Tno 2020 R12005

Princetonlaan 6
3584 CB Utrecht
P.O. Box 80015
3508 TA Utrecht
TNO report The Netherlands
www.tno.nl
TNO2020 R12005
T +31 88 866 42 56
Inventory of risks associated with underground F +31 88 866 44 75
storage of compressed air (CAES) and hydrogen

(UHS), and qualitative comparison of risks of UHS
vs. underground storage of natural gas (UGS)
Date 30 August 2020
Author(s) Kaj van der Valk, Marianne van Unen, Logan Brunner, Remco
Groenenberg.
Copy no
No. of copies
Number of pages 47 (incl. appendices)
Number of appendices 2
Sponsor NAM, Gasunie, Gasterra, Nouryon, EBN, Rijksdienst voor
Ondernemend Nederland (RVO)
Project name Large-Scale Energy Storage in Salt Caverns and Depleted Gas
Fields (Acronym: LSES)
Project number 060.36821, subsidy reference: TGEO118002
All rights reserved.

No part of this publication may be reproduced and/or published by print, photoprint, microfilm
or any other means without the previous written consent of TNO.
In case this report was drafted on instructions, the rights and obligations of contracting parties
are subject to either the General Terms and Conditions for commissions to TNO, or the
relevant agreement concluded between the contracting parties. Submitting the report for
inspection to parties who have a direct interest is permitted.
© 2020 TNO
TNO report | TNO 2020 R12005 | Project LSES - WP4 2 / 47
Preface
This report details the results of the activities performed in work package 4 of the
research project “Large-Scale Energy Storage in Salt Caverns and Depleted Gas
Fields”, abbreviated as LSES. The project, which was given subsidy by RVO, had
two main goals:
1. Improve insights into the role that large-scale subsurface energy storage
options can play in providing flexibility to the current and future transitioning
energy system;
2. Address techno-economic challenges, identify societal and regulatory barriers
to deployment, and assess risks associated with selected large-scale
subsurface energy storage technologies, in particular compressed-air energy
storage (CAES) and Underground Hydrogen Storage (UHS).
The research was carried out by TNO in close collaboration with project partners
EBN, Gasunie, Gasterra, NAM and Nouryon. Activities were divided over 4 work
packages that ran in parallel:
1. Analysis of the role of large-scale storage in the future energy system: what
will be the demand for large-scale storage, when in time will it arise, and where
geographically in our energy system will it be needed?
2. Techno-economic modelling (performance, cost, economics) of large-scale
energy storage systems, focusing in CAES and UHS in salt caverns, and UHS
in depleted gasfields - analogous to UGS (Underground natural Gas Storage).
3. Assessment of the current policy and regulatory frameworks and how they limit
or support the deployment of large-scale energy storage, and stakeholder
perception regarding energy storage.
4. Risk identification and screening for the selected large-scale subsurface
energy storage technologies.
In this report, the results of the activities performed in work package 4 on risks
associated with CAES and UHS are detailed.
The results of the other work packages are detailed in three other reports.
Project details
Subsidy reference: TGEO118002
Project name: Large-Scale Energy Storage in Salt Caverns and
Depleted Gas Fields
Project period: April 16, 2019 until August 30, 2020
Project participants: TNO (executive organization), EBN, Gasunie, Gasterra,
NAM and Nouryon
Acknowledgements
All project partners (EBN, NAM, Gasunie, GasTerra, and Nouryon) are thanked for
their reviews of the final draft of this report.
Het project is uitgevoerd met subsidie van het Ministerie van Economische Zaken en
Klimaat, Nationale regelingen EZ-subsidies, Topsector Energie uitgevoerd door
Rijksdienst voor Ondernemend Nederland.
Summary
Energy storage can play a pivotal role in the energy transition by adding flexibility to
the sustainable energy system. Large-scale storage of energy underground, in salt
caverns, depleted gas fields and (potentially) aquifers, is an attractive option to store
large amounts of energy, and can help to secure supply in prolonged periods of
several days to more than a week with calm winds and no or little sunshine.
However, the use of the subsurface for energy storage may introduce risks that can
negatively impact health, safety and environment, system integrity, economics and
the public perception towards this technology. The risks associated with Underground
natural Gas Storage (UGS) in the subsurface are well-known from decades of
experience. However, the risks associated with Underground Hydrogen Storage
(UHS) and Compressed Air Energy Storage (CAES) are relatively underexplored.
In this study the potential risks associated with UHS and CAES in salt caverns, and
UHS in depleted gas fields (porous reservoirs) were inventoried, and possible
mitigation measures were explored. Risks were inventoried by conducting a literature
review, and supplemented with expert knowledge. All risks were included in a risk
inventory that categorizes the risks into their relevant project phase, system
component, reservoir storage type and TEECOPS1 category. In total, 159 risks were
derived from 40 references, of which about half (75) pertain to operating the storage
facility. The purpose of the risk inventory is to serve as a starting point and checklist
to identify and manage risks in development projects, and to provide guidance on
potential mitigation measures to reduce the risks.
In order to improve our understanding of the significance of the risks associated with
underground hydrogen storage (UHS), a selection of six key risk themes associated
with storage of hydrogen was made: material integrity/durability, leakage of
hydrogen, blow-out, diffusion and dissolution, loss and/or contamination of hydrogen,
and ground motion (subsidence, induced seismicity). A qualitative non site-specific
comparison was made for these risk themes between UHS and underground storage
of natural gas (UGS, with methane as a proxy for natural gas), primarily based on
differences in gas properties. Overall aim of this comparison was to leverage the
experience from UGS to provide useful information to better understand and reduce
risks and consequences, increase control and inform stakeholders. Although in
general, UGS and UHS have a similar risk profile, there are also differences that were
highlighted in this study:
• Hydrogen has a much wider flammability range and a much lower ignition energy
compared to methane, and is therefore more prone to ignite when released in air.
Hydrogen is therefore classified as a high reactive 2 gas, while methane is
classified as a low reactive gas. On ignition methane radiates heat and creates a
flame that is clearly visible. Ignited hydrogen on the other hand radiates little
(infrared) heat (IR), but emits substantial UV (ultraviolet) radiation. The lack of IR
gives little sensation of heat but the exposure to a hydrogen flame still causes
severe burns because of the UV radiation. Because a burning hydrogen flame is
1
TEECOPS: technical, economic, environmental, commercial, organisational, political and societal
2
“Reactive” here refers to the ability to ignite
also not easily detectable (contrary to methane), it increases the risks associated
with hydrogen when it ignites to form a flame. Detection sensors validated for
hydrogen should be used to detect possible hydrogen releases.
• In case of leakage of hydrogen or methane in confined spaces, where leakages

