydrogen-refuelling stations by 2025, 5  million total canes. HFCB fleets offer greater range, lower weight, faster zero-emission vehicles on the roads and a 60% share of refuelling times and higher resiliency, but the high capital renewable electricity generation by 2030, a ban on sales of cost of refuelling stations and nascent supply chains for internal-combustion passenger vehicles starting in 2035 zero-carbon hydrogen pose significant barriers during the and carbon neutrality by 2045 [27]. The scale of public early stages of adoption [19, 20]. The supply issue might be transit and the variety in transit agencies across the state solved by using electrolysis powered by renewable energy; makes California an especially informative example. Our a ‘power-to-gas’ (P2G) approach may help to manage inter- analysis takes advantage of the relative clarity in problem mittency on the electric grid and provide long-term energy definition (i.e. articulated policy commitments and tar - storage [21, 22]. gets, well-defined segment of the transportation sector Progress towards zero-emission public transit requires with predictable refuelling profiles), availability of rele- cooperation across a range of stakeholders. Governments vant public data (i.e. transit-agency operating profiles and provide the overall policy framework and motivate action transition roadmaps, electricity supply and demand) and through regulations and incentives. They also exercise complementary studies of transition pathways in other convening power to gather other stakeholders and pro- sectors (i.e. light and heavy vehicles, electricity, industry) mote the sharing of information and experience [23]. to examine this issue in detail [28–30]. Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 494 | Clean Energy, 2021, Vol. 5, No. 3 The paper is organized around two questions. First, pm to 4 am) ranged from 90 to 120 GWh/night. For context, what will it take to deliver energy in the right form at the the successful introduction of 5 million light-duty BEVs by right times? Second, how can the necessary investments 2030 would increase the daily demand by an additional in energy delivery for public transit fleets be used to sup- 34–70 GWh/day (viz. 5M LDV x 0.22–0.45 kWh/mi/LDV x 31 port broader decarbonization efforts? We begin with a mi/day). California’s current grid has the ability to satisfy statewide energy balance to quantify the energy-delivery the incremental demand from an all-BEB fleet in 2040, but requirements at different stages of the transition; we also ~75–90% of night-time generation comes from natural gas, examine energy demand from ZEB use profiles at the level nuclear and imports that are expected to be phased out of individual transit agencies. We then provide a brief intro- over the same time horizon during which ZEB demand duction, for non-expert readers, of the investment needs emerges [32–34]. Renewable portfolio standards requiring for the refuelling infrastructure of two leading options for 60% renewable generation by 2030, en route to a net-zero BEB and HFCB fleets. This leads to a discussion on how the grid by 2045, mean investment in both zero-carbon night- predictable and long-term nature of the refuelling profile time generation and energy storage might be needed to might be used to support the scale-up of the fuelling in- ensure adequate electricity supply [26]. frastructure in a manner that can benefit the wider efforts California’s energy infrastructure will also need to to decarbonize the transportation sector and electric grid. evolve to deliver sufficient zero-carbon H for electrified Two examples are given to illustrate how this might be ac- transportation. Over 95% of H production in the USA cur - complished in California. The first involves the expansion rently comes from steam methane reforming (SMR). In a of overnight electricity supply in support of BEBs and the net-zero future, H production will need to shift towards second relates to establishing medium-scale low-carbon alternatives such as biogas reforming, SMR with CO cap- H supply for HFCB fleets. Although we focus on the spe- ture and storage (CCS) or electrolysis using electricity cific case of public transit in California, the possibility of from zero-carbon sources [21, 30]. Given its large share mobilizing capital using large, predictable revenue streams in California’s generation mix, the use of solar power in could be applied more generally; we conclude with some a P2G capacity could become part of the long-term solu- observations on how the lessons from California might be tion [21]. Assuming a daily H demand of 16–30 kg/HFCB/ extended to other locations. day, statewide supply would need to reach 160–290 tons per day (tpd) to support a fleet of 10  000 FCBs. Curtailed solar energy in 2019 ranged from 0.5 (August) to 7.3 GWh/ day (April and May), with an average of ~2.5 GWh/day. 1 State-level energy supply and delivery Assuming 55 kWh/kg for electrolysis and compression, requirements this corresponds to an H potential of ~45 tpd, but monthly Fig. 1a shows the size and share of the public transit bus variations in solar power generation means production fleet relative to the entire vehicle population in California. will vary between 8 and 133 tpd [32]. Curtailed solar power Despite rapid projected growth in the absolute number of is forecasted to grow to ~15 GWh in 2030 and 100 GWh in ZEBs, they would only account for <1% of all vehicles after 2045 [35]. In this scenario, P2G-derived H could meet the full adoption in 2040. For context, success in deploying forecasted demand from HFCBs while also supporting up 5 million total zero-emission vehicles (ZEVs) by 2030 would to 1 million light-duty HFCVs by 2045. For context, 43 retail result in an ~15% vehicle share. In this study, we perform a H -refuelling stations had a total capacity to deliver 12 tpd thought experiment around the two limiting cases of fully at the end of 2019 [36]. BEB or HFCB fleets by 2040. Mixed fleets favouring BEBs are Regardless of fleet composition, California has the re- more likely, but this approach allows us to investigate a sources to generate adequate energy for both types of ZEB full range of possibilities with respect to bus technology fleets throughout the entire transition and beyond 2040. [31]. Despite the limited numbers of vehicles in absolute The timely mobilization of capital investment in delivery terms, the regimented service schedules of ZEBs create a infrastructure is more likely to be the limiting factor. significant and predictable energy demand. Fig. 1b shows Beyond the investment in fleets themselves, BEBs will re- the energy footprint for different vehicle types. The height, quire energy storage or additional night-time generation, width and area of each box correspond to the energy in- while HFCBs will need a supply chain capable of delivering tensity per mile (with the range indicating the average ~100 tons per day