5. Hydrogen applications in practice

The energy transition is transforming the hydrogen landscape by spurring new ways of producing, transporting and using hydrogen. While countries currently mainly use hydrogen from fossil fuels in industrial processes, wider and “greener” hydrogen applications with larger-scale electrolyser production using renewable energy sources are foreseen in the future (IEA, 2021[1]).

While there are many ways in which hydrogen can be applied, this chapter focuses on six main scenarios covering the entire hydrogen value chain. These scenarios are based on hydrogen applications that could support the Dutch climate targets. For each scenario, the report discusses current practice and regulation and zooms in on specific situations with a higher level of risk.

Findings on safety risks, measures and regulation are based on other outputs from the project on Precaution in the Energy Transition and Improved Knowledge for Hydrogen Risk Regulation. These outputs have been included as separate Parts to the current report.

The scenarios included in the report cover different parts of the hydrogen lifecycle from production to usage and have been selected at the request of the Dutch Ministry of Economic Affairs and Climate Policy. They are of particular interest as they cover use of hydrogen technology in densely populated areas requiring safety and risk management techniques. The six scenarios are:

  1. 1. Production through water electrolysis

  2. 2. Pipeline transport

  3. 3. Road transport

  4. 4. Mobility and partially confined spaces: tunnels

  5. 5. Mobility and partially confined spaces: hydrogen refuelling stations

  6. 6. Domestic use

Not all these cases have the same degree of technological maturity, and domestic use is really at a pilot stage only right now. Several cases, however, are well known and understood, present risks that are readily manageable with good practices, and are generally not higher (or lower) than their closest comparable “hydrocarbon fuel” competitors. Nonetheless, they often are subject to more constraining regulations than the said “hydrocarbon competitors”, because of not having a specific regulatory framework (ending up regulated as high-risk industrial processes) and in some cases no provision in zoning.

In this context, governments and regulators should identify hydrogen innovations that are a priority for scaling up, and present difficulties in the existing zoning and permitting frameworks. This should lead, where needed, to revisions of zoning and permitting for selected hydrogen applications. Incorporating lessons from practice and research should allow to define zoning rules that enable the development of hydrogen in a safe way, and define permitting processes that are risk-proportionate, particularly for lower-risk facilities and uses – for which high-risk industrial permitting requirements are likely to be disproportionately burdensome.

As the level of knowledge on safety varies for different hydrogen technologies, it makes sense to regulate them differently. When scientific knowledge is more limited and risks are not obvious or simply unknown, additional pilots could be carried out to improve scientific knowledge. Hydrogen technologies can be divided into three broad categories, based on the level of existing knowledge and scientific research:

  • Category 1 – Mature technologies on which there is extensive scientific knowledge and data on safety, such as hydrogen production through electrolysis (in particular alkaline and PEM) and hydrogen refuelling stations. These technologies often do not require additional caution compared with conventional fuels because of the accumulation of operational experience relating to risk management within the technology, and can be facilitated and managed using existing risk management approaches and findings from recent research and good practices;

  • Category 2 – Technologies for which a significant level of scientific knowledge exists but additional practical risk data may be needed, such as driving hydrogen-powered vehicles through tunnels and blending of hydrogen in gas networks. These technologies can be handled through risk management approaches using available scientific knowledge and experience with comparable technologies. However, they will require additional regulatory scrutiny and oversight, using iterative approaches, as scientific knowledge and technology advance;

  • Category 3 – Technologies for which risks are not yet completely understood, such as the domestic use of hydrogen, because the technology is in its infancy or not currently widely deployed and as such there is a lack of standardisation of design and operation. These technologies require further investigation and research through pilot projects, to more reliably assess risks, identify suitable policy approaches, define regulatory requirements, and build public awareness.

This scenario focuses on low-emission hydrogen derived from water electrolysis. This production method is given a prominent role in the hydrogen ambitions of many countries, as — when produced with low-carbon electricity — it allows them to produce hydrogen in a way that supports net zero targets. In particular, the scenario looks at the safety considerations of a hydrogen leakage from pipes connected to electrolysers.

Hydrogen production can be onsite or offsite, depending on the distance to external hydrogen sources and the available transport infrastructure, among other factors. Onsite production can be more suitable for hydrogen consumption without nearby hydrogen sources, whereas offsite production generally involves larger scale production delivered through tube trailers, liquid hydrogen trucks or hydrogen pipelines (Tian et al., 2021[2]).

Hydrogen produced by water electrolysis using low-carbon electricity features prominently in many hydrogen strategies, as it is the main method of producing hydrogen in a way that aligns with green ambitions. However, it currently represents only a small part of hydrogen production; the majority is produced from natural gas and coal (see section on Status quo and future trends in hydrogen use worldwide in Chapter 2).

Because hydrogen can be cheaply produced with steam methane reforming, electrolysis has been mostly applied on a smaller scale (around 1 MW), although in recent years production capacities have increased in response to a drive towards green hydrogen. Most large-scale electrolysis plants were built for the chemical industry (e.g., ammonia production) between the 1920s and 1980s, using alkaline electrolyser technology. (Krishnan et al., 2020[3]). Other electrolyser technologies, including proton exchange membrane (PEM) and solid oxide electrolyser cell (SOEC), are gaining market traction due to their increased flexibility (IRENA, 2018[4]) (Hu et al., 2022[5]) (Sebbahi et al., 2022[6]). PEM technology have more recently shown significant cost reductions, with costs now approaching the costs of alkaline technologies, and are increasing its share in overall installed electrolyser capacity (IEA, 2022[7]). Self-pressurising electrolysers (i.e., operating at pressure) can produce hydrogen at a higher pressure, which can avoid the cost of an additional stage of mechanical compression (Hancke, Holm and Ulleberg, 2022[8]). Hydrogen is typically delivered at 30 bar, although pressure in storage systems and compressors is higher (350 to 700 bar).

In general, an electrolyser production unit consists of the following elements:

  • An electrical power source, along with power electronics

  • An electrolyser stack, that uses electricity to split water (H2O) into oxygen (O2) and hydrogen gas (H2) (Figure 5.1).

  • An external compressor that increases the hydrogen pressure to reduce its volume (for electrolysers other than self-pressuring electrolysers).

  • Pipelines to transport the hydrogen gas for onsite use and/or temporarily stored within pressure cylinders with subsequent use onsite or at another location (Zarei, Khan and Yazdi, 2021[9]).

Water electrolysis accounted for roughly 0.03% of hydrogen production for energy and chemical feedstocks in 2021, but this share is expected to increase significantly. The total global electrolyser capacity in 2020 was 290 Megawatt (MW), with roughly 40% located in the EU1 (Figure 5.2). The total installed capacity increased to over 500 MW in 2021 and was expected to increase to over 1 300 MW by 2022 (IEA, 2022[10]). Alkaline electrolysers accounted for almost 70% of global electrolyser capacity, followed by one quarter for PEM. Production by SOEC and anion exchange membrane (AEM) electrolysis only accounts for a marginal share of overall production. The company Sunfire has recently finished the construction of a 2.6 MW SOEC electrolyser in the Netherlands – the largest SOEC electrolyser in the world (IEA, 2022[7]) (Sunfire, 2023[11]).

Production by electrolysis needs to rapidly increase its share of overall hydrogen production to realise hydrogen’s potential in the fight against the climate crisis and get on track with IEA’s Net Zero by 2050 scenario. Hydrogen production through electrolysis is taking place at an increasingly large scale, with the largest production site reaching a 150 MW capacity and starting operation in 2021 (IEA, 2022[10]). By 2030, total electrolysis capacity could increase to 134-240 GW based on capacity under construction or planned (although this includes a significant share of projects at earlier stages of planning without a final decision).2 Alkaline electrolysis continues to feature more predominantly in new hydrogen production projects due to its lower costs, although new research is driving significant cost reductions especially for PEM (IEA, 2022[7]).

To support the scaling up of hydrogen production, many countries are developing new projects with larger-scale production sites, which may reduce unit costs for low-emission hydrogen. Although these larger-scale production sites can use various raw material sources to produce hydrogen, many put a strong focus on the development of low-emission hydrogen with larger electrolysers. For example, the United Kingdom is expecting to develop a 60 MW hydrogen production capacity through electrolysis by 2025 (BP, 2021[12]). China holds a significant share of hydrogen production through electrolysis, with a 150 MW electrolyser that became operational in 2021 and a new 260 MW electrolyser expected in 2023 (Upstream, 2022[13]). In the Netherlands, a number of larger-scale electrolyser projects are also envisaged, such as the H2.50 project developing a 250 MW capacity electrolyser (expected to become operational by 2025) (IEA, 2021[1]).

Within Europe, Germany is the largest producer of hydrogen by electrolysis, with 33 installations out of the total 114 installations in Europe. France has the second-largest number of installations (23), although most of these have a relatively low production capacity. The Netherlands had four electrolysis production installations in 2020. With a significant planned deployment of new electrolyser capacity in 2021, total European electrolysis capacity was expected to increase to 136 MW in 2021 (FCHO, 2022[14]). The Clean Hydrogen Monitor 2022 estimated the operational electrolysis production capacity in Europe in August 2022 at 162 MW (Revolve, 2022[15]).

In 2021, (blue) hydrogen production with carbon capture, utilisation and storage (CCUS) accounted for 0.7% of total hydrogen production (IEA, 2022[10]). This type of hydrogen often features less prominently in the ambitions for a net zero 2050 because it still relies on fossil fuels to produce hydrogen. However, it is considered by countries in certain cases to reduce carbon emissions especially in the medium term, such as through the retrofitting of existing fossil-based hydrogen production plants (European Commission, 2020[16]). In 2021, the IEA reported 40 projects for producing hydrogen with CCUS that were under development at that time. Of these, 19 were in Europe, predominantly in the Netherlands and the United Kingdom.

The Netherlands is taking significant steps to increase its production of blue hydrogen. It recently committed EUR 2 billion to fund the Portus project in the Port of Rotterdam, which will store 2.5 megatons of CO2 annually, with a significant share coming from hydrogen production (Porthos, 2022[17]). Moreover, two projects with blue hydrogen production are already operational in the EU, one of which is at the Pernis refinery in Rotterdam (with CCUS expected from 2024). In December 2022, the Netherlands also announced the funding of seven projects for the production of hydrogen through electrolysis, for a total additional capacity of 1 150 MW (Rijksoverheid, 2022[18]).

Given the large ambitions for hydrogen production through water electrolysis, there will be a need for a robust and conducive regulatory framework that can facilitate them. A particular challenge will be to ensure risks are managed effectively – but not excessively – through clear direction and guidelines. An up-to-date overview of scientific knowledge on the risks and measures and experience from other countries facing similar challenges will be crucial.

Not all components at a hydrogen electrolysis production site are equally prone to the risk of failures. The typical events that could result in a hydrogen leak during production are mechanical failure of components such as compressors or pipework, overpressure in one of the components, corrosion or damage due to impact, and human error such as an accidental opening of valves.

Research suggests that current hydrogen production presents a lower normalised3 fatality risk as compared to the production of oil, coal and natural gas (see Part 4 – Review on incident database and lessons learnt) (Brook et al., 2014[19]). Risk analyses on hydrogen production show that most causes of accidents (and therefore the likelihood of harmful consequences) can be either eliminated or reduced by following recommended safety measures, such as safety valves, leakage detectors, fire walls, “spark-free” components and the siting of facilities at a safe distance from vulnerable populations (Zarei, Khan and Yazdi, 2021[9]) (Kasai et al., 2016[20]) (Schefer et al., 2009[21]).

Key factors in the likely occurrence of an explosion or other hydrogen accident such as a fire are the gas concentration in air, as well as the degree of confinement of the released flammable gas. The flammable mixture that is created by a high pressure leak (at 350 bar)extends to three times the distance than that of a low pressure (at 30 bar) release. Flammable mixtures around stoichiometric composition (~30% vol. hydrogen-air mixture) are more hazardous due to their higher burning velocities . However, they are limited to a smaller area around the leak, even with high pressure releases. Higher degrees of congestion can also lead to increased risks of explosion.