can remain undetected, or in case of large volume releases (e.g. a blow-out, see
below) there is an elevated risk of explosion for both hydrogen and methane,
however, the effects of a hydrogen explosion are different compared to methane.
When a mixture of hydrogen and air explodes, the higher flame propagation
speed potentially generates high pressures that could result in an explosion (a
pressure shock wave) with massive burst damage, i.e., damage to buildings or
even collapse. In contrast, when a mixture of methane and air explodes, the
potential for burst damage is lower, but the longer duration of the flame, in
combination with the heat that it radiates, can potentially lead to lasting harm. In
the absence of confinement and congestion though, no overpressures are
generated, and the consequence of an explosion is limited to a flash fire.
• A catastrophic event on the wellpad (e.g. an accident with a heavy truck, or a

dropped object) could lead to complete or partial removal of the wellhead and/or
Xmas tree with all valves, which could lead to uncontrolled outflow of gas (also
referred to as a blow-out). When ignited, both hydrogen and methane will form a
jet flame (flare), but the hydrogen flame is expected to be narrower and reach
higher, which together with the lower energy content, likely reduces the effect of
heat radiation. A properly installed and operationally tested SSSV 3, which is
mandatory for gas (production and) storage wells, must prevent significant
outflow in case of such catastrophic event. Although SSSV’s are extensively used
in oil and gas industry, their effectiveness in shutting in a flowing hydrogen
storage well is yet to be confirmed.
• Contrary to methane, which is an inert gas, hydrogen is a reactive gas4. It can

potentially react with rocks and reservoir fluids and may interact with microbes in
the reservoir. This might affect reservoir performance (e.g. by pore clogging due
to precipitation of minerals or rapid bacterial growth in the near-wellbore region)
and/or could result in loss of hydrogen and/or contamination of the production
stream due to the formation of H2S, a toxic, corrosive gas that degrades wellbore
materials and poses a threat to human health when released to the atmosphere.
Although the risks associated with UHS are generally known, further research is
required in particular on a) the long-term durability of rocks and (well) materials (steel
alloys, cement, elastomers, etc.) when subjected to hydrogen under an alternating
pressure regime that causes mechanical and thermal stresses, and b) interactions of
hydrogen with rocks, fluids and microbes in reservoirs and their effects on reservoir
performance, quality and retrievability of the stored hydrogen, and integrity and
durability of materials subjected to products of such interactions (e.g. H 2S).
3
SSSV: subsurface safety valve
4
“Reactive” here refers to the ability to react with other chemicals (in the reservoir and/or casing)
Contents
Preface ...................................................................................................................... 2
Project details ............................................................................................................ 2
Acknowledgements.................................................................................................... 2
Summary .................................................................................................................. 3
Contents ................................................................................................................... 5
1 Introduction .............................................................................................................. 6
2 Risk Inventory .......................................................................................................... 7
3 Qualitative risk comparison - UHS vs. UGS ........................................................ 10