of clean H along with the construction of and maximum values reported), average daily mileage and refuelling stations. Investment in electricity-transmission total daily energy demand per vehicle [25, 29]. and H -distribution networks will also be needed. Fig. 1c overlays the electricity demand for overnight charging of BEBs and the solar-power-based electrolysis 2 Local energy-delivery requirements production of H in support of HFCBs relative to current generation profiles. The aggregate electrical demand The actual path to full ZEB adoption involves decisions by during overnight hours is projected to be 2.6–4.8 GWh/ individual transit agencies concerning bus technology and night by 2040 (viz. 10k BEB x 2–3.6 kWh/mi/BEB x 132 mi/ schedule; these determine the nature and timing for infra- day). In 2020, the average overnight power generation (10 structure projects. Some agencies have already conducted Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 Ku et al. | 495 A C Vehicle Population (CA, 2019) Future cumulative ZEB energy use vs 2019 generation profile transit bus 1.6M 10K 30M BED overnight charging (2019 total, 10 pm to 4 am) max avg min Light and medium duty Heavy duty Heavy-heavy Incremental energy 2.0 demand (10k BEB) Estimated ZEV energy use (per vehicle) (3.6) 2.0 1.0 12:00 AM 6:00 AM 12:00 PM 6:00 PM 12:00 AM 3.0 (2.5) (1.9) 0.55 2.0 0.22 (0.74) (0.45) 1.0 H from solar P2G production 0.0 (2019 solar, 8 am to 4 pm)* Daily 31 33 74 186 132 mileage LDV HDV HDV HDV HDV (mi/d ) MDV (L) (M) (H) (BEB) 9.1 Incremental max energy (22) 10 (21) 6.7 demand avg 4.2 (10k HFCB) (9.1) (6.7) 0.8 min (1.0) 31 33 74 186 132 Daily 12:00 AM 6:00 AM 12:00 PM 6:00 PM 12:00 AM LDV HDV HDV HDV HDV mileage * P2G = 55 kWh/kg MDV (L) (M) (H) (FCEB) (mi/d ) Fig. 1: Energy impacts of a zero-emission California public transit fleet. (a) Size and share of transit bus fleet relative to other vehicles. (b) Estimated ZEV energy use per vehicle, with the width of each box on the x-axis showing the average daily mileage, and the average and maximum reported energy efficiencies shown on the y-axis. Ranges for energy efficiency are from refs [25] and [29]. The area of each box corresponds to the daily energy use per vehicle (kWh/d or kg H /day). (c) Comparison of overnight electricity demand for BEBs and H using solar-based P2G (at 55 kWh/kg for elec- 2 2 trolysis and compression) to current generating profiles (average, maximum and minimum monthly averages). The x-axis shows the time of day for each 24-h period and the y-axis shows the power generated; the curves show the average, monthly maximum and minimum, generation at 5-min intervals. The unshaded region in the upper plot indicates the amount of power available to support real-time overnight BEB charging. The unshaded region in the lower plot corresponds to the solar generation that can be used to support P2G H production; the curves were calculated using 2019 historical power data published by CAISO [32]. Overlaid boxes show the relative magnitude of future energy demand versus existing generating cap- acity. The boxes in the plots corresponds to the total energy needed to support 10 000 BEBs (upper plot) or produce sufficient H to support 10 000 HFCBs (lower plot). See the online Supplementary Information for details. field evaluation of ZEBs, published strategic roadmaps and current fleet size and average route mileage of transit fleets; begun transitioning their fleets [37]; others are still in the the marker size indicates the total mileage travelled within planning stage. As of December 2020, 10 transit agencies each county and is proportional to the energy demand. On had filed formal plans and another five had announced this basis, Los Angeles and Alameda counties stand out roadmaps for their ZEB transitions. An important feature owing to their larger fleet size and longer average route dis- that has emerged is the role of route distance and fleet tance, respectively. The shaded inset in Fig. 2a is enlarged size in the decision between BEBs and HFCBs. In an ideal in Fig. 2b to highlight use profiles for the next eight largest world without fuel supply or capital investment barriers, county fleets. Together, the top 10 counties account for transit agencies with longer routes and larger fleets would >83% of the total miles travelled by public transit buses on tend towards HFCBs due to operating-cost advantages. The an average day (Fig. 2c)—indicative of the outsized impact exact threshold at which HFCBs become favourable varies of urban areas. The data also show use profiles in the other based on additional factors such as the frequency of stops, 39 counties with transit agencies that must also comply elevation changes, climate and vehicle speeds (see the on- with state targets. Distributions of average route distances line Supplementary Information) [37]. by county and bus population are shown in Fig. 2d and e. Fig. 2 compares use profiles across transit fleets at the Buses in ~80% of counties travel on average <100 mi/ county level. Fig. 2a and b show differences based on the day, but the skew associated with urban areas means that HFCV efficiency BEV efficiency (kg H /100 mi ) (kWh/mi ) Solar generation (GW) Generation (GW) Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 496 | Clean Energy, 2021, Vol. 5, No. 3 ~90% of the buses in California reach this average daily near larger fleets may be able to benefit from infrastruc- mileage. Counties with large fleets (>200 buses) may host ture investment and expertise from their neighbours. multiple transit agencies and the large fleet sizes allow the possibility of hybrid fleets with both BEBs and HFCBs. 3 Infrastructure investment for zero- Medium-sized fleets tend to support moderate population emission public transit centres or large service areas. These fleets are expected The transition to ZEB fleets will require capital investment to be a single bus-technology type due to the desire for across the entire energy supply chain. Fig. 3 summarizes transit agencies to simplify the fuelling logistics and main- the types of investment needed and the relevant actors tenance operations. Small fleets account for only 10% of for each area: transit authorities for bus- and refuelling- the total bus population, but operate in 26 counties; two- station decisions; and utilities and energy providers for thirds of these counties operate fleets with <50 buses. Most energy-delivery decisions. The top panel shows the supply have limited resources to support transition planning and chain for BEBs, where electricity from selected zero- fleet-conversion decisions for this group are likely to be carbon sources is delivered via bus-recharging stations. An constrained by logistics considerations. However, counties Total mileage CDF by county A C Daily mileage per bus vs fleet size 100% 75% 120 50% AC 25% OC 0% SD 0% 20%40% 60%80% 100% LA Fraction of counties Sac SF D Mileage CDF by county 0 1000 2000 3000 