The main causes of hydrogen accidents at installations are typical of the type and range seen in conventional hydrocarbon-based industry sectors (see Part 4 – Review on incident database and lessons learnt).4 The component most prone to failure in the past is the hydrogen pipework, often related to a valve failure. However, the frequency of pipeline failures decreased with the introduction of modern valve design and safety regulation, and pipework accidents are less likely to have fatal consequences than in the past. The hydrogen compressor has the highest ratio of number of deaths to number of accidents, followed by accidents in storage. Looking at the root causes of accidents, the most frequent cause was equipment failure, followed by deficiencies in procedures leading to human error. This emphasises the importance of guidance on the expected lifespan of critical components such as safety valves.

The databases, as well as findings in scientific literature, confirm that the compressors and high-pressure storage vessels are the major risk contributors, whereas the risks associated with the electrolyser itself are relatively small in comparison (Pan et al., 2016[22]) (FCH 2 JU, 2020[23]). Flash fires caused by a rupture to a stationary high-pressure storage vessel can lead to harm over a wider distance of tens of metres, placing the compressor and storage systems at higher risk than other components.

Calculations by Sandia National Laboratories (based on historical hydrogen data for compressors, cylinders, hoses, joints, pipes and valves) showed that the leaking frequency of connecting pipes (between 2.99 X 10-9 m-1 year-1 for very small leakage and 3.13 X 10-10 m-1 year-1 for rupture at 95% confidence level) falls within the acceptable range set by Purple book – a guideline for quantitative risk assessment (QRA) in the Netherlands (Glover, Baird and Brooks, 2020[24]).

Annex Box 1.A.1 presents the safety measures that can be considered to decrease the risks related to hydrogen production through electrolysis. Risk measures should be decided upon based on desired risk targets and taking into account countervailing risks (see Chapter 1 – “Managing risks”).

There is no hydrogen-specific safety regulation for hydrogen production in the Netherlands. A hydrogen production site in the Netherlands is considered as a standard chemical manufacturing facility for licensing and safety regulation purposes. This treatment does not differ depending on whether the hydrogen is produced through electrolysis with alkaline, PEM, or SOEC, or through natural gas reforming (HyLAW, 2018[25]). Hydrogen production and storage is also governed by European Commission directives on chemical processes involving emissions,5 including the Seveso Directive.

There is no distinction in licensing procedures between fossil fuels and greener forms of hydrogen production, or between onsite or offsite production. Furthermore, there is no simplified process whereby the procedures laid down in the Act on the general provisions environmental law (the so-called “Wabo-procedure” — Wet Algemene bepallingen omgevingsrecht, Wabo) are further defined according to the scale or type of production or scale (HyLAW, 2018[25]). This could slow down the rollout of smaller-scale, onsite low-emission hydrogen production, such as that linked to hydrogen refuelling stations, as they will face the same procedures as larger producers (see Part II – “Regulatory review”).

The authority in charge of licensing, safety inspections and enforcement may differ depending on the scale of production (or more specifically, the storage capacity). The province is the responsible authority for installations with quantities of hydrogen storage above five tonnes. In other cases, the municipality is the responsible authority. In general, the authority in charge of licensing will also be responsible for inspections and enforcement. Moreover, the respective authority can assign to the ODs the tasks to issue licences, inspect and enforce (see Chapter 4 – “Licensing and inspections for hydrogen activities”).

In general, as part of the process to apply for land use and environment permits, hydrogen production plants need to adhere to the following requirements:

  • Conducting risk assessments in line with the Brzo 2015 decree (the implementation of the Seveso-III Directive in Dutch legislation)6 and the Bevi decree.

  • Meeting health and safety requirements (in line with the EU Equipment Directive intended for use in explosive atmospheres (Appareils destinés à être utilisés en Atmosphères Explosibles, ATEX) and other relevant legal codes).

  • Aligning with integrated environmental obligations (following from the EU IED Directive).

  • Conducting environmental impact assessments and strategic environmental assessments (based on the EU EIA and SEA Directives) (HyLAW, 2018[25]).

There are differences across countries in the way in which hydrogen is regulated, and in particular, whether rules are specifically tailored to hydrogen. As a somewhat more mature and well-developed technology, hydrogen production via electrolysis is often subject to a more advanced regulatory framework than some of the other scenarios. In China and Korea, codes and standards on hydrogen are in force, whereas in many other countries the requirements for hydrogen production follow general regulations for flammable gasses. Moreover, for jurisdictions such as the EU and Japan, hydrogen requirements differ according to the storage capacity rather than the production capacity. A few highlights:

  • China has legally binding standards for the safe design and maintenance of hydrogen production stations. The relate to restrictions on maximum allowable storage capacity, operational conditions, safety equipment, technical specifications of pipework and safety requirements, such as the minimum ventilation rate, separation and safety distances, etc..

  • In the EU, notification of the regulatory authority is required for production or storage of more than five tonnes based on the Seveso Directive (Lexparency, 2008[26]). There is a requirement to draw up a written safety policy for the prevention of hazardous accidents. Storage greater than 50 tonnes requires a safety report and emergency plan to be prepared, submitted to and assessed by the competent authority.

  • In Japan, the requirements for hydrogen production facilities are set under the regulation of high-pressure gas facilities.

  • Korea has developed codes that cover most of the requirements for hydrogen production and storage facilities.

  • In the United States, hydrogen production facilities are governed by Occupational Safety and Health Administration’s (OSHA) standards and the National Fire Protection Association’s NFPA-2 standard which, among other issues, define safety and separation distances as well as requirements for safety systems.

Hydrogen production through water electrolysis is, in many cases, regulated under the more general regulatory requirements for flammable gases. Going forward, the increasing scale at which this technology is expected to be applied may prompt countries to review existing requirements, to assess if they will facilitate the expected rapid increase in electrolysis production. This could create a need for more specific guidelines, standardisation, and risk management good practices to support its smooth deployment.

A further discussion of regulatory practices of hydrogen production across countries is presented in Part 2 – “Regulatory review”.

The international standard ISO 22734:2019 has been developed to cover construction, safety and performance requirements for hydrogen gas generation appliances, i.e. electrolysing water to produce hydrogen. The standard applies to electrolysers for industrial and commercial use, and indoor and outdoor residential use in sheltered areas (such as garages, utility rooms and similar residential locations).

The current scenario focuses on the transport of compressed gaseous hydrogen through high-pressure pipelines.

Hydrogen can be transported in gaseous form by pipelines in cases where there is a sufficiently large, sustainable and localised demand (IEA, 2019[27]). Analysis by the IEA shows that transporting gaseous hydrogen by pipeline can often be a cost-efficient option for a wide range of distances, depending on pipeline capacity and whether it is a new or repurposed pipeline (Figure 5.3) (IEA, 2022[10]). Recent research in the area suggests pipeline transport may even be, under certain conditions, the most cost-efficient option for distances up to 10 000 kilometres (Perey and Mulder, 2023[28]). For longer distances, transport of hydrogen as a liquid or converted into ammonia may be more attractive due to the higher energy densities per volume for these substances (IEA, 2021[1]).

Historically, carbon steel or stainless-steel pipelines have been used for high-pressure hydrogen transmission. This is because higher grades of steel (above 100 ksi7) can more easily lead to hydrogen embrittlement of pipelines (IEA, 2021[1]). Hydrogen embrittlement is due to the technical properties of hydrogen gas (see Chapter 2 – “Understanding and managing hydrogen risk”).

Transport of hydrogen by pipeline is a mature technology, which was first employed in the Rhine-Ruhr area in Germany in 1938. Pipelines can be newly-built or repurposed or retrofitted8 natural gas pipelines (IEA, 2022[10]). Similar to the construction of natural gas pipelines, constructing hydrogen pipelines is capital-intensive and requires long-term investments with high upfront costs.

Investment costs of newly-built hydrogen infrastructure tend to be higher than for natural gas. At similar pipeline diameters, the IEA estimates that capital expenditure for hydrogen-specific steel pipelines can be 10 to 50% higher (IEA, 2021[1]). Additionally, the same amount of energy transported requires a higher volume of hydrogen gas as compared with natural gas (by a factor three). This is due to the low energy density by volume of hydrogen (see Chapter 2 – “Understanding and managing hydrogen risk”) (IEA, 2019[27]).

However, costs can be reduced by repurposing existing natural gas infrastructure. The HyWay27 project estimates that reusing natural gas pipelines is four times more cost-effective than constructing new hydrogen pipelines (HyWay27, 2021[29]).9 The suitability of gas infrastructure to be repurposed as hydrogen pipelines will depend on the type of steel used in the pipeline and the purity of hydrogen being transported (where higher concentrations of hydrogen may lead to embrittlement of pipelines) (see Scenario 6 – Domestic use) (IEA, 2019[27]).

In 2021, roughly 5 000 kilometres of hydrogen pipeline were in operation worldwide, compared with roughly 3 million kilometres of natural gas pipeline (IEA, 2019[27]). More than 90% of total hydrogen infrastructure is in Europe and the United States.10 The current infrastructure is mostly made up of closed systems owned by large hydrogen producers within or near industrial sites with chemical plants or oil refineries (IEA, 2021[30]).

Repurposing of natural gas pipelines is a relatively new trend, where earlier hydrogen pipelines are mostly newly built or converted pipelines originally for other fuels. In the 1970s and 1990s, pipelines for crude oil and related products were repurposed to transport hydrogen. Gasunie in the Netherlands was the first to repurpose a natural gas pipeline for hydrogen transport with a total length of 8 kilometres, put into commercial service in 2018 (IEA, 2021[30]). Similar repurposing projects with longer hydrogen pipelines are envisaged in other countries such as Germany and Australia (IEA, 2021[1])

A consortium of 31 European infrastructure operators, the European Hydrogen Backbone (EHB) initiative, aims to roll out a trans-European hydrogen grid (EHB, 2022[31]). Its 2020 proposal envisages 39 700 kilometres of hydrogen pipeline across 21 countries by 2040. The initiative foresees a significant role for the repurposing of natural gas infrastructure to reduce cost and make new use of existing infrastructure, with 69% of the envisaged network consisting of repurposed gas pipelines compared with 31% newly built (IEA, 2021[30]).

Two studies from 2017 and 2018 indicate that the existing Dutch natural gas infrastructure could be used to transmit hydrogen with certain modifications (DNV, 2017[32]) (Netbeheer Nederland, 2018[33]). The studies argue that the pipeline grades used in the Netherlands are suitable for hydrogen transport. As the utilisation of the Dutch natural gas transmission network decreases as natural gas is substituted by other energy forms, otherwise “stranded assets” may be converted to hydrogen pipelines. Many countries report on-going pilots to determine the impact of hydrogen on different gas pipeline materials, including in subsea transmission.

Hydrogen may also be injected into existing natural gas infrastructure as a blend with natural gas. Research on the impact of blending hydrogen into natural gas pipelines is still at an early stage, with further evidence needed on the performance and durability of pipelines for different levels of blends and required maintenance activities. There is an increasing scientific interest to understand the impact of different blends of hydrogen on different pipeline materials, including so-called “killed” steels that improve the toughness of steel pipelines (EIGA, 2004[34]).

The main risk related to hydrogen transport by pipeline is the possibility of hydrogen leakage and subsequent ignition. Leakage can occur either through failure of flanged pipe joints or due to damage to pipelines such as corrosion or due to impact. An incident with hydrogen leakage can have several consequences including fire, explosions or unignited releases. The exact risk level and consequence depends on factors including the type of failure, hole size, pipeline pressure, ignition probability, time to ignition, meteorological conditions, pipeline condition and soil type (for buried pipelines. The successful operation of safety systems will also be a factor.11

New hydrogen transmission pipelines are usually buried underground. This can support the safety and reliability of hydrogen transport and protect against accidental damage and frost. However, their underground location means there is also a requirement for pipeline protection against excavation accidents, the impact of shifting soil and heavy loads imposed on the soil due to heavy-duty vehicles or equipment.