3.1 Gas properties ......................................................................................................... 10
3.2 Qualitative comparison of selected risks for H2-storage vs. CH4-storage ............... 10
3.2.1 Risk theme 1: Material integrity and durability ......................................................... 10
3.2.2 Risk theme 2: Leakage of hydrogen ........................................................................ 14
3.2.3 Risk theme 3: Uncontrolled outflow at the wellhead (blow-out) .............................. 18
3.2.4 Risk theme 4: Diffusion and dissolution................................................................... 19
3.2.5 Risk theme 5: Loss of H2 and/or contamination of the production stream.............. 19
3.2.6 Risk theme 6: Subsidence and induced seismicity ................................................. 21
4 Discussion and conclusions ................................................................................ 23
References ............................................................................................................. 26
Appendix 1 – Risk Inventory ................................................................................ 30
Appendix 2 – Consequence & Probability matrix .............................................. 46

1 Introduction
Energy storage can play a pivotal role in the energy transition by adding flexibility to
the sustainable energy system. Storage of energy needs to be deployed at both
small-scale (low power and fast response solutions) and large-scale (longer-term
balancing for grids). While batteries are ideally suited to store and deliver energy with
fast response for a short period, they are not capable of storing the large amounts of
energy that must be supplemented to secure supply in prolonged periods of several
days to more than a week with calm winds and no or little sunshine. At this timescale,
large-scale storage of energy underground, in salt caverns, depleted gas fields and
aquifers, is an attractive option. Underground energy storage provides flexible bulk
power and energy management and offers essential services to society in the form
of strategic energy reserves and balancing solutions for unavoidable seasonal
variations.
The use of the subsurface for energy storage however may introduce risks that can
negatively impact health, safety and environment, system integrity, economics and
the public perception towards this technology (Evans, 2008). Risks associated with
natural gas storage are well-known from decades of operational experience (CMEO,
1993; Pudlo et al., 2013). In contrast, the risks associated with Underground
Hydrogen Storage (UHS) and Compressed Air Energy Storage (CAES) are relatively
underexplored. In this study we identified potential risks and mitigation measures
associated with CAES in salt caverns, and UHS in salt caverns and depleted gas
fields (phase 1), and qualitatively compared selected key technical risks of UHS with
Underground (natural) Gas Storage (UGS; phase 2).
In the first phase a literature review was conducted to inventory risks associated with
hydrogen storage and CAES, and identify potential mitigations to reduce risks. Risks
were included in an Excel-based risk inventory, categorized by project phase, system
component, reservoir type, and classified according to the TEECOPS criteria (i.e.
technical, economic, environmental, commercial, organisational, political and
societal). The purpose of the risk register is to serve as a starting point and/or
checklist to identify and manage risks in underground energy storage projects, and
to provide guidance on potential mitigation measures to reduce the risks.
In the second phase, a qualitative non site-specific comparison between natural gas
storage and hydrogen storage was performed for a selection of key risks. The
selection was made by careful evaluation of the risks and mitigation measures in the
risk inventory, supplemented with expert judgement. This comparison is deemed
valid because UHS essentially uses the same technology as UGS. UGS has been
done for many decades and the risks are well-understood. As such, it can serve as a
point of reference for assessing the risk associated with UHS.
2 Risk Inventory
A risk inventory was created in Excel, which lists the risks and ways to mitigate them.
The aim of the Risk Inventory (RI) is to serve as an instrument to: 1) visualize and
increase awareness of important risks, and 2) indicate the impact of mitigations
relevant for communication and/or permitting. It is a structured template, which is self-
explanatory, has a clear scope and boundaries, and has the possibility to filter risks
on relevance. Each risk is categorized by system component it pertains to, and
project phase where the risk is present, and to classified into the TEECOPS1 criteria.
The Risk Inventory can be found in Appendix 1 – Risk Inventory and can be shared
on request.
Literature survey
The RI was compiled from risks found in literature, and supplemented by internal
TNO expertise and expertise from partners in the LSES consortium. Risks were
ordered by system component to which the risk pertains, project phase during which
the risk is present, classified according to the TEECOPS criteria and incorporated in
the RI template. An important step was cleaning up the inventory by regrouping,
merging and deleting risks. Ultimately, this resulted in an RI with 159 risks associated
with UHS and/or CAES. Furthermore, the RI template also allows for a first qualitative
ranking of the risks, based on their consequence and probability rating.
Inventory structure
The RI template allows to filter by project phase, system component and TEECOPS
criteria, which makes it an efficient template for the determination of specific risks
within different fields of interest.
The risks in the RI are categorized in five project phases (Figure 1) and one general
category for risks that apply to all (or the majority of the) phases:
− Pre-execution phase: the phase during which all work is done in preparation for
the execution phase, including analysis, design, permitting, stakeholder
engagement and contracting;
− Execution phase: the phase during which the facility is constructed /adapted;
− Operational phase: the phase during which the storage operations take place,
i.e., the charging and discharging of energy in the form of compressed air or
hydrogen;
− Decommissioning phase: this phase includes all activities required to abandon
wells, remove surface facilities and clear the site for future use;
− Post-abandonment phase: the phase after decommissioning, during which the
abandoned site is monitored for early detection of failure of barriers that might
lead to the occurrence of a potentially harmful event with negative consequences;
− All phases: risks that apply to all (or most) of the above defined project phases.
Of the 159 risks, 17 are categorized as being relevant during the pre-execution
phase, 32 as relevant during the execute phase, 75 as relevant during the operational
phase, 21 as relevant during the decommissioning phase, and 12 as relevant during
the post-abandonment phase
Figure 1 The structure of the Risk Inventory in Appendix 1 is composed of five project phases, which
are consistent with the typical project workflow.
Additionally, the risks are categorized into the three main (groups of) components of
an underground energy storage system they can pertain to:
1) “Surface Facilities” group: includes compressors, piping, instrumentation,