4000 5000 Fleet size (buses) Daily mileage per bus vs fleet size 0% 20%40% 60%80% 100% Fraction of counties SBr AC Mileage CDF by bus SCI SD OC SM Sac SF 40 AC LA OC SD 0 300 600 900 1200 SF Fleet size (buses) 0 0% 20%40% 60%80% 100% Fraction of buses Fig. 2: County-level segmentation of bus fleets and route profiles. (a) Daily mileage vs fleet size for 43 counties with transit agencies operating buses. For labelled counties, the areas of the circles are proportional to the total miles driven. (b) Expanded view of daily mileage vs fleet size, using the same shading scheme as in the first panel. (c) Cumulative distribution function (CDF) for share of total miles/day, segmented by county. The x-axis indicates the fraction of counties and the y-axis shows the cumulative share of daily miles driven across the entire state. The top 20% of counties account for ~75% of miles driven. (d) CDF for average daily mileage, by county. (e) CDF for average daily mileage versus number of buses across the entire state. Five counties are shown for illustration: San Francisco (SF), Orange (OC), San Diego (SD), Los Angeles (LA) and Alameda (AC). Data from ref. [25]. Average daily mileage (mi/day/bus) Average daily mileage (mi/day/bus) Cumulative share of Average daily mileage Average daily mileage total miles/day (%) (miles/day/bus) (miles/day/bus) Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 Ku et al. | 497 important feature in station costs is the non-linear invest- Among the options for central production, SMR facilities ment profile as the fleet size grows; a new transmission with CCS will take time to develop, leaving biogas-derived infrastructure (e.g. substations) may be required to deliver H as the leading option in the near term. Air Liquide is in adequate power for the fast charging of large fleets. The the process of commissioning a 30-tpd biogas-to-liquid H bottom panel shows the supply chain for HFCBs. There are facility in Las Vegas, NV [44]; by 2030, aggregate demand two primary paths for station design, based on whether from ZEVs in California could consume all of the supply the fuel is stored as a compressed gas or a cryogenic liquid, from multiple facilities of this scale. Depending on project and three representative options for production and distri- details and distance from transit-agency depots, electri- bution [38, 39]. city production for a BEB division (50 MWh/night) would Bus fleets are the largest capital expense: ~$100 million is require $5–20 million and zero-carbon H supply (2–3 tpd) needed for a single division, increasing proportionally with could range from $6 to $15 million. These values would in- larger numbers of divisions. Some of this required capital will crease by an order of magnitude for 10 divisions and an- have already been committed for the routine replacement other order of magnitude to support the entire ZEB fleet of existing assets. Since ZEBs cost ~$200–500  k more than in 2040. diesel buses, the incremental cost over the existing capital commitments would be ~$20–30  million for a division be- yond what would be spent under business-as-usual replace- 4 Public transit demand as a catalyst for ment schedules. ZEB fleets will also require the construction investment of BEB-charging or HFCB-refuelling stations. Capital invest- The central hypothesis of this study is that investments ment for stations varies depending on technology; we es- made to support energy delivery for public transit can timate a depot-based fast-charging facility for a full BEB also aid efforts to decarbonize the electricity grid and division could require $5–14 million, while an HFCB station broader transportation sector. The predictable and could be less expensive, ranging from $2 to $6  million [14, long-term energy demand from ZEB fleets is a valu- 39]. During a phased adoption in which buses are added over able attribute that can be used to facilitate long-term multiple years, HFCB fleets are more expensive to operate planning for grid evolution and also mobilize capital initially because stations must be constructed at their full- for infrastructure development. Increasing variability service capacity; however, they become more affordable as from growth in renewable-energy generation and expan- the number of buses in the fleet increases [31]. California has sion from electrified transportation will tend to amplify recognized this barrier and offers policy support to help fund the reliability challenges for the electricity grid. Whilst the construction of H stations for early adopters. Over time, time-of-use demand-response programmes can help to market mechanisms will need to provide the additional $20– spread charging loads throughout the day for passenger 100 million needed for 10 divisions and $200 million–$1 bil- vehicles or commercial fleets, there may be limits to lion needed for full conversion by 2040 [28]. their ability to fully rebalance energy requirements for Capital projects will also be needed to upgrade night-time charging [18, 24, 45]. With regard to capital California’s electricity-transmission and H -delivery net- mobilization, the predictability of demand from BEB works. Expansion of night-time electricity capacity to sup- fleets could allow electric utilities to negotiate commer - port BEB fleets could include combinations of generation, cial supply contracts (e.g. power purchase agreements) storage and transmission, as shown in Fig. 3. For electric designed around stable overnight-charging schedules. charging, we focus only on the depot option; interested Predictability in H-refuelling demand could be leveraged readers can consult an extensive literature on alternate in a similar way. In both cases, long-term contracts offer charging options [6, 11, 13–17]. H supply chains can be con- secured revenue streams that can be used for capital re- figured for local production at refuelling stations or central pro- covery in the financing of infrastructure projects, as il- duction with distribution. The first option occurs entirely at lustrated schematically in Fig. 4T . able 1 shows estimates the station and involves electrolysis, small-scale SMR, gas of how much capital might be mobilized by such supply compression, high-pressure ground storage and, possibly, agreements; assuming simple amortization schedules, refrigeration. Electrolysis is currently more expensive than a single 100-bus division could support in the order of alternatives; cost-reduction roadmaps have been devel- $10–50  million in capital investment. Applied to 10  000 oped, but it could take a decade to fully realize the savings ZEBs at 100 bus divisions across California, this has the [40]. For production scales and transport distances relevant potential to mobilize $1–5 billion for infrastructure pro- to California, supply chains using centralized production jects; on a national basis across the USA, the sums could with liquid-H (LH ) distribution is increasingly favoured by 