The properties of hydrogen gas in comparison with natural gas affect its risk profile, where its relative risk depends on the context in which it is released. Hydrogen is lighter than air or natural gas and its volume leakage from pipelines is generally approximately 1.3 to 2.8 times larger than methane leakage (Rigas and Amyotte, 2013[35]). For underground hydrogen releases, the pipe depth, release orientation and soil properties will determine whether a crater is formed due to the release pressure and how quickly the hydrogen disperses. Due to its properties, in the case of leakage, hydrogen diffuses more quickly in air compared with natural gas. Moreover, for the same mass flow, hydrogen leaks are greater in volume flow than those of natural gas. The subsequent risk depends on the level of hydrogen concentration built up following an incident. For hydrogen concentrations in air below 10% vol., hydrogen has a minimum ignition energy similar to that of natural gas and its combustion results in hardly any overpressure. For concentrations above 10% vol., hydrogen presents a greater risk as it is more likely to deflagrate with the pressure building up faster.

An analysis of incident data shows a slightly lower normalised12 incident rate for hydrogen compared with natural gas, although this value could change once hydrogen pipelines become more widespread (see Incident database report). Reported incidents in incident databases equate to 0.09 incidents per 1 000 km of pipeline per year, compared with 0.13 (Europe) to 0.16 (United States) incidents for natural gas. However, this value is based on relatively few reported incidents for hydrogen, as its transport through pipelines is still limited. As hydrogen pipeline networks will grow, this will provide for additional data to assess the exact incident rate more accurately.

A comparative (theoretical) risk study of hydrogen and methane in pipelines found that, in the case of immediate ignition of a hydrogen leak, the increase in expected risks for hydrogen is negligible when compared to methane (see Part 7 – Quantitative risk assessment: Hydrogen versus conventional fuel). However, the modelling did not consider cases of delayed ignition and showed that assumptions on the ignition probability have an important impact on results, both of which should be explored further.

The types of hydrogen incident are typical of those hazards observed with other major pipelines such as natural gas or liquid hydrocarbons. The reported root causes of hydrogen incidents include design errors, human error, inadequate maintenance and deficiencies in procedures. Most incidents resulted in hydrogen fires, with others resulting in explosions or the unignited release of hydrogen.

Annex Box 1.A.2. presents the safety measures that can be considered to decrease the risks related to hydrogen transport by pipeline. Risk measures should be decided upon based on desired risk targets while taking into account countervailing risks (see Chapter 1 – Managing risks).

The Dutch government has not yet developed an overarching regulatory framework for high-pressure hydrogen transport through pipelines in the Netherlands, as it awaits a broader EU directive first (EZK, 2021[36]). However, the Decree external safety pipelines (Bevb) applies also for hydrogen. Within this framework the National Institute for Public Health and the Environment (Rijksinstituut voor Volksgezondheid en Milieu, RIVM) did provide advice on the failure frequency of hydrogen pipelines and the preferred calculation method to assess the risk of hydrogen pipelines (RIVM, 2021[37]).

On 15 December 2021, the European Commission adopted a legislative proposal to update the 2009 EU Gas Regulation that includes a legislative framework for hydrogen networks (European Commission, 2021[38]). The proposal aims to support “the development of a cost-effective, cross-border hydrogen infrastructure and competitive hydrogen market”.

To achieve this, it proposes a number of rules for hydrogen networks and markets. These span:

  • Tariffs for the transmission and distribution of hydrogen by system operators will be approved by a regulatory authority.

  • Ownership unbundling for hydrogen network operators.

  • Regulated third party access to hydrogen networks.

  • A transition period until 31 December 2030, until which existing private hydrogen networks may be exempt from certain access requirements.13

The existing legal framework in the Netherlands already offers certain opportunities for infrastructure companies to develop hydrogen infrastructure, as was done by Gasunie in 2018 (see State of play) (EZK, 2021[36]). The envisaged new Energy Act specifies that infrastructure companies can be involved in activities including the development and maintenance of hydrogen networks, the transport of hydrogen and metering activities. However, it does not allow system operators, or the holding companies to which they belong, to produce, trade or supply hydrogen (as is also the case for their involvement with electricity and gas) (EZK, 2021[39]).

Hydrogen pipelines do not require a specific licence. However, developers and operators of larger hydrogen pipelines14 need to comply with the Decree on the external safety pipelines (Besluit externe veiligheid buisleidingen, Bevb). Among other things, this decree requires operators of hydrogen pipelines to implement a safety management system. The construction or replacement of pipelines is only allowed when it is aligned with the zoning plan or an environment permit has been issued, so as to prevent major accidents or disproportional risks to vulnerable people or buildings. The local risk15 (plaatsgebonden risico) for vulnerable objects in the proximity a of pipeline shall not exceed a set threshold of 10-6 (or one in a million) per year. Moreover, the pipeline operator is required to construct or replace the pipeline in such a way that the local risk does not exceed the 10-6 per year threshold at a distance of five meters from the heart of the pipeline.16 Additionally, the Bevb also defines the criteria and thresholds for the group risk.17 Acceptance criteria for hydrogen pipelines are similar to those for natural gas pipelines.

There is no common approach across countries as to how hydrogen pipelines are regulated:

  • In some countries, such as Australia and Germany, regulations have been amended to allow hydrogen to be transmitted though existing pipelines. In Australia, injection of up to 10% hydrogen into natural gas pipelines has been allowed, whereas in Germany an ordinance was passed to allow operators to use existing natural gas infrastructure for hydrogen.

  • In the UK, hydrogen transport through pipelines requires permission and must adhere to pipeline requirements for design, safety systems, construction, installation, operation, maintenance, and decommissioning as well as to industry codes such as the Pipeline Safety Regulations Act of 1996.

  • In Japan, even though the transport of hydrogen is limited to short-distance uses, there are safety regulations for the pipe layout and pipe materials. However, many of them are still being verified.

  • In the United States, regulations for flammable gases in hydrogen pipelines are applied. The American Society of Mechanical Engineers (ASME) provides standards for piping and transportation pipelines. Requirements for piping in gaseous and liquid hydrogen service, and for pipelines in gaseous hydrogen service can be found in the ASME B31.12 Standard on Hydrogen Piping and Pipelines. This standard covers the requirements for materials, brazing, welding, heat treating, forming, testing, inspection, examination, operating and maintenance.

  • China has developed a national code that sets general requirements for pipelines.

A further discussion of regulatory practice across different countries for hydrogen transport by pipeline is presented in Part II – “Regulatory review”.

This scenario involves the transport of hydrogen by road. This includes both hydrogen-powered vehicles (such as FCEVs) and vehicles transporting a hydrogen cargo that is not intended as fuel for the vehicle itself.18 For ease of reference, the latter will be referred to as ‘vehicles transporting hydrogen’, although technically hydrogen-powered vehicles also transport hydrogen but with the distinction that this serves only for the consumption by the vehicle itself. The scenario looks in particular at the presence of hydrogen in road transport within built-up areas – including the potential for incidents such as hydrogen leakage in parking garages and road accidents.

Vehicles transporting hydrogen

In the absence of pipelines or onsite production, hydrogen can be transported by road, supplying hydrogen from production sites to consumption sites such as industrial users and hydrogen refuelling stations. Transport by road includes the transport of hydrogen in gas tanks, metallic cylinders, tubes or composite vessels. Hydrogen may be transported as a compressed gas or as liquid19 (IEA, 2019[27]).

Transporting hydrogen as a liquid has certain advantages over gaseous hydrogen especially for longer-distance transport, as a liquid tanker truck is able to hold larger quantities of hydrogen than gaseous tube trailers can. The liquefaction involves cooling hydrogen to temperatures as low as minus 253 degrees Celsius. However, hydrogen liquefaction is a process that is energy intensive, consuming as much as 30% of the hydrogen’s energy content when using current technologies (US Department of Energy, n.d.[40]).

Hydrogen-powered vehicles

Hydrogen-powered vehicles can be passenger cars, as well as medium to heavy-duty vehicles such as buses, commercial vehicles and trucks.

FCEVs use a fuel cell (or “fuel cell stack”) to produce electricity (Figure 5.4). Hydrogen is stored in a fuel tank – usually as a compressed gas for more efficient storage20 – and is converted in the fuel cell into electricity and water.21 The electricity is then used to power the motor. This can be used in combination with a battery pack, which smooths out the power delivered from the fuel cell, recaptures braking energy and provides extra power for acceleration. (US Department of Energy, n.d.[41]). The most common type of fuel cell for FCEVs is the PEM fuel cell (US Department of Energy, n.d.[42]).

Most local distribution of hydrogen currently takes places by trucks carrying hydrogen gas (IEA, 2019[27]). However, there is no information available on the exact volume of hydrogen transported by road in the Netherlands, Europe or worldwide.

Hydrogen-fuelled vehicles still makes up only a small share of the transport sector, but the number of hydrogen-powered vehicles has experienced significant growth levels recently. In 2020, FCEVs made up less than 0.01% of total road vehicles worldwide and 0.3% of total electric vehicles. However, the number of FCEVs grew at an average annual rate of 70% over the period from 2017 to 2020 and there were 51 600 FCEVs on the road in 2021, as well as 730 hydrogen refuelling stations (Figure 5.5). Due to substantial subsidies to increase the adoption of FCEVs, Korea is the largest stockholder globally when it comes to the number of vehicles, although Japan has the largest network of public refuelling stations.

Regional priorities in FCEV deployment differ. Efforts to increase the use of FCEVs focus mostly on passenger cars in Korea, Japan and the United States, whereas China and Europe focus more medium- and heavy-duty commercial vehicles such as buses and trucks. China has the largest fleet of both fuel cell buses and trucks, with a total of more than 8 400 vehicles (or 90% of fuel cell buses and 95% of fuel cell trucks worldwide) (IEA, 2022[43]). Countries are also increasingly showing an interest in using hydrogen in other non-road transport, with plans and pilots for the use of hydrogen in trains, trams, ferries, ships and aviation (IEA, 2021[30]).

The total FCEV fleet in Europe in March 2022 consists of 4 050 vehicles, of which roughly a third was deployed in Germany. The Netherlands held the third largest fleet with around 550 vehicles (FCHO, 2022[45]).

Current numbers of FCEVs across the world are still far from ambitions for the future. The IEA Net Zero by 2050 Scenario will require a large increase in the number of vehicles, from the current number of 51 600 in 2021 to 15 million vehicles by 2030 (see Chapter 2 – “Future trends”).

Incidents with hydrogen-powered vehicles and vehicles transporting hydrogen can be caused by the leakage of hydrogen from tanks or cylinders. Such leaks can be due to equipment failure, inadequate maintenance of components, tank ruptures, corrosion or the release of hydrogen through a pressure relief device. In many cases, incidents are caused by external factors such as traffic accidents.

The risks related to hydrogen leakage in transport depend on the level of confinement in which accidents occur. Hydrogen is less likely to cause a fire or explosion in open or well-ventilated spaces, where it can disperse more easily. Within covered and poorly ventilated spaces, hydrogen concentrations can — due to the gas’ buoyant nature — build up close to the ceiling. Natural or mechanical ventilation can reduce the level of hydrogen concentration and thereby reduce the risk of fires or explosions. Therefore, the safety of hydrogen vehicles in confined spaces, such as parking garages, maintenance workshops and covered refuelling stations, may be especially relevant, making an urgent case for efficient ventilation.

In total, 71 incidents were reported in the HIAD 2.0 and H2tools incident databases (see Part 4 – Review on incident database and lessons learnt). Fifty-three incidents (or 75%) involved vehicles transporting hydrogen, whereas 18 involved hydrogen-powered vehicles. The causes of such accidents are relatively comparable to those for liquefied petroleum gas (LPG) vehicles. Forty-two per cent of incidents were caused by traffic accidents, whereas equipment failure was the cause in 15% of cases. Other less frequent causes include design error, human error, inadequate maintenance, deficiencies in procedures and external factors.22

Consequences of incidents differ between hydrogen-powered vehicles and vehicles transporting hydrogen. For vehicles transporting hydrogen, 53 incidents were reported, of which 58% resulted in no or only an unignited release, whereas 28% resulted in a fire and 13% in an explosion. For hydrogen-powered vehicles, 18 incidents were reported, of which 17 resulted in no or only an unignited release and only one resulted in a fire. (It should be noted, however, that most of these hydrogen-powered vehicles were busses in use as part of a pilot project. As a result, minor incidents were reported that may not have been reported otherwise). Looking forward, the deployment of hydrogen in road transport is projected to increase, providing additional data as to the causes and consequences of accidents.