process facilities;
2) “Well” group: includes the X-mas tree, wellhead, well (completion and
cemented casings), sand-face completion;
3) “Subsurface” group (storage reservoir): includes the target storage reservoir,
the caprock and overburden.
A “General” component group is included, for risks that pertain to all (or multiple) of
the system components. An unfilled section for project specific risks is also present
in the template. In this section risks that are project specific and probably not relevant
for (most) other projects can be noted.
Finally, individual risks have been classified into the TEECOPS criteria (based on the
Peterhead CCS project, 2016):
− Technical: (Sub)surface, Infrastructure, Technology, Operability, Availability,

Integrity, Sustainability, Maintenance
− Economical: Life-Cycle Cost, Phasing, Valuation method, Capacity, Economic
model, Regret costs
− Environmental: Surface exposure, Subsurface environment
− Commercial: Contracting & Procurement, Financing, Business controls, Legal,
Terms & Conditions, Competition, Marketing, Liabilities, Collaboration Agreement
− Organizational: Structure, Resources, Procedures, Project Controls, Knowledge
Management, Systems & IT, Interfaces, Partners, Governance
− Political: Government, Stakeholders, Employment, Regulation, Security,
Reputation, NGOs, Export Control, Localization
− Societal: Community, Public opinion, Social License to Operate
Application
A similar RI was made for the purpose of High Temperature - Aquifer Thermal Energy
Storage (HT-ATES) in the context of the HEATSTORE5 project (Van Unen et al.,
2020). In order to test the robustness and added value of the RI the template was
used in the preparatory study of the HT-ATES demonstration in Middenmeer in The
Netherlands. Prior to the workshop experts were asked to select (but not rank) the
most important risks for each system component from the RI. The participants could
5
https://www.heatstore.eu/
provide their input through a questionnaire (Mentimeter). TNO assessed the expert
input and selected the 10 most relevant risks (or risk themes), which were then
discussed in more detail and ranked in a dedicated workshop. For the ranking of the
risks the consequence – probability matrices from DAGO (DAGO, 2019) were used
(Appendix 2). For a full explanation of this workflow, the reader is referred to van
Unen et al., 2020. The application of the RI in the Middenmeer study was received
very positively by the participants. A similar application process is suggested here to
select the most relevant risks for CAES and UHS and rank them in a risk matrix, as
a precursor to e.g. a more detailed bowtie analysis of causes and consequences of
specific undesired events.
3 Qualitative risk comparison - UHS vs. UGS
The qualitative risk assessment presented in this section focuses on hydrogen

storage in salt caverns and depleted fields. This is done by taking Underground Gas
Storage (UGS) as reference and extrapolating selected risks or risk themes from
UGS to Underground Hydrogen Storage (UHS). In the Netherlands, there are four
UGS facilities where natural gas is stored in depleted fields (Grijpskerk, Norg,
Alkmaar and Bergermeer) and one facility that stores natural gas in salt caverns
(Zuidwending). From the literature study and with the help of relevant experts and the
consortium partners, six high-level risk themes were selected for which to analyze
the risks and compare them to natural gas:
1. Material integrity/durability
2. Leakage (above ground and below ground)
3. Blow-out (uncontrolled flee outflow at the wellhead)
4. Diffusion and dissolution
5. Loss/contamination of production stream
6. Ground motion: subsidence and induced seismicity
3.1 Gas properties
The difference in gas properties between methane and hydrogen forms the basis of
the comparison. Natural gas can be produced in many qualities and the composition
varies per field. The main constituent of natural gas is methane (70-90%). To reduce
complexity pure methane was assumed as representative gas for natural gas. Table
1 (next page) highlights the main properties of methane and hydrogen.
3.2 Qualitative comparison of selected risks for H2-storage vs. CH4-storage
Risks associated with natural gas storage are well understood from decades of
(industrial) experience with natural gas production and storage. This knowledge and
expertise will serve as a point of reference for assessing the risks associated with
underground hydrogen storage, which will be presented in the following sub-sections
for the six risk themes. Evans (2008) suggests that it is important to determine
potential points of failure in an underground gas storage system as these could harm
health, environmental, and economic aspects. By identifying potential points of
failure, putting in place barriers and measures to monitor them, and mitigating the
consequences when failure occurs, the impact to health, safety and environment can
be minimized.
3.2.1 Risk theme 1: Material integrity and durability