2 2 approach an order of magnitude larger. industry and offers competitive economics. The costs from liquefaction are more than offset by savings from the re- duced cost of distribution and simplification of station de- 4.1 Example 1: overnight charging for BEB fleets sign (i.e. liquid pumping in lieu of gas compression, direct filling to reduce ground storage, heat integration with li- Historical patterns in California wind generation show quid vaporization to reduce refrigeration) [39, 41–43]. peak power in the evening through to the early morning Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 498 | Clean Energy, 2021, Vol. 5, No. 3 Transit agencies Energy providers Bus fleet Recharging station Electricity transmission Electricity production Solar: $2000/kW Storage: $250/kWh (NREL, 2020) $4-10M $25-60M/100 km (150 or 350 kW fast charger) NGCC + CCS $1-10M (MISO, 2019) $2400/kW 100 bus division (upgrade) (650 MW) HFCB Capital cost factors: CO (NETL, 2016) 2020: $0.8-1.2M/bus $5-30M (new) - rated voltage 2030: $0.5-0.6M/bus (MISO, 2019) - structures (Transit agencies, 2020; Wind: $1400/kW - land/rights costs BEB DOE, 2020; BNEF, 2018) Storage: $250/kWh (NREL, 2020) Fleet size (buses) Bus fleet H deliveryH supply Refueling station 2 2 Gaseous H 2 $0.9-3.4M/tpd (0.4 to 50 tpd) (see SI) $4-8M (3 tpd station) (ANL HDRSAM, 2020) $60-80M/100 km $0.5-0.6M/tpd (US DOE EERE, 2018) CH 4 (380 tpd) 100 bus division (NREL, 2018) CO 2020: $0.9-1.4M/bus Liquid H $1-3M/tpd (10-30 tpd ) 2030: $0.6M/bus $1.7-2.5M/tpd (50 tpd) $1.1M (4-5 tH trailer) (Transit agencies, 2020; $2-5M (3 tpd station) $3-11M/tpd (7.5 tpd) (US DOE-EERE, 2016) DOE, 2019; 2012) (ANL HDRSAM, 2020) (CEC, 2020) CO BEB or HFCB Electrical Transmission Solar + storage NGCC + CCS Wind + storage Fast charging station 100 bus division Substation lines CH CO Gas compressor Liquid pump Liquefier Pipeline Biogas Electrolyzer + cascade storage + cascade storage + tanker SMR + CCS + reformer + refrigeration + refrigeration Fig. 3: Estimated ranges for capital investment in BEB and HFCB supply chains. Investment options for transit agencies and energy providers in BEB and HFCB supply chains. The options illustrated here are indicative and not exhaustive. See the online Supplementary Information for details. hours [46]. Wind generation in 2019 was ~14 TWh from an require ~1.6 GWh/night. (A detailed discussion of aggre- installed base of ~6 GW, with generation during the winter gated wind generation and BEB demand is included as on- months about half that of summer production. For con- line Supplementary Information.) While the overbuilding text, the California Energy Commission (CEC) expects that of wind to ensure adequate supply year-round is not in- 3 GW of wind capacity will be added between 2020 and consistent with CEC forecasts, a more robust approach 2030. This begs the question of whether this resource can is to invest in energy storage under the assumption that help meet the overnight-charging demand, initially for the daily energy generation is adequate to meet transit public transit and in the longer term for electrified trans- needs. Fig. 5b shows the cumulative distribution function portation generally. for the estimated incremental overnight wind generation Fig. 5a shows the seasonal variation in hourly gen- in 2030, assuming a profile similar to historical generation eration at the four most productive wind power sites in (2019). Assuming an hourly generation profile consistent California, which accounted for 90% of the wind gener - with 2019, <1.6 GWh/night of added generation would be ation of the state in 2019 [47, 48]. A  fully electrified ZEB expected during 17% of the year (63 nights). Fig. 5c and d fleet in the Bay Area and greater Los Angeles would shows the effect of storage on overnight reliability and Legend HFCB BCB Capital cost Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 Ku et al. | 499 Transit agency Total cost of operation (transit agency, $/mi) TA operating revenues Incentives Operating Transit agency Fuel cost costs capital recovery Predictable demand allows long-term contracts Fuel price Capital for financing energy supply Operations Supplier infrastructure (incl profits) capital recovery Capital markets Energy supplier Fig. 4: Use of predictable fuelling schedules to support capital mobilization for energy infrastructure. From the transit-agency perspective, the total cost of operation (top bar) is the sum of the operating revenues and government incentives (second bar) as well as the sum of the operating costs (third bar). Operating costs include contributions from fuel, other transit-agency operating costs and capital-recovery costs for transit-agency in- vestments. Fuel purchased from external providers (fourth bar), in turn, supports the operating costs for energy suppliers and capital recovery for infrastructure projects (fifth bar). Predictable demand allows the negotiations of long-term contracts that can be used by energy providers to mo- bilize larger sums for capital financing. Table 1: Estimates of capital that could be supported by energy contracts Financing terms 8% 15% 8% 15% (rate of return [%], period [yrs]) 10 yrs 10 yrs 20 yrs 20 yrs BEB 100-bus division $15 million $11 million $22 million $14 million Short routes: 200 MWh/night Energy sales value: $30/MWh 100-bus division $37 million $27 million $54 million $34 million Long routes: 500 MWh/night Energy sales value: $30/MWh HFCB 100-bus division $15 million $11 million $22 million $14 million 3 tpd; energy value: $2/kg the investment required for different levels of reliability This simplified analysis uses 2019 data. As such, it does not (number of days in a year in which wind generation can account for annual variations in wind generation. Moreover, cover the incremental charging demand) from different the productivity of added wind capacity may be lower, as the levels of energy storage. Fig. 5c shows the effect of energy most productive sites are developed first. These effects are ex- storage on reliability. This use profile indicates that energy pected to impact the exact number of days on which gener - storage in support of BEB demand would be needed for ation is inadequate, but the general range is expected to be ~17% of the year, primarily during the winter months and consistent with the results obtained from our initial estimate. occasionally on days with below-average wind generation. Further work will be needed to identify specific projects and During the remaining days, the facilities would be avail- fully quantify their benefits for renewable power integration. able as a general capability to support grid stability. Fig. 5d shows the value of investment in energy storage on re- 4.2 Example 2: Medium-scale zero-carbon H liability. Revenue from bus energy-supply contracts could supply for fuel-cell vehicle fleets support investment at a level of $90–170 million (cf. T able 1, A key difference between BEB and HFCB supply chains is the scaled to 1.6 GWh/night); spending this on energy storage timing of infrastructure deployment. Whereas BEB-charging could increase reliability to ~93%. Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 500 | Clean Energy, 2021, Vol. 5, No. 3 AB Variability in wind generation Incremental wind generation by 2030 (2019 wind, 10 pm to 4 am) due to forecasted growth Q3 Q2 Avg Q1 Q4 0 0 0 20 40 60 80 12:00 AM 6:00 AM 12:00 PM 6:00 PM 12:00 AM Ranked day C D Energy storage capacity needed Impact of energy storage to satisfy overnight demand investment on reliability 100% 95% 300 90% 85% 80% 0 400 800 1200 1600 80% 85% 90% 95% 100% Installed ES capacity (MWh) Reliability (% days/yr demand met) Fig. 5: Prospects for meeting overnight BEB demand using wind and energy storage. (a) Quarterly variability in daily wind-generation profiles in California, plotted at 5-min resolution. (b) CDF for incremental wind generation forecasted by 2030. The x-axis shows days, in ascending order. The y-axis shows the incremental generation potentially available from a 3-GW increase in wind capacity. The dotted line indicates the threshold needed to meet the 1.6-GWh/night incremental demand from all-BEB fleets in the Bay Area and Southern California; a gap exists for 63 nights of the year. (c) Impact of increasing energy-storage capacity (x-axis) on reliability (y-axis), as indicated by the number of days on which the overnight-charging demand would be satisfied. (d) Investment (y-axis) needed to deploy sufficient energy storage to satisfy different reliability thresholds (x-axis), at an assumed energy-storage cost of $250/kWh. The shaded band indicates the level of investment ($90–170 million) potentially available from transit- agency-backed contracts. facilities can be built in a stepwise manner, H supply chains Contributions from the Low Carbon Fuel Standard (LCFS) require significant upfront investment. Since full conversion are considered for each case. For the on-site electrolysis to ZEB fleets by 2040 requires 100% of bus purchases and de- case shown in Fig. 6b, electricity from the grid still includes liveries by 2028, a medium-scale H supply chain is necessary a significant contribution from fossil energy; we assume an to preserve the option of HFCB adoption in the long term. LCFS credit of ~$2/kg assuming a basis of 33% renewable H Fig. 6a compares two options for clean H supply in the and $199/tCO (2020 average) [49, 50]. Assuming financing 2 2 intermediate term. The first involves on-site electrolysis to scenarios from Table 1, ~$3/kg of the revenue would be meet the needs of individual stations. While this is pos- needed to support capital recovery for the refuelling station sible for small-scale pilots involving <20 buses, full-scale and the electrolyser. In this case, the transit agency is re- deployment for a 1000-bus division will be challenging due sponsible for the infrastructure and is the main beneficiary. to footprint and electricity-delivery constraints in many The biogas-H with LH distribution case uses a dif- 2 2 urban areas. The second option uses biogas-to-H with LH ferent arrangement. Here, H is produced at a central 2 2 2 distribution. This path requires central production facil- facility, liquefied and delivered to the station. Fig. 6c dif- ities of ≥9 tpd (enough to supply three bus divisions) for ferentiates between the revenues available to the transit economics, but these facilities can also support broader agency and the H supplier. As with on-site electrolysis, demand from the transportation sector. The Air Liquide the transit agency allocates revenue and incentives to Las Vegas project suggests that this path is being taken operating costs and capital recovery; one difference is that seriously by industry [44]. the transit agency buys fuel from the H supplier, which Fig. 6b and c illustrates how capital recovery can be sup- then allocates sales revenue to its own operations and ported by the value streams from refuelling operations. capital recovery for the production and liquefaction in- The top bar in each case shows the revenue accrued by the frastructure costs. A second difference is the higher LCFS transit agency from operations and government incentives. credit; biogas-derived H is 100% renewable and qualifies Reliability (% of days/yr demand met) Wind generation (GW) Energy storage Incremental wind Investment ($M) generation by 2030 (GWh/night) Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 Ku et al. | 501 AC Medium-scale zero-carbon H supply 2 Option 2. Biogas-H + Liquid delivery TA Operating revenues LCFS credit ($4/kg) Option 1. Option 2. On-site electrolysis Biogas-H + Liquid delivery 2 Station Fuel cost costs H supply primarily Expandable supply supports single TA 10 tpd central facility For a 100-bus division Financing potential 3 tpd onsite facility (support 3-4 divisions) (3 tpd station), (serves 1 division) TA covers station costs $1/kg (allocated) TA pays for station H supplier(s) pay for and electrolyzer biogas-H production LH refueling 10-20 yrs; 8 to 15% and liquifier/tankers station $3-6 M $6-11 M B Option 1. On-site electrolysis Long-term fuel contract LCFS credit TA Operating revenues ($2/kg) Fuel price Operations Supplier Station Electrolyzer + station operating costs capital recovery (incl profits) capital recovery Supply for 3 bus divisions Financing potential Financing potential For a 100-bus division (3 tpd station), (prorated 3 tpd from 9 tpd facility) $1/kg + $2/kg $1/kg + $4/kg (allocated) (LCFS) (allocated) (LCFS) Onsite cgH Refueling 10 yrs ; 8 to 15% 10-20 yrs ; 8 to 15% station electrolyzers Liquefier + tanker Biogas-H $4-8 M+ $3-10 M $17-33 M $4-10 M+ $9-33 M $28-54 M Fig. 6: Financing pathways for medium-scale zero-carbon H supply chains. (a) Overview of two options for supplying clean H . (b) Financing poten- 2 2 tial for on-site electrolysis. The top bar represents the revenue from transit-agency operations and government (LCFS) credits; the second bar shows allocations among station-operating costs and capital recovery. The capital-recovery potential is calculated using scenarios frT om able 1 and the panel shows that ~$3/kg allocated to capital recovery could cover the capital costs associated with the approach. (c) Financing potential for biogas-H with liquid delivery. The top two bars indicate the revenue and cost allocations from the perspective of the transit agency. In the near term, the higher renewable content of biogas-derived H corresponds to a larger LCFS credit. A fraction of the total value is used by the transit agency to sup- port capital recovery for the refuelling station. The lower two bars show the revenue available to H suppliers. The revenue available is simply the fuel price, and it is allocated to operations (including profits) and capital recovery among the suppliers. The fuel price is a negotiated value between the transit agency and energy supplier. Predictability