In Japan, a comparison between LPG and hydrogen-powered vehicles shows a somewhat higher risk for hydrogen-powered vehicles (see Part 4 – Review on incident database and lessons learnt). Evidence on traffic incidents for both vehicle types shows an incident probability for hydrogen-powered vehicles of 0.0026% per vehicle per year (i.e. 1 in every 38 461 vehicles being involved in an accident each year). At the same time, it found an incident probability of 0.0003% (i.e. 1 in every 333 333 LPG vehicles being involved in an accident). However, these probabilities may not yet be very precise, given the low overall number of FCEVs currently in operation. Nonetheless, causes of accidents for both vehicles were comparable, and a further rollout of FCEV will likely provide further insights into the exact probability of accidents and allow more meaningful comparisons.

Annex Box 1.A.3 presents the safety measures that can be considered to decrease the risks related to hydrogen transport by road. Risk measures should be decided upon based on desired risk targets while taking into account countervailing risks (see Chapter 1 – Managing risks).

The regulation of vehicles transporting hydrogen in the Netherlands is in accordance with the Carriage of Dangerous Goods by Road Regulation (ADR) from 1957 under the auspices of the United Nations Economic Commission for Europe (UNECE). The ADR is an international regulation that regulates the transportation of hydrogen in cylinders, tubes, trailers and tank vehicles. It specifies packaging types, load security, the classification and labelling of dangerous goods, and the training of drivers. The Economic and Social Council Committee of Experts on the Transport of Dangerous Goods, organised by the UNECE, develops and updates safety provisions for the transport of hydrogen by all modes of transport. These provisions are included in the UN Model Regulations on the Transport of Dangerous Goods. The ADR is frequently revised, with a new edition that came into force on 1 January 2023 (UNECE, 2022[46]).

The municipality is charged with a number of licensing functions regarding vehicles transporting hydrogen. Vehicles transporting hydrogen require an exemption to transport hydrogen across roads other than those designated by authorities for the transport of dangerous substances, as well as for loading and unloading purposes. This exemption can be requested from the municipality (Rijksoverheid, 2015[47]). Additionally, staging areas for tube trailers require an environment permit from the municipality.

Hydrogen-powered vehicles are not subject to the ADR Regulation, which exempts the carriages of “gases contained in the fuel tanks or cylinders of a vehicle performing a transport operation and destined for its propulsion or for the operation of any of its equipment used or intended for use during carriage” (UNECE, 2022[46]). There are no specific regulations, codes and standards for hydrogen-powered mobility. The parking of hydrogen-powered vehicles in parking garages is currently not subject to the Publication Series Dangerous Substances Guideline PGS26 (H2Platform, 2020[48]).

Most countries currently apply regulations that were developed for other flammable gases in their regulation of vehicles transporting hydrogen. Countries in Europe apply the ADR, ensuring consistency between the Dutch and European systems. Australia applies the Dangerous Goods Safety (Road and Rail Transport of Non-explosives) Regulations of 2007 and the Australian Dangerous Goods Code. Training of transport company employees on the associated risks of these goods is obligatory in France.

The Global Technical Regulation (GTR) No. 13 defines vehicle requirements for hydrogen FCEVs, including equivalent (or higher) levels of safety as those required for conventional, fuel-powered vehicles. It includes specifications on the allowable hydrogen levels within vehicle enclosures during in-use and post-crash conditions and on the allowable hydrogen emissions levels of vehicle exhaust during certain modes of normal operation. GTR can be applied globally; however, the regulatory bodies in each country decide its incorporation into national regulations.

The International standard ISO 11623 provides requirements for the periodic inspection of certain composite transportable gas cylinders (ISO, 2015[49]).

This scenario involves the transport of hydrogen by road through tunnels. In particular, it looks at situations where traffic accidents may lead to hydrogen releases in tunnels, and the corresponding risks. An example of this scenario would be a hydrogen bus driving through a tunnel and being involved in a collision.

The transport of hydrogen by road through tunnels includes two distinct categories: hydrogen-powered vehicles and vehicles transporting hydrogen (see Scenario 3 – Road transport). Both categories involve vehicles that carry an amount of hydrogen that might be released inside the tunnel in the case of a traffic accident or an involuntarily leak due to a mechanical malfunction or human error. For the current scenario, these, categories are referred to together as hydrogen vehicles.

At present, there are no available figures on the volume of hydrogen transported by road through tunnels, although volumes are expected to be relatively modest given the current level of hydrogen deployment in road transport (see Scenario 3 – Road transport). However, as ambitious targets for the hydrogen transition and its deployment in transport are set by many countries, the use of hydrogen vehicles is expected to increase (see Status quo and future trends in hydrogen use worldwide in Chapter 2). In turn, this will likely increase the number of hydrogen vehicles using tunnels in the future.

Hydrogen vehicles can pose different levels of risk in enclosed environments such as tunnels, due to the properties of hydrogen and the high levels of confinement. In the open air, hydrogen releases will disperse quickly due to the low weight of hydrogen. However, in tunnels and other closed spaces, accidental leaks of hydrogen from vehicles can be trapped or accumulate below the ceiling or in cavities, leading to the build-up of higher hydrogen concentrations. Therefore, particular attention should be paid to the safe use of hydrogen vehicles in tunnels.

The risks associated with accidents involving hydrogen vehicles inside tunnels depend on several conditions, including whether hydrogen is leaked, the volume of hydrogen being released, the presence of any ignition source, the shape and length of the tunnel, the presence of effective tunnel ventilation and other prevention and mitigation systems, and the properties of the vehicle’s thermal pressure relief device and tank. In a scenario with thermal pressure relief device (TPRD) activation, an immediate ignition poses fewer hazards compared to a delayed ignition. Where hydrogen releases are severe and unignited, the high level of confinement in tunnels may result in overpressures that can maintain their strength for long distances (Venetsanos et al., 2008[50]). Where concentration levels are sufficiently high, they could result in explosions in the presence of an ignition source.

While hydrogen may pose different risks in tunnel environments, a risk assessment by LaFleur et al. (2017) and Ehrhart et al. (2019) shows that the most likely outcome of a FCEV crash inside a tunnel is that there will be no additional hazard due to the hydrogen fuel – with a probability of 98.1-99.9% (see Literature review report) (LaFleur et al., 2017[51]) (Ehrhart et al., 2019[52]). In cases where the hydrogen does ignite, the most likely consequence, with a probability of 0.03-1.8%, is a jet flame from the pressure relief device.

Existing research identifies how exact tunnel conditions can impact the risks of accidents with FCEVs in tunnels, with on-going research expected to shed further light on these consequences. A number of existing studies indicate that the worst-case scenario for a hydrogen-powered bus accident would involve the release of the entire hydrogen volume, followed by ignition when the maximum flammable volume inside the tunnel is achieved, resulting in unacceptably high levels of overpressure (Venetsanos et al., 2008[50]) (Middha and Hansen, 2009[53]). However, more realistic scenarios woud involve lower levels of harm that correspond to the eardrum rupture threshold and moderate building damage (or less). Other research also highlights how the use of protective measures – in particular thermal pressure relief devices, ventilation, or leak detection with safety shutdown – can reduce risks related to hydrogen accidents in tunnels.

The preliminary results of a quantitative risk assessment provide some insights into the comparative risks associated with hydrogen and methane city bus accidents in tunnels, as categorised by different types of accident. In the case of an accident involving a jet fire from a thermal pressure relief device with immediate ignition, the study showed that the incident frequency (events per year) and hazard distances are higher for hydrogen, as compared with methane. Similarly, in the case of a catastrophic tank rupture, the study observed a similar profile for both hydrogen and methane but a higher individual risk and hazard distances for hydrogen.

Further research into hydrogen accidents in tunnels will support a deeper understanding of risks and the safe application of hydrogen vehicles in tunnels. One promising project that has improved the understanding of risks for hydrogen vehicles in tunnels is the HyTunnel-CS project. This project, funded by the Clean Hydrogen Partnership, conducted pre-normative research on the safe use of hydrogen-driven vehicles and transport through tunnels and confined spaces and provides a set of safety recommendations (HyTunnel, 2022[54]).

Annex Box 1.A.4 presents the safety measures that can be considered to decrease the risks related to hydrogen vehicles in tunnels. Risk measures should be decided upon based on desired risk targets while taking into account countervailing risks (see Chapter 1 – Managing risks).

Access to tunnels for vehicles transporting hydrogen in cylinders, tubes, trailers and tank vehicles in the Netherlands is restricted by the ADR regulation (see Vehicles transporting hydrogen). Restrictions on tunnel access are based on the assumption that there are three main hazards that could lead to victims or serious damage in tunnels: explosions, releases of toxic gas or volatile toxic liquid, and fires. As a consequence, vehicles carrying dangerous goods that are expected to pose a higher risk in terms of these three hazards face stronger restrictions (UNECE, 2022[46]). The regulation uses a classification of road tunnels that includes five classes:

Vehicles transporting hydrogen in tanks are allowed to enter tunnels with a class A classification, but cannot enter those tunnels classified as B, C, D or E. In practice, this means that hydrogen in tanks can only be delivered through five tunnels in the Netherlands.23 Such transport is expected to follow the obligatory Hazmat routing in order to avoid water tunnels. Similar restrictions apply to the transport of LPG by road – but, of course, LPG does not present the climate benefits that low-emission hydrogen has (UNECE, 2022[46]).

There is no comprehensive framework regulating the access of hydrogen-powered vehicles to tunnels, as the ADR agreement does not apply to the transport of hydrogen in fuel tanks used to power the vehicle. For this reason, hydrogen-powered vehicles are currently allowed to enter all tunnels in the Netherlands.

The regulation of access to tunnels for vehicles transporting hydrogen in the Netherlands is aligned with other European countries, which also regulate tunnel access based on the ADR regulation. The ADR regulation specifies the restrictions for hydrogen, in compressed and liquid form, and as a mixture with methane. It therefore does not require further amendments to incorporate new hydrogen applications in road transport.

Outside Europe, countries often apply more general national regulations developed for flammable gases to the transport of hydrogen through tunnels. For example, in Japan, vehicles carrying explosive or flammable dangerous goods are prohibited or restricted from entering long tunnels over five kilometres long, as well as underwater and waterfront tunnels.

Similar to the case in the Netherlands, there are, in general, no specific restrictions on hydrogen-powered vehicles entering tunnels in the other countries that were analysed.

There are currently no international standards on the access of vehicles transporting hydrogen and hydrogen-powered vehicles to tunnels. However, for FCEVs specifically, the Global Technical Regulation No. 13 includes specifications for the safe design of vehicles (see Scenario 3 – Road transport), which can reduce the risks involved with using FCEVs in tunnels.

The potential of hydrogen in road transport relies on the availability of an infrastructure to refuel FCEVs. Similar to the case for the rollout of battery-powered electric vehicles, this will require a network of hydrogen refuelling stations at the national and international level to allow sufficient mobility for users of FCEVs.

Hydrogen refuelling stations can operate with liquid hydrogen or compressed (gaseous) hydrogen. Unlike the case for battery-powered electric vehicles, hydrogen refuelling may take around as much time as refuelling with conventional fuels. However, supplying refuelling stations may require more time and labour than for conventional fuels (IEA, 2019[27]).

Although the exact configuration and design of a hydrogen refuelling station may differ depending on the regulations, capacity and type of hydrogen, it may consist of the following components (Haskel, n.d.[55]) (Iberdrola, n.d.[56]):

  • An electrolyser, if hydrogen is produced onsite (see Scenario 1 – Production through water electrolysis).