The chemical properties of methane and hydrogen are different (see Table 1) and
this could affect the integrity and durability of the materials that are commonly used
for underground storage. In the next subsections, the comparison of the effects of
hydrogen on materials vs. natural gas is split into materials used in pipelines and
surface facilities and well materials because of their difference in characteristics and
technological readiness.
Table 1: Gas properties of methane and gaseous hydrogen (H2Tools, 2020; Hyde & Ellis, 2019;
Klebanoff et al., 2016; Uehera, 2013; Maytal & Pfotenhauer J.M., 2013).
Property Methane (CH4) Hydrogen (H2)
Molecular weight [g/mol] 16.0 2.02
Kinetic Diameter (Å) 3.8 2.89
Diffusion coefficient in air at (NTP6) [cm2/s] 0.16 0.61
Normal boiling point7 (NBP) [°C], 1 atm -162 -253
Solubility in water [mg/ml] 0.022 0.0016
Viscosity at NTP [g/(cmꞏs)] 1.10 E-4 8.81 E-5
Physical state at NTP Gas Gas
Normal density at NTP [kg/m3] 0.668 0.0838
Explosive limits in air [vol%] 6.3 – 13.5 18.3 – 59.0
Minimum spontaneous ignition pressure [bar] 100 41
Heating Values (energy density)8 at 0 °C, 1 bar LHV - HHV [kJ/g] 50 - 55.5 120 - 142
Flammability range in air [vol%] (LEL & UEL) 5.3 - 15 4.0 - 75
Burning velocity at NTP in air [m/s] 0.37 – 0.45 2.6 – 3.2
Flame temperature in air9 [°C] 1875 2045
Max. laminar flame speed gas/air mixture [m/s] 0.374 2.933
Minimum ignition energy at NTP [mJ] 0.29 0.02
Flash point [°C] -188 <-253
Auto ignition temperature in air [°C] 540 585
Thermal conductivity at NTP [W/(mꞏ°C)] 0.0339 0.1825
Quenching distance [mm] 2.0 0.64
Specific volume at NTP [m3/kg] 1.52 11.94
Enthalpy of vaporization at NTP [J/mole] 8.5 0.92
Energy content per unit mass [MJ/kg] 50.02 119.96
Energy content per unit volume [MJ/L] 21.1 7.9
Vapor specific gravity at 25°C, 1atm (air=1) 0.555 0.0696
Joule-Thomson max. inversion temperature [°C] 736 -72
Wobbe index (interchangeability) [MJ/Nm3] 47.91-53.28 40.65-48.23
Calorific value (energy of flame) [MJ/m3] 39.8 12.7
Pipelines and surface facilities