can facilitate long-term arrangements, which can mobilize larger sums of capital; the larger LCFS credit is accrued by the transit agency (as the user of the fuel) but affords flexibility in setting the price. This figure shows a situation in which the value from the LCFS is effectively passed along to H suppliers to help offset their capital requirements. for a credit of $4/kg (2020 average), and this value can be competitive alternatives as cost and regulatory hurdles are effectively allocated between the transit agency and H addressed. As with the BEB example, predictable demand, supplier through supply contract negotiations. Although sustained over decades, from transit agencies is a key en- the value of LCFS credits is expected to decline over time, abler for investments in infrastructure that can simultan- their current value provides an avenue for intermediate- eously meet immediate needs and act as a bridge to larger, term development. Predictability of the fuelling profile is system-wide changes. important in that it allows the longer-term agreements that can extend time horizons and access to the longer 5 Broader implications and conclusion capital-recovery scenarios. The numbers in Fig. 6c reflect an allocation of revenues that can support near-term de- Although we have focused on the situation in California, velopment of this type of supply chain. In contrast to the the predictability of energy demand from large ZEV fleets distributed-electrolysis approach, this option is compat- can be a useful lever for promoting transitions to carbon ible with future growth of the H  supply chain. Once an ini- neutrality in other locales. There are two complementary tial supply chain has been established, market forces can principles: the leveraging of large, predictable, long-term augment it in response to the growing HFCV population aggregate demand from vehicle fleets to support capital and improving technology. Over time, electrolysis with recovery in the financing of infrastructure projects; and on-site storage for P2G and SMR with CCS may become the alignment of projects to support broader system-level Transit agency value map H suppliers(s) Transit agency value map Value map Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 502 | Clean Energy, 2021, Vol. 5, No. 3 evolution. This approach could be particularly relevant for Third, government support can take different forms. urban areas in North America, Europe and East Asia where California is aggressively pursuing carbon neutrality across commitments to carbon neutrality intersect with suffi- multiple sectors using a combination of mechanisms. ciently large public transit systems or corporate fleets to Within this context, some counties have taken concrete support meaningful numbers of projects. We note four nu- steps to pilot ZEBs and develop detailed transition plans ances that should be considered in applying these prin- and roll-out schedules coordinated across multiple public ciples to other situations. agencies (see the online Supplementary Information for First, scale matters. Transportation and energy infra- details); others are still in the early stages of defining their structures typically involve large projects; the associated specific roadmaps. An even wider range of postures exists capital investments are generally tens of millions of dol- across the jurisdictions that can apply the principles iden- lars or more (cf. Fig. 3). Large fleets will be needed to gen- tified in this study. For example, Shenzhen, China, com- erate sufficient revenue to support capital recovery for this pleted the transition of its entire bus fleet of >16  000 magnitude of investment. In the California example, we vehicles to electric buses in 2018 [4]. This was accom- focused on bus divisions with 100 ZEBs. This is a natural plished through significant government support across the scale relevant to transit-agency operation that also appears entire supply chain, leverage of the unique regulatory en- capable of generating appropriate levels of revenue; many vironment in China [5, 52]. At the other extreme, there are larger projects will require cooperation between multiple jurisdictions that have announced commitments but are bus divisions. In 2018, the UN identified 42 metropolises still in the early stages of defining their path forward. For with populations of >5 million (along with 264 areas with municipalities in this latter group, calibrating across the populations of >1  million) across North America, Europe examples from California and other leading locales could and East Asia [51]. The size of the public bus fleets in these provide guidance for their planning activities. cities is more than sufficient to participate in the financing Finally, commercial fleets might also be able to con- model suggested here, and most of these cities have made tribute. Revenue streams from commercial fleets would direct commitments to climate action or are covered by be backed by business operations, so the financing terms national-level pledges. and time horizons would depend on the details of each Second, predictability can help in multiple ways. In the situation and could be more aggressive than for public California example, predictability offered benefits on two transit. Assuming financing terms and vehicle-use profiles levels. On one level, well-defined energy-demand profiles similar to the California situation (cf. Fig. 1b and Table 1), in the form of stable, daily charging schedules for BEBs or fleet sizes in the order of 3400–3900 light-duty (passenger known growth rates in cumulative H demand over time cars), 340–360 medium/heavy-duty trucks and 70–100 can be used by energy providers to plan for operations and Class 8 heavy-duty trucks would have similar energy de- growth. Coordination across the supply chain is a natural mand and revenue potential. In reality, commercial fleets consequence due to the mutual benefits. This can be as operate under a wide range of conditions, so these thresh- simple as sharing information between transit agencies, olds are indicative; separate analysis will be needed for energy suppliers and government about strategic inten- each situation. tions in a market-oriented framework or involve active Reaching carbon neutrality will require signifi- participation by government through public incentives cant changes across multiple sectors of the economy. for private capital decisions. Specific projects can then be Interactions, within and across sectors, can create feed- selected to preserve or even increase strategic flexibility back loops that introduce complexity and make the tran- across the energy system. On a second level, the long-term sition more difficult. However, these interactions are not nature of public transit demand unlocks the ability to fi- always negative and this Perspective article has attempted nance capital over extended periods. Government agen- to show situations in which interactions