  • Storage tanks of intermediate pressure.

  • A compressor, to increase the pressure of hydrogen for dispensing.

  • High-pressure buffer storage tanks (in cascade) to store the available hydrogen before dispensing.

  • A cooling system, to remove excess heat from the compression process and cool the hydrogen for dispensing.

  • A hydrogen dispenser to supply the hydrogen to FCEVs.

Investment costs for hydrogen refuelling stations may depend on the pressure and capacity, and the country’s safety and licensing requirements. The two largest cost components are the station’s compressor (up to 60% of total costs) and the storage tanks (IEA, 2019[27]). Required station capacities will depend on the number of FCEVs as well as the types of vehicles being refuelled (passenger vehicle, buses or trucks).

There are strong economies of scale in terms of the capacity of hydrogen refuelling stations. Increasing the capacity of a station from 50 to 500 kilograms of hydrogen per day could cut costs per kilogram by three quarters. For stations with hydrogen at a pressure of 350 bar, investment costs are estimated in the range of 0.15 to 1.6 million USD, whereas at 700 bar investment costs are estimated within the range of USD 0.6 to 2 million. The lower end of these ranges applies for stations with a lower capacity (50 kg of hydrogen per day) and the higher end for stations with a higher capacity (1 300 kg per day) (IEA, 2019[27]).

In 2021, 730 hydrogen refuelling stations worldwide were in operation, to supply a total of 51 600 FCEVs (see Figure 5.5). Japan holds the largest share of the total number of stations (23%), followed by China (20%) and Korea (16%). Between 2020 and 2021, the global number of hydrogen refuelling stations increased by 35% (IEA, 2022[43]).

Within Europe, Germany has the largest network of public hydrogen refuelling stations and the largest number of FCEVs, while deployment in the Netherlands is rising. The total number of hydrogen refuelling stations in Europe by March 2022 was 170. At that date, Germany had 90 public hydrogen refuelling stations in operation, making up 53% of the total number of public stations in Europe. The Netherlands had ten stations in operation, which translates to a 100% increase since 2020 (FCHO, 2022[45]).

Deployment of hydrogen refuelling stations is still far from future goals (see Scenario 3 – Road transport). The potential of FCEVs as a road transport alternative relies on a robust refuelling infrastructure. This therefore creates a need for a rapid scaling up of available refuelling stations to support the envisaged increase in FCEVs worldwide.

Within Europe, the current number of 170 refuelling stations in Europe is still some way off from the international network of refuelling stations as envisaged by the European Commission (European Commission, 2020[16]). The European Commission therefore plans for a strong increase in the number of hydrogen refuelling stations in the EU. At present, hydrogen refuelling stations exist mainly in only a few European member states and are usually not suitable for heavy-duty vehicles such as trucks, thereby limiting the possibility for using hydrogen in heavy-duty transport. To improve this situation, the Commission drafted a proposal for a revised directive on an Alternative Fuels Infrastructure, which would require all publicly accessible stations to serve gaseous hydrogen at 700 bar, with a minimum number of stations also serving liquid hydrogen. The 2014 version of the directive already envisaged one hydrogen refuelling station every 400 kilometres along the Trans-European Transport Network (TEN-T) by 2025. The new proposal outlines plans for one hydrogen station serving compressed hydrogen every 150 kilometres along the network by the end of 2030 and one station that serves liquid hydrogen every 450 kilometres (European Commission, 2021[57]). Furthermore, In March 2023, the European Parliament and the Council reached a political agreement to increase the number of publicly accessible electric recharging and hydrogen refuelling stations. The agreement defines that hydrogen refuelling infrastructures, which can serve both cars and trucks, are to be installed from 2030 in all urban nodes and every 200 km along the core TEN-T network (European Commission, 2023[58]).

The risks of hydrogen refuelling stations depend on a number of factors, including whether production and compression is onsite, the amount of hydrogen stored onsite, the type of hydrogen (liquid or compressed gas), the facility layout, equipment and the population density in the area surrounding the station. Risks are therefore likely to differ between individual refuelling stations, which will affect the need for specific safety measures.

The literature identifies a number of elements that contribute to the particular risks of hydrogen refuelling stations. Pan et al. (2016) identified the compressor as the main risk contributor of all the elements that make up a refuelling station, whereas Khalil (2017) noted that a small leakage from a compressor is associated with unacceptable risks (Pan et al., 2016[59]) (Khalil, 2017[60]).

For hydrogen refuelling stations with onsite production, Tchouvelev et. al showed that production through water electrolysis presents a lower individual and societal risk than production through methane reforming (Dash, Chakraborty and Elangovan, 2023[61]). A comparative risk assessment conducted by (Yoo et al., 2021[62]) indicated that hydrogen refuelling stations that supply liquefied hydrogen have a lower risk than those that supply gaseous (compressed) hydrogen, but with only small differences (Yoo et al., 2021[62]).

Historical data on accidents related to hydrogen refuelling stations show that most accidents at stations have only minor consequences. Data from accident databases include a total number of 25 accidents. A majority (56%) resulted in no release of hydrogen, whereas in another 24% of cases the accident led to an unignited release. Five accidents (or 20% of all accidents) resulted in more serious consequences, where the hydrogen release resulted in a fire or explosion (see Incident database report).

Most incidents at hydrogen refuelling stations are due to equipment failure, especially the malfunction of the dispenser or compressor. The dispenser-related accidents are usually due to flexible hose failures. However, accidents caused by equipment failure often do not result in a hydrogen leak. Other accidents in accident databases have been caused by deficiencies in procedures, design errors, inadequate maintenance or human error by FCEV users. Most of the cases of hydrogen leakage occurred at joint sections in the installations and were due to inadequate torque or sealing. A Japanese study found that hydrogen leakage was often caused by screw joints, highlighting how the use of welded joints may reduce hydrogen leakage (Sakamoto et al., 2016[63]).

A first comparison of accident rates for hydrogen refuelling stations found that, in their current state, these stations may be considered slightly safer than LPG stations. This comparison, using historical incident data, found a normalised accident rate of 1.19 x 10-7 per time of refuelling a hydrogen-powered vehicle (or one in every 8 million times of refuelling) against an accident rate of 2.52 x 10-7 (or one in every 4 million times of refuelling) per time of refuelling an LPG-powered vehicle (see Incident database report). However, it should be noted that the number of hydrogen refuelling stations at the moment is still relatively low, which can affect the accuracy of these estimates based on historical accident data.

A first quantitative risk assessment comparing the risks between hydrogen and compressed natural gas (CNG) found a lower average individual risk for hydrogen refuelling stations than for CNG stations (see summary QRA 5). For hydrogen refuelling stations, risks are expected to be lower for stations with production via electrolyser onsite than those stations with a supply by pipeline or tube trailer. For both gases, the risk was lower for continuous supply via gas pipeline than for discontinuous supply via tube trailer.

Annex Box 1.A.5 presents the safety measures that can be considered to decrease the risks related to hydrogen refuelling stations. Risk measures should be decided upon based on desired risk targets while taking into account countervailing risks (see Chapter 1 – Managing risks).

The regulation of hydrogen refuelling stations in the Netherlands is governed by a number of documents:

  • The Decree Quality Living environment (Besluit Kwaliteit Leefomgeving, Bkl) defines certain safety distances for hydrogen refuelling stations, in particular the “fire attention zone” and “explosion attention zone” (Staatsblad, 2018[64]).

    • The Publication series Dangerous Substances (Publicatiereeks Gevaarlijke Stoffen, PGS) 35 defines the guidelines for the safe use of hydrogen installations supplying gaseous hydrogen at a maximum pressure of 700 bar to vehicles (PGS, 2021[65]). The guidelines can be used by licensing authorities, ODs and inspections as a reference framework (IFV, 2019[66]);

    • The Netherlands norm (Nederlands Norm, NEN-norm) 17124 defines the quality characteristics of gaseous hydrogen fuel dispensed at hydrogen refuelling stations for FCEVs (NEN, 2022[67]).

    • The Act on general provisions environmental law (Wet Algemene bepallingen omgevingsrecht, Wabo) and the Decree environment law (Besluit omgevingsrecht) define the rules regarding the environment permit (Rijksoverheid, 2023[68]).

    • The Spatial planning act defines rules regarding the spatial planning requirements for hydrogen refuelling stations (Rijksoverheid, 2021[69]).

The development of a hydrogen station in the Netherlands requires an environment permit, which brings together one or multiple licenses related to spatial planning, construction and environmental impact (see Chapter 4 – “Licensing”). Hydrogen refuelling stations often do not align with prevailing land use plans, as these plans do not consider the possibility of using hydrogen as a fuel. A quantitative risk assessment will need to be developed as part of this procedure, to inform the licensing process. For a hydrogen filling station, a “Wabo” environmental permit must be applied for in all cases and usually also a building permit.

Existing norms and guidelines can be used as a basis for the licensing process, although these mainly focus on the supply of gaseous hydrogen to FCEVs. The scope of the PGS 35 guideline is limited to those stations supplying gaseous hydrogen at a pressure not higher than 700 bar, although it considers the delivery of hydrogen to these stations in both gaseous and liquid condition. The guideline focuses on occupational, environmental and fire safety aspects for installations and related equipment, and defines potential risks, scenarios and safety measures (PGS, 2021[65]). A similar standard for hydrogen refuelling stations supplying liquid hydrogen is currently lacking. A second PGS guideline (PGS 38) on multifuel stations is currently in consultation. This includes stations supplying both gaseous hydrogen and other fuels, but excludes stations supplying liquid hydrogen (PGS, 2023[70]).

The Decree Quality Living environment defines safety distances for hydrogen refuelling stations, based on an analysis of risk and effect distances for hydrogen refuelling stations by the RIVM. In line with the PGS 35 guideline, this analysis identifies three types of hydrogen delivery to refuelling stations: (1) in gaseous condition via pipeline or local production; (2) in gaseous condition via tube or cylinder trailer and (3) in liquid condition via tank truck. In all the three cases, the hydrogen that is supplied to FCEVs is in gaseous condition. The analysis defines safety distances based a set of risk scenarios (RIVM, 2016[71]).

At the time the decree was written, only gaseous hydrogen was supplied by hydrogen refuelling stations in the Netherlands, and stations supplying liquid hydrogen were not expected in the coming years. Therefore, the decree does not include safety distances for the supply of liquid hydrogen. The decree notes the following safety distances (Staatsblad, 2018[64]):

  • The distances for the local risk24 are set at:

    • 30 metres from the storage unit, for stations where the hydrogen is delivered to the refuelling station via pipeline or is produced onsite.

    • 35 metres from the point of dispensing for stations where the hydrogen is delivered via tank trucks.

  • The distances for the fire attention zone (brandaandachtsgebied)25 are set at 55 metres from the storage unit.

There are significant differences between the countries analysed in the extent to which hydrogen refuelling stations are subject to a comprehensive regulatory framework:

  • In the EU and Australia, in the absence of solid regulatory frameworks for hydrogen refuelling stations, national and international standards or codes are often used as a reference. In other cases, hydrogen refuelling stations are sometimes considered a par with LPG and liquefied natural gas (LNG) stations.

  • Japan, China and the United States have regulations in place for hydrogen stations and their equipment (dispensers, compressors and storage) used for the supply of both compressed and liquid hydrogen.

    • Japan and China have regulations that indicate the technical specifications for materials and equipment They also include prevention and mitigation measures, detailed safety distances from site boundaries and different components of the stations, vulnerable objects, as well as oxygen facilities.

    • The state of California, has developed a comprehensive set of rules for hydrogen refuelling stations, including requirements for dispensing systems and approved equipment (cylinder, containers, tanks, pressure relief devices, hoses, compressors, hydrogen generators, dispensers, detection systems, electrical equipment and others). The NFPA-2 standard defines among other things the separation distances for hydrogen refuelling stations, as well as other fundamental safeguards for the generation, installation, piping, use and handling of hydrogen.