Natural gas storage surface facilities include equipment to compress (compressors,
intercoolers) the gas prior to injecting it into the storage reservoir, for cleaning the
gas (e.g. Pressure-Swing Adsorbers, Thermal-Swing Adsorbers), and for drying the
gas (e.g. glycol-based dryers) upon withdrawal, prior to feeding it back into the grid.
Additionally, pipelines are used to transport the gas from production sites to storage
and from storage to consumers. DNV-GL (2017) together with GTS (Gasunie
Transport Services) assessed the re-use potential of the natural gas transmission
and distribution network for hydrogen, and concluded that this is technically feasible,
which makes it an attractive option to reduce the costs associated with the integration
of hydrogen into our energy system. However, they point out that large and frequent
operational pressure variations should be avoided to minimize the risk of crack
growth, and that non-metallic (e.g. plastic) parts of valves should be replaced.
6
Normal Temperature and Pressure (as defined by NIST, USA) = 20°C (68°F) and 1 atm.
7
The boiling point at 1atm pressure
8
Heating values are the energy, per gram of fuel, generated by a combustion reaction. Higher
heating value (HHV) is obtained when all of the water formed by combustion is liquid. Lower heating
value (LHV) is obtained when all of the water formed by combustion is vapor.
9
Experimentally determined flame temperatures are shown in the table. These values do not differ
significantly from theoretical adiabatic flame temperatures
Furthermore, recent studies demonstrated that the durability of metal pipes could
degrade when they are exposed to hydrogen over long periods of time, particularly
with hydrogen in high concentrations and at high pressures (Melaina et al., 2013), for
which the material durability (primarily of pressure regulators and valves) remains to
be proven (Weidner et al., 2016). González Díez et al. (2020) investigated the
compatibility of hydrogen in mixtures with natural gas, in particular the influence of
hydrogen on the fatigue properties of relevant steel grades and the resulting crack
propagation. Although they concluded that no significant effects due to hydrogen-
enhanced fatigue crack growth are expected for the typical operating conditions and
material types (X42-X70) used in pipeline for natural gas, they stress the importance
of assessing the current condition of the integrity of the pipelines prior to transporting
hydrogen through it.
Surface facilities for hydrogen storage are expected to be very similar to those that
are used for UGS. Operating conditions of underground hydrogen storage are also
similar to those of natural gas storage. As an industrial gas, hydrogen has been
produced (from natural gas, by steam methane reforming), transported (through
pipelines), stored (in high pressure cylinders) and used (e.g. in the petrochemical
industry) for decades, and the risk and safety aspects are well-known. In fact (as is
described in the report of work package 2 of the LSES project), hydrogen storage in
salt caverns is already operational at 4 locations in the world, and no safety incidents
have been reported. As such, there is confidence in the technology and years of
handling hydrogen on an industrial scale have provided the experience to safely
operate facilities where hydrogen is produced, stored, or used.
Nonetheless, when using hydrogen in a mix with natural gas in existing equipment
such as compressors, care must be taken. Concentrations up to 10%vol of hydrogen
have been claimed to be acceptable in existing mechanical compressors without
complicating operation and/or degrading performance. However, particular attention
must be given to material compatibility and fugitive losses through the seals
(González Díez et al., 2020). Furthermore, to compress a (near) pure hydrogen
stream with existing (mechanical) compressors such as are currently used will require
extensive changes because many more impellors will be required. In fact, to
compress a pure hydrogen stream, a reciprocating compressor is a more suitable
compressor type than a mechanical compressor. As such, re-use of existing UGS
surface facilities for (near) pure hydrogen storage is not straightforward and
replacement of existing gas processing units by new ones is likely to be required 10.
Well materials
UGS wells are very similar to gas production wells and use similar materials which
are specified by mature standards and guidelines based on decades of experience
with production of natural gas. The wells used for hydrogen storage would have to
be designed with completion materials that are compatible with hydrogen.
Additionally, they would have to be compatible with the products that could be
generated from chemical and microbiological reactions with hydrogen during storage
operations, e.g. H2S (see Section 3.2.5). A schematic diagram of a design of a gas
(storage) well is shown in Figure 2. One of the well materials that is in direct contact
with hydrogen is (alloy) steel. In gas wells the steel (alloy) components of the
completion are in direct contact with the storage medium (e.g. inner casing tubing,
10
Personal communication NAM
production casing / liner, SSSV, packers, etc.), which all need to be hydrogen
resistant under a wide range of temperatures (DBI-GUT, 2017). Often recognized
processes involving hydrogen that affect the integrity and/or durability of steel alloys
are: hydrogen blistering, hydrogen-induced cracking and hydrogen embrittlement
(Gilette & Kolpa, 2007; Gonzales-Diez et al., 2020). These processes are influenced
by temperature, pressure, hydrogen concentration and stress fields (Reitenbach et
al., 2014). Additionally, corrosion could play a role and if that takes place defects can
be created on which cracks could develop due to tensile stress build up, which could
subsequently result in leakage.
Figure 2: Schematic well installation used for gas wells, representing different failure scenarios of
the well (SODM, 2019)
Hydrogen may activate such defects when using the existing UGS infrastructure
(Gonzales-Diez et al., 2020). Lastly, hydrogen is expected to be able to flow at higher
velocities through the same well, which increases the likelihood of erosion in
situations where (solid) particles are present in the production stream. In order to
prevent significant erosion of the materials it should be ensured that the flow velocity
is always below the erosional velocity specified by the manufacturer, in the industry
a velocity of 100 m/s is commonly used (see report of work package 2 of the LSES
project).
The second major material used for well construction is cement, which is used to seal
off the annulus between the casing and the formation (see Figure 2). In order to
prevent migration (or leakage) of gas along the outside of the casing and/or through
the cement, it has to be chemically resistant for the stored gas and tight enough to
make sure that the gas cannot penetrate through it into the (shallower) rock
formations. Therefore, the cementation operations are a critical aspect of the well
integrity and specific requirements are stated in the NOGEPA 41 standard on well
integrity (NOGEPA OPCOM, 2016). As hydrogen has a smaller size and higher
diffusivity compared to methane, the cement must be adjusted in order to prevent
migration of the hydrogen through the cement. The risk of chemical alteration of the
cement by contact with hydrogen is considered to be low (DBI-GUT, 2017). Lastly,
the potential effects of cyclic loading of the well, and especially on the cement, on the
integrity (sealing function) of the well have to be taken into account.
The last important material in UGS wells are elastomers, which are used in the
packers and fittings. Further investigation on these elastomers must be done in order
to define their resistance to the higher diffusivity of hydrogen compared to methane
(DBI-GUT, 2017). Penetration of hydrogen through these elements could lead to
integrity loss as a result of fast decompression and inner blister fracturing
(Reitenbach et al., 2014). It must be noted that during normal operations such sudden
decompressions do not take place in gas storage facilities.
Based on the above sections it can be concluded that surface facilities for hydrogen
are at similar technology level as for UGS and that they are not expected to increase
the risk profile with respect to material integrity and durability. Additionally, re-using
the natural gas transmission and distribution network could provide an opportunity to
accelerate the implementation of hydrogen at reduced cost, but requires additional
research. Lastly, the current state of the art well completion materials for natural gas
wells are not confirmed to be fully hydrogen resistant. Therefore, the sealing
effectiveness and corrosion resistance of all materials used for the completion (steel
alloys, cement, elastomers and seals) have to be subjected to a technical integrity
evaluation prior to storing hydrogen using the existing oil and gas infrastructure or
well materials.
3.2.2 Risk theme 2: Leakage of hydrogen