can be used to aid cies play a critical role in establishing markets and have a the transition to a more sustainable future. vested interest in the creation of mechanisms that engage private capital markets. The public sector plays a key role in setting transformational targets and providing the seed Supplementary data funding to nucleate action; mechanisms that can de-risk Supplementary data is available at Clean Energy online. investments, such as the one presented in this paper, can help to mobilize private capital to supplement public funds. While transit agencies are subject to the budgetary prior - Data availability ities of government and their capital expenses are often Data and analysis for figures and tables are included as online supported by subsidies (e.g. federal grants, sales taxes, Supplementary Information. Further information and requests for parcel taxes, and road and bridge tolls), public transit is resources and reagents should be directed to and will be fulfilled by seen as a public good that provides stability to its revenue the Lead Contact, Anthony Ku (anthonyku@nicenergy.com). Data streams. Although the specific details related to these two on vehicle populations were obtained from the California Transit facets will vary, the general principles should translate Authority and verified at individual transit-agency websites. Data across jurisdictions around the world. on energy production, including capacity and generation, were Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 Ku et al. | 503 obtained from the California Energy Commission and California com/sites/default/files/2020-07/Electrification%20%26%20 ISO (CAISO) websites. Other data were obtained from references Infrastructure%20Guidebook-Final_06.29.20.pdf (23 December as cited in the text. 2020, date last accessed). [14] Chen  Z, Yin  Y, Song  Z. A cost-competitiveness analysis of charging infrastructure for electric bus operations. Transportation Research, Part C: Emerging Technologies, 2018, Conflict of interest statement 93:351–366. [15] Lajunen A. Lifecycle costs and charging requirements of elec- None declared. tric buses with different charging methods. Journal of Cleaner Production, 2018, 172:56–67. References [16] He  Y, Song  Z, Liu  Z. Fast-charging station deployment for battery electric bus systems considering electricity demand [1] National, Regional, and Local Government Climate Pledges. charges. Sustainable Cities and Society, 2019, 48:101530. Paris Agreement Nationally Defined Contributions. https://unfccc. [17] An  K. Battery electric bus infrastructure planning under de- int/process-and-meetings/the-paris-agreement/nationally- mand uncertainty. Transportation Research, Part C: Emerging determined-contributions-ndcs/nationally-determined- Technologies, 2020, 111:572–587. contributions-ndcsClimate . Ambition Alliance: Net Zero by 2050. [18] van Triel F, Lipman TE. Modeling the future California electri- https://unfccc.int/climate-action/race-to-zero-campaign (23 city grid and renewable energy integration with electric ve- December 2020, date last accessed). hicles. Energies, 2020, 13:5277. [2] Organizational Climate Pledges. Net Zero Carbon by 2040. [19] Reddi  K, Mintz  M, Elgowainy  A, et  al. Building a hydrogen https://www.theclimatepledge.com/. Pledge to Net Zero. infrastructure in the United States. In: Kundu P, Dutta K. https://www.pledgetonetzero.org/. Climate Neutral Now. (eds.) Compendium of Hydrogen Energy, Volume 4: Hydrogen Use, https://unfccc.int/climate-action/climate-neutral-now/i-am- Safety and the Hydrogen Economy. Cambridge, UK: Woodhead a-company/organization/climate-neutral-now-signatories Publishing, 2016, 293–319. (23 December 2020, date last accessed). [20] Kurtz J, Sprik S, Bradley TH. Review of transportation hydrogen [3] Comello  S, Glenk  G, Reichelstein  S. Cost-efficient Transition infrastructure performance and reliability. International Journal to Clean Energy Transportation Services. 2020. https://madoc. of Hydrogen Energy, 2019, 44:12010–12023. bib.uni-mannheim.de/55376/ (23 December 2020, date last [21] Schoenung SM, Keller JO. Commercial potential for renewable accessed). hydrogen in California. International Journal of Hydrogen Energy, [4] Li M, Ye H, Liao X, et al. How Shenzhen, China pioneered the 2017, 42:13321–13328. widespread adoption of electric vehicles in a major city: im- [22] Wang  D, Muratori  M, Eichman  J, et  al. Quantifying the flexi- plications for global implementation. Wiley Interdisciplinary bility of hydrogen production systems to support large-scale Reviews: Energy and Environment, 2020, 9:e373. renewable energy integration. Journal of Power Sources, 2018, [5] Li Y, Zhan C, de Jong M, et al. Business innovation and govern- 399:383–391. ment regulation for the promotion of electric vehicle use: les- [23] US Department of Transportation. Federal Transit Administration. sons from Shenzhen, China. Journal of Cleaner Production, 2016, https://www.transit.dot.gov/. California Air Resources Board. 134:371–383. Innovative Clean Transit. https://ww2.arb.ca.gov/our-work/ [6] Kunith  A, Mendelevitch  R, Goehlich  D. Electrification of a programs/innovative-clean-transit (23 December 2020, date city bus network—an optimization model for cost-effective last accessed). placing of charging infrastructure and battery sizing of [24] Sovacool  BK, Kester  J, Noel  L, et  al. Actors, business models, fast-charging electric bus systems. International Journal of and innovation activity systems for vehicle-to-grid (V2G) tech- Sustainable Transportation, 2017, 11:707–720. nology: a comprehensive review. Renewable and Sustainable [7] Rogge M, van der Hurk E, Larsen A, et al. Electric bus fleet size Energy Reviews, 2020, 131:109963. and mix problem with optimization of charging infrastruc- [25] California Transit Authority. https://caltransit.org/about/ ture. Applied Energy, 2018, 211:282–295. transit-data/ (23 December 2020, date last accessed). [8] Gallet  M, Massier  T, Hamacher  T. Estimation of the en- [26] California State Legislature. Senate Bill 1. Transportation Funding. ergy demand of electric buses based on real-world data for 2018; Senate Bill 100. California Renewables Portfolio Standard large-scale public transport networks. Applied Energy, 2018, Program: Emissions of Greenhouse Gases. 2018. California Air 230:344–356. Resources Board, Innovation Clean Transit regulation. 14 [9] Bakker  S, Konings  R. The transition to zero-emission buses December 2018. in public transport–the need for institutional innovation. [27] State of California. Senate Bill-100 California Renewables Transportation Research Part D: Transport and Environment, 2018, Portfolio Standard Program: Emissions of Greenhouse Gases (2018); 64:204–215. Executive Order-B-48–2018; Executive Order-55–2018; Executive [10] Pelletier  S, Jabali  O, Mendoza  JE, et  al. The electric bus fleet Order-N-79–2020. transition problem. Transportation Research, Part C: Emerging [28] California Air Resources Board. 