  • Korea developed codes with technical standards for hydrogen refuelling stations.

The international standard ISO 19880-1:2020 covers the technical specifications for public and non-public fuelling stations that supply gaseous hydrogen to light-duty vehicles (but does not apply to stations supplying liquid hydrogen or hydrogen to heavy-duty vehicles). The standard includes the minimum design, installation, commissioning operation, inspection and maintenance requirements for station safety and performance. The standard also applies to:

  • fuelling stations for motorcycles, fork-lift trucks, trams, trains, fluvial and marine applications

  • fuelling stations with indoor dispensing

  • residential applications to fuel land vehicles

  • mobile fuelling stations

  • non-public demonstration fuelling stations (ISO, 2020[72]).”

This scenario looks at the use of hydrogen in residential and other buildings, mainly for heating and cooking purposes. The discussion of this scenario will also consider the use of existing distribution networks to transport hydrogen blends or pure hydrogen to households. The safety and appropriateness of existing household appliances to run on hydrogen is outside the scope of this scenario.

The potential of hydrogen use in buildings depends on several factors, including the existence of a natural gas infrastructure, other energy needs within the building, energy efficiency and safety concerns (IEA, 2019[27]) (IEA, 2021[1]). As the use of natural gas for heating and cooking in buildings is expected to decrease, the use of hydrogen can be an alternative to electricity-based solutions such as heat pumps and electric stoves. This may be especially the case for those situations where heat has to be provided to existing (older) buildings where a gas infrastructure already exists and other green solutions are less feasible. In those cases, local low-emission hydrogen applications could support the decarbonisation of the domestic use of energy. In other cases, the co-existence of hydrogen and other heat technologies can add flexibility in colder climates to cover peak demand if heat pumps cannot meet the heating demand (IEA, 2021[1]).

While hydrogen provides a green alternative to existing heating solutions, its prospects in the heating of buildings at present remain limited to specific contexts. This is in part due to the green alternatives that are already available, in particular photovoltaic (PV) powered heat pumps. These heat pumps operate at a higher efficiency and do not face the same energy losses that result from converting hydrogen. As such, they require five to six times less electricity than a boiler running on hydrogen produced through electrolysis to deliver the same amount of heating (IEA, 2021[1]). Other challenges to the use of hydrogen in buildings relate to safety concerns and consumer acceptance (IEA, 2019[27]). For these reasons, it remains controversial whether low-emission hydrogen will be able to play a significant role in the future of building heating (Weidner and Guillén-Gosálbez, 2023[73]).

Technologies for the use of hydrogen in buildings include hydrogen boilers, fuel cells to co-generate heat and electricity, hybrid heat pumps26 and gas-driven heat pumps27 (IEA, 2021[1]).

At present, the share of hydrogen in the energy mix for residential and other use in buildings is still negligible. In 2020, it was estimated at below 0.005% of total heating energy demand, with many countries piloting its use through demonstration projects (IEA, 2021[1]). These projects look at the injection of hydrogen into gas infrastructure and the use of hydrogen appliances in households, with the ultimate goal of developing hydrogen networks for heating and cooking purposes.

Pilot projects have been reported in countries including China, France, Germany, the Netherlands and the United Kingdom, to assess and demonstrate the safe use of hydrogen in residential and commercial buildings (IEA, 2021[1]). Some notable pilot projects include:

  • Blending hydrogen into gas networks was first piloted on the Dutch island Ameland, where from 2007 to 2011 hydrogen was blended into the existing natural gas network, with injection volumes of up to 20% for heating and cooking with standard appliances (Kiwa, 2012[74]). More recently, injection volumes up to 20% were also demonstrated in the Grid Management by Hydrogen Injection to Decarbonise Energies (Gestion des Réseaux par l’injection d’Hydrogène pour Décarboner les Énergies, GRHYD) project in France and the HyDeploy project in the United Kingdom (IEA, 2021[1]).

  • Other projects to pilot the use of pure hydrogen in existing networks are under development. The delivery of pure hydrogen to 300 households in the United Kingdom through the H100 project, initially planned for 2022, is expected to commence in 2024 (SGN, 2022[75]) (The Guardian, 2022[76]). Projects in the Netherlands that will pilot the delivery of pure hydrogen to households include projects in Rozenburg, Hoogeveen, Stad aan ’t Haringvliet and Wagenborgen (IEA, 2021[1]).

The overall injection of hydrogen into natural gas networks has grown sevenfold between 2013 and 2020, but overall volumes remain low. Almost all blending into natural gas network takes place in Europe, with Germany accounting for 60% of the hydrogen volume blended into natural gas grids (IEA, 2021[1]). The EU JRC estimates that with a 5% blending threshold, 18.4 GW of electrolyser capacity could be integrated across the EU, or 40-70.8 GW with a 20% blending threshold. The maximum amount of annual hydrogen blended into the EU network is estimated at 49.5 TWh at a 5% blending threshold and 220 TWh at a 20% blending threshold (EU JRC, 2022[77]).

Risks and safety measures regarding the use of hydrogen in residential or other buildings distinguish between aspects related to the transport of hydrogen through distribution networks to buildings and those related to the transport and use of hydrogen inside buildings.

A main point of attention when looking at the distribution of hydrogen blends or pure hydrogen through existing low-pressure distribution networks is the impact that hydrogen may have on pipelines. Hydrogen has different properties than the natural gas for which existing networks were initially designed. In particular, there is a risk that hydrogen can lead to embrittlement28 of carbon steel pipelines. This may be valid in particular for the transport of pure hydrogen or blends with a high share of hydrogen through carbon steel pipelines, but less so for blends with a small share of hydrogen or for plastic pipelines (see Part 6 - Lessons learnt and preliminary findings regarding hydrogen safety elements).

The main risk factors that come with the distribution of gas through distribution networks are due to the possibility of hydrogen leakages from the network. The properties of hydrogen, in particular its lower weight as compared with natural gas, mean that leakage volumes may be larger under the same conditions. This information can feed into the design of networks and safety measures, to ensure comparable risk levels for hydrogen as is currently the case for natural gas.

Data on natural gas leak incidents can nevertheless still provide useful indications regarding risk factors as well as mitigating measures. Data on past gas leak incidents in the United Kingdom show that most leaks occur in the connecting pipe, at the gas meter or the indoor piping, especially where network components are made of materials such as grey and ductile iron, asbestos cement and steel (Van den Noort et al., 2020[78]). Findings from other research indicate that a switch from the current composition of the UK network of 74% polyethylene and 26% metal parts to a 100% polyethylene network could reduce flammable gas leakages by a factor 2.5, and “gas in building” incidents by a factor 3.5, for both natural gas and hydrogen (Mouli-Castillo, 2021[79]).

The risks related to hydrogen leakages from distribution networks differ in their impact depending on the level of hydrogen concentration and whether the leakage occurs in open air or underground (see also Scenario 2 – ). Overall, the risks of natural gas and hydrogen releases from distribution networks are comparable in the case of free flow in open air (Van den Noort et al., 2020[78]). Risks associated with hydrogen releases underground depend on the soil type and its permeability.

Similar to the case of natural gas inside buildings, the risks related to the use of hydrogen inside buildings depend on the possibility of hydrogen concentrations building up inside the house, which could potentially ignite. Research in the United Kingdom found that hydrogen (meeting quality standard ISO 14687 Type A) was compatible with all the domestic gas fittings and pipes tested, where components that displayed no leakages with natural gas also showed no leakages with hydrogen (Ryan and Roberts, 2020[80]). Leakage volumes from damaged components were, however, larger for hydrogen than for natural gas. Hydrogen can more easily ignite, but at the same time it has a lower energy content per volume and dissipates more quickly due to its lower weight. The impact is therefore likely to differ depending on the presence of effective ventilation at the location of a leak.

Experiments with the use of hydrogen blends of up to 20% in natural gas showed that these are likely to result in only small increases in overpressure in the event of a leakage compared with natural gas (by a factor 1.2) (Lowesmith et al., 2011[81]). Other experiments in the HyHouse found that, for the scenarios considered, the associated potential to cause severe structural damage was comparable for hydrogen and natural gas. A study to assess the likelihood of household items causing the ignition of hydrogen found that only a few of the domestic appliances – that did not cause natural gas to ignite – caused hydrogen to ignite (Crewe, Johnson and Allason, 2020[82]). Nearly all these items require a human operator present, who would most likely smell a gas release provided an odorant is added to the gas.

Annex Box 1.A.6 presents the safety measures that can be considered to decrease the risks related to domestic use of hydrogen. Risk measures should be decided upon based on desired risk targets while taking into account countervailing risks (see Chapter 1 – Managing risks).

The existing legal framework in the Netherlands related to the distribution of hydrogen to households is expected to change with the new Energy Act (see Scenario 2 – ). The act specifies the activities related to hydrogen that network operators are allowed to engage in, which would include the operation and maintenance of hydrogen networks.

Transmission and distribution system operators are currently not allowed to blend any amount of hydrogen into the natural gas infrastructure, although this is expected to change with the proposed Energy Act (EZK, 2021[39]). Within the proposed new law, system operators are required to accept gases other than natural gas (including hydrogen) on their network. This is contingent on them being able to reasonably blend the additional gas into their system and maintain the quality of gas delivered (in line with quality criteria set out in a ministerial regulation). Moreover, hydrogen transportation through newly constructed hydrogen pipelines is regulated under the Decree External Safety Pipelines.

While there is currently no regulatory framework that determines the conditions for the supply of hydrogen to consumers or the safe use of appliances inside buildings, the ACM has developed a framework to facilitate pilot projects (ACM, 2022[83]). This framework will allow network operators and energy retailers that adopt adequate safety measurs to already test and gain experience with domestic hydrogen applications before new legislation is expected to come in place. The ministry has appointed the SodM as the body to supervise the safety of these pilots (SodM, 2022[84]).

In other countries, there is usually no or only limited regulation regarding the distribution and domestic use of hydrogen:

  • In Australia, the government is conducting a review of the volume of hydrogen that can be blended into gas networks. There is no regulation allowing pure hydrogen as gas appliances are only suitable for a blend of up to 10% or 20%.

  • China’s policies and regulations support hydrogen blending in existing natural gas grids and the government has published a group of standards for natural gas and hydrogen mixing stations. It is currently completing a review on how to bring hydrogen into the gas network.

  • In Japan and South Korea, the domestic use of hydrogen involves fuel cell systems, which are subject to regulations that apply to fuel cells in general.

  • In the United Kingdom, in the absence of hydrogen related rules and regulations, the concentration of hydrogen that can be injected into the gas network and consequently be supplied to domestic homes should be no greater than 0.1% molar.29

  • In the United States, there are no regulations specifically targeting the domestic use of hydrogen, although such use is not prohibited as can be seen from the existence of small-scale pilot projects.

As the application of hydrogen for domestic use is currently still at the stage of piloting, there is no international standard that applies to this scenario.


[83] ACM (2022), ACM develops framework to enable pilot projects with hydrogen [ACM stelt kader op om pilotprojecten met waterstof mogelijk te maken], https://www.acm.nl/nl/publicaties/acm-stelt-kader-op-om-pilotprojecten-met-waterstof-mogelijk-te-maken (accessed on 15 October 2022).

[87] Air Liquide (n.d.), Storing hydrogen, https://energies.airliquide.com/resources-planet-hydrogen/how-hydrogen-stored#:~:text=A%20volume%20of%20around%2011,storage%20in%20the%20gaseous%20form (accessed on 28 October 2022).

[12] BP (2021), BP plans major green hydrogen project in Teesside, https://www.bp.com/content/dam/bp/business-sites/en/global/corporate/pdfs/news-and-insights/press-releases/bp-plans-major-green-hydrogen-project-in-teesside.pdf (accessed on 19 October 2022).

[19] Brook, B. et al. (2014), “Why nuclear energy is sustainable and has to be part of the energy mix.”, Sustainable Materials and Technologies, Vol. 1, pp. 8-16, https://www.sciencedirect.com/science/article/pii/S2214993714000050.