Leakage is defined here as the accidental escape of gas out of the facilities. Leakage
is one of the most important hazards in any underground storage system and has
impact in all of the TEECOPS aspects. Here we focus on causes of potential leakage
from the well system and/or the storage reservoir (either a salt cavern or a depleted
field), and how these might differ between hydrogen and methane. Surface facilities
for hydrogen are advanced and operational and therefor their influence on the risk
profile for leakage is expected to be low, therefore these have not been included in
the assessment. Furthermore, differences in leakage detection and the potential
impact of leakage between hydrogen and methane are discussed. Because of the
potential impact of uncontrolled outflow at surface (blow-out), this has been treated
as a separate risk theme and will be discussed in Section 3.2.3.
Well leakage
Wells are considered an important and critical component in the infrastructure to
produce, store and distribute natural gas. As long as the integrity of the well is not
compromised, stored product cannot leak to the environment. UGS wells are
constructed to have multiple barriers to prevent leakage, such that when one barrier
(e.g. casing section or valve) fails, other barriers are still in place. Figure 2 displays a
typical well configuration and the potential failure scenarios (SODM, 2019). The
barriers can be divided into primary and secondary barriers. Primary barriers are the
production casing, liner and liner cement below the production packer, the production
packer, the completion string below the (Surface-controlled) Subsurface Safety Valve
(SSSV), and the SSSV itself. The SSSV, which can be controlled from the surface,
is a key component of a gas storage well. In case of an unwanted event (e.g. a
flowline rupture) the SSSV will be closed to prevent an uncontrolled, prolonged
outflow from the well. It is failsafe, i.e., hydraulic pressure is required to force the
valve to its open position, and when the pressure in the hydraulic system drops, the
valve will automatically return to the closed (safe) position (EnergyStock, 2017).
Secondary barriers are the production casing, liner and liner cement above the
production packer, the completion string above the SSSV, the wellhead (incl. casing
hanger with seals and wellhead valves), and the production tree (body and master
valves). The performance of the well barriers must be verified through appropriate
functional testing. Monitoring of the well barrier elements and application of any
changes to the well are to be documented through the entire storage lifecycle
(including the abandonment phase). It is the responsibility of the well operator to
ensure the well barriers can withstand the anticipated loads (Opedal et al., 2020).
In order to avoid uncontrolled gas leakages in any part of the well, careful evaluation
of the well integrity is required under operational storage conditions, for which many
technical, operational and organizational procedures must be applied (DBI-GUT,
2017). To our knowledge no specific standards or guidelines exist for well designs for
hydrogen storage. SODM shows that storage wells and gas producing wells have a
very similar configuration. However, they do regard gas storage as a separate sector
and subject it to an additional legal regime (SODM, 2019). Standards for well design
of comparable gas wells are mature, e.g. by ISO, API and Norsok. In the Netherlands
the NOGEPA standard 41 describes well design requirements and also requires the
use of multiple barriers.
The decades of experience in the oil and gas industry, dissemination of new
techniques and best practices, advances in well design (e.g. double barrier policy),
and testing led to an ongoing reduction in the frequency and severity of incidents
(Bérest et al., 2019). However, there is uncertainty about the hydrogen resistance of
borehole completions materials currently used for UGS wells, if subjected to
hydrogen for a long period of time (see section 3.2.1), and about the tightness of
casing and tubing connections. The smaller molecular size and higher diffusion
coefficient (more than 3 times higher) of hydrogen compared to methane, in
combination with the embrittling nature of hydrogen, increases the probability that it
permeates through steels and materials used in wells and distribution systems
(Melaina et al., 2013). This could result in higher leakage probabilities of hydrogen
compared to methane when using the existing UGS infrastructure and/or materials
(H2Tools, 2020). In order to be able to withstand damage (e.g. corrosion) and
subsequently penetration of hydrogen through pipe walls, the well completion
components that are in direct contact with hydrogen (steel alloys, cement, seals and
elastomer packers) have to be made hydrogen resistant (DBI-GUT, 2017), also see
Section 3.2.1 and Figure 2. Furthermore, the tightness of casing and tubing
connections must be tested specifically for hydrogen. Finally, while SSSV’s are
extensively used in oil and gas industry, their effectiveness in shutting in a flowing
hydrogen storage well is yet to be confirmed. In fact, it is likely that acceptable leakage
rates for hydrogen and hence verification criteria will be different, in which case an
SSSV may have to be designed specifically for the purpose of hydrogen storage.
Leakage from salt cavern or porous reservoir