2020 Annual Evaluation of Fuel Technologies, 2019, 109:174–193. Cell Electric Vehicle Deployment & Hydrogen Fuel Station Network [11] Lin  Y, Zhang  K, Shen  ZJM, et  al. Charging network planning Development (Report Pursuant to AB-8; Perea, Chapter 401, for electric bus cities: a case study of Shenzhen, China. Statutes of 2013). 2020. https://ww2.arb.ca.gov/sites/default/ Sustainability, 2019, 11:4713. files/2020-09/ab8_report_2020.pdf (23 December 2020, date [12] Logan  KG, Nelson  JD, Hastings  A. Electric and hydrogen last accessed). buses: shifting from conventionally fueled cars in the UK. [29] Forrest K, Mac Kinnon MM, Tarroja B,et  al. Estimating the Transportation Research Part D: Transport and Environment, 2020, technical feasibility of fuel cell and battery electric vehicles 85:102350. for the medium and heavy duty sectors in California. A pplied [13] Southern California Edison. Take Charge: A  Guidebook to Energy, 2020, 276:115439. Fleet Electrification and Infrastructure. 2020. https://www.sce. Downloaded from https://academic.oup.com/ce/article/5/3/492/6375879 by DeepDyve user on 28 September 2021 504 | Clean Energy, 2021, Vol. 5, No. 3 [30] Energy Futures Initiative and Stanford University. An Action [42] Mayer  T, Semmel  M, Morales  MAG, et  al. Techno-economic Plan for Carbon Capture and Storage in California: Opportunities, evaluation of hydrogen refueling stations with liquid or gas- Challenges, and Solutions. 2020. https://sccs.stanford. eous stored hydrogen. International Journal of Hydrogen Energy, edu/sites/g/files/sbiybj7741/f/efi-stanford-ca-ccs-sfpm- 2019, 44:25809–25833. rev1-10.25.20_0.pdf (23 December 2020, date last accessed). [43] Reed J, Dailey E, Shaffer B, et al. Roadmap for the Deployment and [31] California Fuel Cell Partnership. Fuel Cell Electric Buses: Enable Buildout of Renewable Hydrogen Production Plants in California. 100% Zero-emission Bus Procurement by 2029. 2019. https://cafcp. CEC. Publication Number: CEC-600-2020-002. 2020. https:// org/sites/default/files/2019-CaFCP-FCEB-Road-Map.pdf (23 cafcp.org/sites/default/files/Roadmap-for-Deployment-and- December 2020, date last accessed). Buildout-of-RH2-UCI-CEC-June-2020.pdf (27 July 2021, date [32] California ISO. http://www.caiso.com/todaysoutlook/pages/ last accessed). supply.html and http://www.caiso.com/informed/Pages/ [44] Air Liquide. Air Liquide committed to producing renewable ManagingOversupply.aspx (23 December 2020, date last hydrogen for the West Coast mobility market with new li- accessed). quid hydrogen plant. Press release. 8 October 2019. https:// [33] California Energy Commission (CEC). Final 2019 Integrated www.airliquide.com/united-states-america/air-liquide- Energy Policy Report. Report19-IEPR-01. pp. 70–113, 227–245. 2020. committed-producing-renewable-hydrogen-west-coast- https://www.energy.ca.gov/data-reports/reports/integrated- mobility-market (23 December 2020, date last accessed). energy-policy-r e port/2019-inte gr ated-energy-policy- [45] Cheng AJ, Tarroja B, Shaffer B, et al. Comparing the emissions report/2019-iepr (23 December 2020, date last accessed). benefits of centralized vs. decentralized electric vehicle smart [34] California Public Utilities Commission. Decision Approving charging approaches: a case study of the year 2030 California Retirement of Diablo Canyon Nuclear Power Plant. 11 January 2018. electric grid. J Power Sources, 2018, 401:175–185. https://docs.cpuc.ca.gov/PublishedDocs/Published/G000/M205/ [46] Hingtgen J, Le D, Davis B, et  al. Productivity of Wind Generation K090/205090240.PDF (23 December 2020, date last accessed). in California 2014 through 2016. California Energy Commission [35] Gerke B, Gallo G, Smith S, et al. The California Demand Response staff report CEC-200-2019-002. 2019. https://ww2.energy. Potential Study, Phase 3: Final Report on the Shift Resource through ca.gov/2019publications/CEC-200-2019-002/CEC-200-2019- 2030. 2020. https://escholarship.org/uc/item/7bx121k6 (23 002.pdf (23 December 2020, date last accessed). December 2020, date last accessed). [47] Wang  M, Ullrich  P, Millstein  D. The future of wind energy in [36] California Energy Commission and California Air Resources California: future projections with the Variable-Resolution Board. Joint Agency Report on Assembly Bill 8: 2019 Annual CESM. Renewable Energy, 2018, 127:242–257. Assessment of Time and Cost Needed to £attain 100 Hydrogen [48] Tarroja  B, Mueller  F, Eichman  JD, et  al. Spatial and temporal Refueling Stations in California. CEC-600-2019-039. 2019. https:// analysis of electric wind generation intermittency and dy- ww2.energy.ca.gov/2019publications/CEC-600-2019-039/CEC- namics. Renewable Energy, 2011, 36:3424–3432. 600-2019-039.pdf (23 December 2020, date last accessed). [49] California Air Resources Board. LCFS Electricity and Hydrogen [37] California Air Resources Board. Innov ative Clean Transit—Rollout Provisions. https://ww2.arb.ca.gov/resources/documents/lcfs- Plans. https://ww2.arb.ca.gov/our-work/programs/innovative- electricity-and-hydrogen-provisions. Appendix H: Low Carbon clean-transit/ict-rollout-plans (23 December 2020, date last Fuel Standard (LCFS) Program and Examples. 2020. https://ww3. accessed). arb.ca.gov/regact/2018/ict2018/apph.pdf (23 December 2020, [38] Reddi  K, Elgowainy  A, Sutherland  E. Hydrogen refueling date last accessed). station compression and storage optimization with tube- [50] California Air Resources Board. Weekly LCFS Credit Transfer trailer deliveries. International Journal of Hydrogen Energy, 2014, Activity Reports. 2021. https://ww3.arb.ca.gov/fuels/lcfs/ 39:19169–19181. credit/lrtweeklycreditreports.htm (6 January 2021, date last [39] Li XJ, Allen JD, Stager JA, et  al. Paths to low-cost hydrogen accessed). energy at a scale for transportation applications in the USA [51] United Nations. World Urbanization Prospects: The 2018 and China via liquid-hydrogen distribution networks. Clean Revision. File 12: Population of Urban Agglomerations with 300,000 Energy, 2020, 4:26–47. Inhabitants or More in 2018, by Country, 1950–2035 (Thousands) [40] Schmidt  O, Gambhir  A, Staffell  I, et  al. Future cost and per - POP/DB/WUP/Rev.2018/1/F12. https://population.un.org/wup/ formance of water electrolysis: an expert elicitation study. Download/ (23 December 2020, date last accessed) (29 January International Journal of Hydrogen Energy, 2017, 42:30470–30492. 2021, date last accessed). [41] Petitpas  G, Aceves  SM. Liquid hydrogen pump performance [52] Dong  D, Duan  H, Mao  R, et  al. Towards a low carbon transi- and durability testing through repeated cryogenic vessel tion of urban public transport in megacities: a case study of filling to 700 bar. International Journal of Hydrogen Energy, 2018, Shenzhen, China. Resources, Conservation & Recycling, 2018, 43:18403–18420. 134:149–155.

Journal

Clean EnergyOxford University Press

Published: Sep 1, 2021

References