[82] Crewe, R., M. Johnson and D. Allason (2020), Ignition Potential Testing with Hydrogen and Methane, DNV GL.

[61] Dash, S., S. Chakraborty and D. Elangovan (2023), “A Brief Review of Hydrogen Production Methods and Their Challenges”, Energies, p. 16.

[32] DNV (2017), Verkenning Waterstofinfrastructuur, https://www.topsectorenergie.nl/sites/default/files/uploads/TKI%20Gas/publicaties/DNVGL%20rapport%20verkenning%20waterstofinfrastructuur_rev2.pdf (accessed on 25 October 2022).

[31] EHB (2022), The European Hydrogen Backbone (EHB) initiative, https://ehb.eu/#mission (accessed on 25 October 2022).

[52] Ehrhart, B. et al. (2019), “Ehrhart, B., Brooks, D., & Muna, A., Lafleur, C. (2019). Risk assessment of hydrogen fuel cell electric vehicles in tunnels.”, Fire Technology, Vol. 56.

[34] EIGA (2004), Hydrogen Transportation Pipelines, https://h2tools.org/sites/default/files/Doc121_04%20H2TransportationPipelines.pdf (accessed on 7 December 2022).

[77] EU JRC (2022), Blending hydrogen from electrolysis into the European gas grid, https://publications.jrc.ec.europa.eu/repository/handle/JRC126763 (accessed on 2 June 2023).

[58] European Commission (2023), European Green Deal: ambitious new law agreed to deploy sufficient alternative fuels infrastructure, https://ec.europa.eu/commission/presscorner/detail/en/ip_23_1867 (accessed on 2 June 2023).

[38] European Commission (2021), Proposal for a DIRECTIVE OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL on common rules for the internal markets in renewable and natural gases and in hydrogen, https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=COM:2021:803:FIN (accessed on 26 October 2022).

[57] European Commission (2021), Proposal for a Regulation of the European Parliament and of the Council on the deployment of alternative fuels infrastructure, and repealing Directive 2014/94/EU of the European Parliament and of the Council, https://eur-lex.europa.eu/legal-content/en/TXT/?uri=CELEX%3A52021PC0559 (accessed on 7 February 2023).

[16] European Commission (2020), A hydrogen strategy for a climate-neutral Europe, https://ec.europa.eu/energy/sites/ener/files/hydrogen_strategy.pdf (accessed on 29 September 2022).

[36] EZK (2021), Explanatory memorandum Energy act [Memorie van toelichting wetsvoorstel Energiewet], https://open.overheid.nl/repository/ronl-295d9f68-a553-4e4c-b9ec-cfa6ab498baa/1/pdf/2a.%20MvT%20Energiewet%20_algemeen%20deel_%20versie%20UHT%20d.d.%2017%20nov%202021.pdf (accessed on 26 October 2022).

[39] EZK (2021), Proposal Energy act - version UHT (17 November 2021) [Wetsvoorstel Energiewet - versie UHT (17 november 2021)], https://open.overheid.nl/repository/ronl-fbad566b-5bfc-499c-9aad-25b433fc1fee/1/pdf/1.%20Wetsvoorstel%20Energiewet%20versie%20UHT%20d.d.%2017%20nov%202021b.pdf (accessed on 26 October 2022).

[23] FCH 2 JU (2020), Minutes of the FCH2JU workshop of electroysis on 18 November 2020.

[45] FCHO (2022), Chapter 1 - Technology and Market, https://www.fchobservatory.eu/sites/default/files/reports/Chapter%201%20-%20Technology%20and%20Market%20-%202022%20Final%20Revised%2007.2022.pdf (accessed on 28 October 2022).

[14] FCHO (2022), Chapter 2 - 2022 Hydrogen Supply Capacity and Demand, https://www.fchobservatory.eu/sites/default/files/reports/Chapter%202%20-%20FCHO%20Market%20-%202022%20Final.pdf (accessed on 28 October 2022).

[24] Glover, A., A. Baird and D. Brooks (2020), Final Report on Hydrogen Plant Hazards and Risk Analysis Supporting Hydrogen Plant Siting near Nuclear Power Plants, https://www.osti.gov/servlets/purl/1678837.

[48] H2Platform (2020), Hydrogen in parking garages [Waterstof in parkeergarages], https://opwegmetwaterstof.nl/waterstof-in-parkeergarages/ (accessed on 28 October 2022).

[8] Hancke, R., T. Holm and Ø. Ulleberg (2022), “The case for high-pressure PEM water electrolysis”, Energy Conversion and Management, Vol. 261, p. 115642, https://doi.org/10.1016/j.enconman.2022.115642.

[55] Haskel (n.d.), How Does a Hydrogen Refuelling Station Work?, https://www.haskel.com/en-fr/blog/how-does-a-hydrogen-refuelling-station-work#:~:text=A%20regular%20HRS%20consists%20of,type%20of%20vehicle%20being%20refuelled. (accessed on 7 February 2023).

[5] Hu, K. et al. (2022), “Comparative study of alkaline water electrolysis, proton exchange membrane water electrolysis and solid oxide electrolysis through multiphysics modeling”, Applied Energy, Vol. 312, p. 118788, https://doi.org/10.1016/j.apenergy.2022.118788.

[25] HyLAW (2018), HyLAW National Policy Paper - Netherlands, https://www.hylaw.eu/sites/default/files/2019-03/HyLAW_National%20Policy%20Paper_Netherlands.pdf (accessed on 15 October 2022).

[54] HyTunnel (2022), Deliverable D6.9, https://hytunnel.net/wordpress/wp-content/uploads/2022/09/HyTunnel-CS_D6.9_Recommendations-for-safer-use-HV.pdf (accessed on 1 March 2023).

[29] HyWay27 (2021), HyWay 27: realisation of a national hydrogen network, https://www.hyway27.nl/en/latest-news/hyway-27-realisation-of-a-national-hydrogen-network (accessed on 19 October 2022).

[56] Iberdrola (n.d.), What are hydrogen stations and how do they work?, https://www.iberdrola.com/sustainability/hydrogen-stations (accessed on 7 February 2023).

[85] IEA (2023), Towards hydrogen definitions based on their emissions intensity, IEA, https://www.iea.org/reports/towards-hydrogen-definitions-based-on-their-emissions-intensity.

[7] IEA (2022), Electrolysers - Tracking report September 2022, https://www.iea.org/reports/electrolysers (accessed on 16 October 2022).

[44] IEA (2022), Fuel cell electric vehicle stock and hydrogen refuelling stations by region, 2021, https://www.iea.org/data-and-statistics/charts/fuel-cell-electric-vehicle-stock-and-hydrogen-refuelling-stations-by-region-2021 (accessed on 4 October 2022).

[43] IEA (2022), Global EV Outlook 2022, https://www.iea.org/reports/global-ev-outlook-2022 (accessed on 7 Februrary 2023).

[10] IEA (2022), Global Hydrogen Review 2022, OECD Publishing, Paris, https://doi.org/10.1787/a15b8442-en.

[1] IEA (2021), Global Hydrogen Review 2021, OECD Publishing, Paris, https://doi.org/10.1787/39351842-en.

[30] IEA (2021), Hydrogen - Tracking report November 2021, https://www.iea.org/reports/hydrogen (accessed on 30 September 2022).

[27] IEA (2019), The Future of Hydrogen: Seizing today’s opportunities, OECD, Paris Cedex 16, https://doi.org/10.1787/1e0514c4-en.

[66] IFV (2019), Administrative guideline on license issuance for hydrogen refuelling stations [Bestuurlijke handreiking vergunningverlening waterstoftankstations], https://opwegmetwaterstof.nl/wp-content/uploads/2020/03/IFV-Bestuurlijke-handreiking-vergunningverlening-waterstoftankstations.pdf (accessed on 15 October 2022).

[4] IRENA (2018), Hydrogen from renewable power: Technology outlook for the energy transition, https://www.irena.org/publications/2018/sep/hydrogen-from-renewable-power.

[72] ISO (2020), ISO 19880-1:2020, https://www.iso.org/standard/71940.html (accessed on 10 February 2023).

[49] ISO (2015), ISO 11623:2015(en), https://www.iso.org/obp/ui/#iso:std:iso:11623:ed-2:v2:en (accessed on 22 March 2023).

[20] Kasai, N. et al. (2016), “The qualitative risk assessment of an electrolytic hydrogen generation system”, International Journal of Hydrogen Energy, Vol. 41/30, pp. 13308-13314.

[60] Khalil, Y. (2017), “A probabilistic visual-flowcharting-based model for consequence assessment of fire and explosion events involving leaks of flammable gases”, Journal of Loss Prevention in the Process Industries, Vol. 50(A), pp. 190-204.

[74] Kiwa (2012), Hydrogen in natural gas on Ameland [Waterstof in aardgas op Ameland], https://www.netbeheernederland.nl/_upload/Files/Waterstof_56_d2bb410cce.pdf (accessed on 14 February 2023).

[3] Krishnan, S. et al. (2020), Power to gas (H2): alkaline electrolysis, Academic Press.

[51] LaFleur, C. et al. (2017), Hydrogen fuel cell electric vehicle tunnel safety study.

[26] Lexparency (2008), Directive 2012/18/EU of the European Parliament and of the Council of 4 July 2012, https://lexparency.org/eu/32012L0018/ (accessed on 7 December 2022).

[81] Lowesmith, B. et al. (2011), “Vented confined explosions involving methane/hydrogen mixtures”, Int. J. Hydrogen Energy, Vol. 36, pp. 2337–2343.

[53] Middha, P. and O. Hansen (2009), “CFD simulation study to investigate the risk from hydrogen vehicles in tunnels”, International Journal of Hydrogen Energy, Vol. 34/14, pp. 5875–5886.

[79] Mouli-Castillo, J. (2021), “A quantitative risk assessment of a domestic property connected to a hydrogen distribution network”, International Journal of Hydrogen Energy, Vol. 46/29, pp. 16217–16231, https://doi.org/10.1016/j.ijhydene.2021.02.114.

[67] NEN (2022), Waterstof brandstof - Product specificatie en kwaliteitsgarantie - Proton Exchange Membrane (PEM) brandstofcel toepassingen voor wegvoertuigen, https://www.nen.nl/nen-en-17124-2022-en-294449 (accessed on 15 October 2022).

[33] Netbeheer Nederland (2018), Toekomstbestendige gasdistributienetten, https://www.netbeheernederland.nl/_upload/RadFiles/New/Documents/Kiwa%20-Toekomstbestendige%20gasdistributienetten%20-%20GT170272%20-%202018-07-05%20-D.._.pdf (accessed on 25 October 2022).

[22] Pan, X. et al. (2016), “Safety study of a wind–solar hybrid renewable hydrogen refuelling station in China”, International Journal of Hydrogen Energy, Vol. 41/30, pp. 13315-13321.

[59] Pan, X. et al. (2016), “Safety study of a wind–solar hybrid renewable hydrogen refuelling station in China”, International Journal of Hydrogen Energy, Vol. 41/30, pp. 13315-13321.

[28] Perey, P. and M. Mulder (2023), “International competitiveness of low-carbon hydrogen supply to the Northwest European market”, International Journal of Hydrogen Energy, Vol. 48/4, pp. 1241-1254, https://doi.org/10.1016/j.ijhydene.2022.10.011.

[70] PGS (2023), 38 Multi-energie stations, https://publicatiereeksgevaarlijkestoffen.nl/publicaties/online/pgs-38/2022/0-1-januari-2023#top (accessed on 21 February 2023).

[65] PGS (2021), Hydrogen applications to supply hydrogen to vehicles and equipment [Waterstofinstallaties voor het afleveren van waterstof aan voertuigen en werktuigen], https://publicatiereeksgevaarlijkestoffen.nl/publicaties/online/pgs-35/2021/1-0-augustus-2021#top (accessed on 15 October 2022).

[17] Porthos (2022), Project, https://www.porthosco2.nl/en/project/ (accessed on 19 October 2022).

[15] Revolve (2022), Hydrogen Outlook, https://hydrogen.revolve.media/2022/features/hydrogen-outlook/#:~:text=While%20in%202019%2C%20the%20largest,installation%20came%20online%20in%202022 (accessed on 2 June 2023).

[35] Rigas, F. and P. Amyotte (2013), Myths and facts about hydrogen hazards.

[68] Rijksoverheid (2023), Act on general provisions environment law [Wet algemene bepalingen omgevingsrecht], https://wetten.overheid.nl/BWBR0024779/2023-04-19/ (accessed on 2 June 2023).

[18] Rijksoverheid (2022), Seven large hydrogen projects in the Netherlands receive subsidy for electrolysis [Zeven grote waterstofprojecten in Nederland krijgen subsidie voor elektrolyse], https://www.rijksoverheid.nl/actueel/nieuws/2022/12/20/zeven-grote-waterstofprojecten-in-nederland-krijgen-subsidie-voor-elektrolyse (accessed on 10 February 2023).

[69] Rijksoverheid (2021), Spatial planning act [Wet ruimtelijke ordening], https://wetten.overheid.nl/BWBR0020449/2021-07-01 (accessed on 2 June 2023).

[47] Rijksoverheid (2015), Law transport dangerous substances [Wet vervoer gevaarlijke stoffen], https://wetten.overheid.nl/BWBR0007606/2015-04-01/#HoofdstukIIIa_Artikel28 (accessed on 8 November 2022).

[37] RIVM (2021), Rekenmethodiek transport waterstof in (aardgas)transportleidingen, https://www.rivm.nl/sites/default/files/2023-02/20211125%20Advies%20rekenmethodiek%20transport%20waterstof%20in%20%28aardgas%29transportleidingen.pdf (accessed on 1 March 2023).

[71] RIVM (2016), Memo Risico- en effectafstanden waterstoftankstations, https://www.rivm.nl/sites/default/files/2021-11/20170711%20Memo%20Risico-%20en%20effectafstanden%20waterstoftankstations_revisie%201.pdf (accessed on 8 February 2023).

[80] Ryan, N. and S. Roberts (2020), “Hy4Heat Work Package 7 – Lot 1: Safety Assessments for the Suitability of Hydrogen in Existing Buildings”, Steer Energy.

[63] Sakamoto, J. et al. (2016), “Leakage-type-based analysis of accidents involving hydrogen fueling stations in Japan and USA”, International Journal of Hydrogen Energy, Vol. 41/46.

[21] Schefer, R. et al. (2009), “Experimental evaluation of barrier walls for risk reduction of unintended hydrogen release”, International Journal of Hydrogen Energy, Vol. 34, pp. 1590-1606.

[6] Sebbahi, S. et al. (2022), “Assessment of the three most developed water electrolysis technologies: Alkaline Water Electrolysis, Proton Exchange Membrane and Solid-Oxide Electrolysis”, Materials Today: Proceedings, Vol. 66, pp. 140-145, https://doi.org/10.1016/j.matpr.2022.04.264.

[75] SGN (2022), H100 Fife, https://www.sgn.co.uk/H100Fife (accessed on 14 February 2023).

[86] Shell (2017), Shell Hydrogen Study: Energy of the Future?, https://epub.wupperinst.org/frontdoor/deliver/index/docId/6786/file/6786_Hydrogen_Study.pdf (accessed on 25 October 2022).

[84] SodM (2022), Supervision arrangement hydrogen distribution pilots [Toezichtarrangement pilots Waterstofdistributie], https://www.sodm.nl/documenten/rapporten/2022/11/01/toezichtarrangement-pilots-waterstofdistributie (accessed on 1 December 2022).

[64] Staatsblad (2018), Decree Quality Living environment [Besluit Kwaliteit Leefomgeving], https://zoek.officielebekendmakingen.nl/stb-2018-292.html (accessed on 15 October 2022).

[11] Sunfire (2023), RENEWABLE HYDROGEN PROJECT “MULTIPLHY”: WORLD’S LARGEST HIGH-TEMPERATURE ELECTROLYZER FROM SUNFIRE SUCCESSFULLY INSTALLED, https://www.sunfire.de/en/news/detail/renewable-hydrogen-project-multiplhy-worlds-largest-high-temperature-electrolyzer-from-sunfire-successfully-installed (accessed on 22 May 2023).

[76] The Guardian (2022), ‘World-first’ hydrogen project raises questions about its role in fuelling future homes, https://www.theguardian.com/environment/2022/sep/20/world-first-hydrogen-project-raises-questions-about-its-role-in-fuelling-future-homes (accessed on 14 February 2023).

[2] Tian, Z. et al. (2021), “Review on equipment configuration and operation process optimization of hydrogen refueling station”, International Journal of Hydrogen Energy.

[46] UNECE (2022), ADR - Volume I, https://unece.org/sites/default/files/2023-01/ADR2023_Vol1e.pdf (accessed on 3 February 2023).

[13] Upstream (2022), Baofeng Energy brings world’s largest green hydrogen project on line in China, https://www.upstreamonline.com/hydrogen/baofeng-energy-brings-worlds-largest-green-hydrogen-project-on-line-in-china/2-1-1161221 (accessed on 19 October 2022).

[42] US Department of Energy (n.d.), Fuel Cell Electric Vehicles, https://afdc.energy.gov/vehicles/fuel_cell.html (accessed on 18 October 2022).

[41] US Department of Energy (n.d.), How Do Fuel Cell Electric Vehicles Work Using Hydrogen?, https://afdc.energy.gov/vehicles/how-do-fuel-cell-electric-cars-work (accessed on 18 October 2022).

[40] US Department of Energy (n.d.), Liquid Hydrogen Delivery, https://www.energy.gov/eere/fuelcells/liquid-hydrogen-delivery (accessed on 28 October 2022).

[78] Van den Noort, A. et al. (2020), Gedrag van waterstof bij lekkages in het gasdistributienet, https://www.netbeheernederland.nl/_upload/Files/Waterstof_56_770e4ce970.pdf.

[50] Venetsanos, A. et al. (2008), “CFD modelling of hydrogen release, dispersion and combustion for automotive scenarios”, Journal of Loss Preventation in the Process Industries, Vol. 21/2, pp. 162-184.

[73] Weidner, T. and G. Guillén-Gosálbez (2023), “Planetary boundaries assessment of deep decarbonisation options for building heating in the European Union”, Energy Conversion and Management, Vol. 278, p. 116602, https://doi.org/10.1016/j.enconman.2022.116602.

[62] Yoo, B. et al. (2021), “Comparative risk assessment of liquefied and gaseous hydrogen refuelling stations”, International Journal of Hydrogen Energy, Vol. 46/71, pp. 35511-35524.

[9] Zarei, E., F. Khan and M. Yazdi (2021), “A dynamic risk model to analyze hydrogen infrastructure”, International Journal of Hydrogen Energy, Vol. 46/5, pp. 4626-4643.


← 1. This share is significantly lower in 2021, when a new 150 MW electrolyser become operational in China.

← 2. Currently, only 4% of projects (in terms of production output in 2030) are at advanced stages of development (under construction or with final investment decision), whereas one third is at the concept stage and the remaining projects are undergoing feasibility and engineering studies (IEA, 2023[85]).

← 3. Normalisation allows for a comparison of fatality risk by adjusting for the total production volume of energy sources, resulting in a fatality risk per energy volume (such as TWh).

← 4. Based on data from three incident databases (Energy-related Severe Accident Database (ENSAD), Hydrogen Incident and Accident Database (HIAD) 2.0 and H2tools).

← 5. Directives 2011/92/EU on the environmental impact assessment (EIA), Directive 2001/42/EC on the strategic environmental assessment (SEA), Directive 2010/75/EU on industrial emissions (IED) and Directive 2012/18/EU on the control of major hazards involving dangerous substances (Seveso Directive).

← 6. Requirements following from the Seveso Directive may differ between so-called “lower-tier” establishments (with quantities above 5 tonnes) and “upper-tier” establishments (with quantities above 50 tonnes) (Lexparency, 2008[26]).

← 7. Ksi stands for kilo pounds per square inch and measures the amount of stress a material such as steel can undergo before failing.

← 8. Retrofitting is the upgrade of existing gas infrastructure to allow the injection of certain amounts of hydrogen as a blend, whereas repurposing involves the conversion of existing infrastructure to a dedicated hydrogen infrastructure.

← 9. Similarly, the European Hydrogen Backbone (EHB) study estimates that conversion costs are 21 to 33% of the cost of new hydrogen pipelines.

← 10. Looking at hydrogen shares by country, in 2017 57% of worldwide hydrogen pipelines were in the United States (2 608 km), 13% in Belgium (613 km), 8% in Germany (376 km), 7% in France (303 km), 5% in the Netherlands (237 km), 3% in Canada (147 km) and 6% in other countries (258 km). In total, hydrogen pipelines amounted to 4542 kilometre in 2017 (Shell, 2017[86]).

← 11. Where hydrogen leakages occur, the corresponding drop in pressure should usually activate installed protection systems such as the automatic closing of safety valves to limit the quantity of release. The magnitude of the consequences of a hydrogen incident will therefore depend on the successful operation of this safety system.

← 12. The incident rate (number of leakage incidents per year) was normalised per 1 000 km pipeline.

← 13. In other cases, a system of negotiated third-party access applies until the implementation of regulated third party access in 2030.

← 14. The decree applies to pipelines for flammable substances with an external diameter of at least 70 millimetres or an internal diameter of at least 50 millimetres and a pressure of 1 600 kPA or higher.

← 15. The local risk is the risk of a fatal accident due to a pipeline incident for a person who is continuously exposed and unprotected at a given location.

← 16. The Minister of EZK can decide in a ministerial regulation to adjust this distance for certain category of pipelines or to accept a different risk level.

← 17. The group risk is defined as the cumulative risk per year and per kilometre that at least 10, 100 or 1 000 persons die as a direct consequence of their proximity to a pipeline that is experiencing an incident.

← 18. While the main focus of the scenario is on vehicles transporting hydrogen, a significant part of the literature on safety risks due to releases in transport concerns hydrogen-powered vehicles. For this reason, both will be considered under this scenario.

← 19. Hydrogen in liquid state is a so-called cryogenic liquid, referring to liquids with a boiling point at extremely low temperatures. Therefore, to transport hydrogen as a liquid rather than a gas, it needs to be cooled to low temperatures.

← 20. Due to its low weight and energy content by volume, unpressurised gaseous hydrogen would require a volume of 11 m3 of hydrogen for 1 kg of hydrogen, roughly needed to drive 100 km by car. By compressing the gas, the energy value per volume increases, allowing hydrogen cars to drive further on a single tank (Air Liquide, n.d.[87]).

← 21. This conversion into electricity and water requires the combination of hydrogen and oxygen from the air.

← 22. For a significant share of incidents (28%), the cause was unknown.

← 23. These tunnels include the Roertunnel, the Schipholtunnel, the Swalmentunnel, the Leidsche Rijntunnel and the Willem-Alexandertunnel.

← 24. The local risk is the risk of a fatal accident due to a pipeline incident for a person who is continuously exposed and unprotected at a given location. The local risk distance is set at such a distance to ensure the local risk for vulnerable buildings and locations does not exceed a threshold of 10-6 (or one in a million) per year (Staatsblad, 2018[64]).

← 25. A fire safety distance limits the zone beyond which the impact of an unusual incident that causes a puddle or flare fire does not lead to a heat radiation higher than 10 kilowatts per square metre (Staatsblad, 2018[64]).

← 26. Hybrid heat pumps use a hydrogen boiler as a supplement to an electric heat pump to meet peak demand.

← 27. Gas-driven heat pumps use a gas engine that produces electricity to run a heat pump.

← 28. Embrittlement is a significant decrease of ductility of a metal, which makes the material brittle.

← 29. 0.1% molar indicates the percentage of moles (or units) of hydrogen as a share of total number of moles in the mixture.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2023

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at https://www.oecd.org/termsandconditions.