In order to store gas in the subsurface, the storage container (depleted field or salt
cavern) must trap the gas such that it cannot migrate out of the storage complex. Salt
is commonly accepted to be impermeable, and therefore a salt cavern, which is
essentially a hole in a massive salt body, is considered to be a leak-tight container in
which liquids and gases can be effectively and safely stored. In fact, they have proven
their great sealing capacity (very low permeability and porosity) by effectively storing
a variety of gases including methane and hydrogen (Kruck et al., 2013). Furthermore,
in work package 2 of the LSES project, where the effects of cyclic pressure operation
on the leak tightness of caverns was investigated, it was shown that leak tightness is
not compromised by such cyclic pressure variations.
In natural gas reservoirs, trapping of gas requires the presence of an impermeable

caprock (salt, tight shale) that acts as a seal to migration of the gas out of the
reservoir. Since the natural gas was trapped in the reservoir for millions of years, the
reservoir has proven its effectiveness as a storage container. However, for hydrogen
this may not necessarily be the case. Its smaller molecular size and higher diffusivity
compared to methane (see Table 1) increase the potential of hydrogen permeation
into and subsequently percolation through the surrounding sealing rocks of the
storage reservoir. Ultimately, this may lead to loss of tightness and hydraulic integrity
of the rock formations (DBI-GUT, 2017). As such, the sealing capabilities of cap rocks
for hydrogen storage in depleted fields would have to be assessed on a case by case
basis prior to deciding to use the reservoir for this purpose.
Furthermore, geological uncertainties could potentially increase the probability of

leakage, because e.g. the occurrence of non-halite interbeds, heterogeneities in
sealing formations, and the sealing capacity of faults for hydrogen cannot always be
observed or determined. However, leakage incidents associated with subsurface
geological uncertainties are relatively unlikely (compared to leakages associated with
well failures).
Leakage detection
In case of any leakage at surface, early detection is important to enable a swift
response, limit the volume of the leak and thereby the potential impact on health,
safety and the environment. However, both gaseous hydrogen and methane are
undetectable by human senses as both gasses are colourless, odourless and
tasteless. In order to be able to detect natural gas a sulphur-containing odorant has
been added. Unfortunately, there are no known odorants light enough to "travel with"
hydrogen at the same dispersion rate, making it difficult to detect hydrogen gas
(H2Tools, 2020). At hydrocarbon (gas) production platforms several types of gas
leakage and flame detectors are used (e.g. ultrasonic gas leak detector, open path
gas detector, IR (InfraRed) gas detector, IR flame detectors; (Koelewijn et al., 2019).
However, these detectors are not suited for detecting hydrogen. Specific detectors
for hydrogen do exist, for example the catalytic bead detector that is suitable for
detecting hydrogen at lower flammable limit (LFL) levels. These sensors can detect
any combustible gas that combines with oxygen to generate heat. Additionally, when
hydrogen ignites it has an almost invisible pale blue flame and has low radiant heat,
which results in decreased detectability of ignited hydrogen compared to methane.
Detectors with the ability to sense hydrogen flames include: thermal detectors, UV
detectors and/or multispectral IR detectors, which have the ability to sense the non-
visible spectrum of electromagnetic radiation (Koelewijn et al., 2020).
Leakage could also happen in the subsurface, either because of unforeseen

geological pathways or (more realistically) along the wellbore. Leakage along the
wellbore occurs when multiple barriers would fail, and by monitoring these barriers
such leakage can be detected at an early stage. For example, to detect leakage of
the packer or the production tubing the pressure in the annular space between
production tubing and casing is monitored (see Figure 2). If methane or hydrogen
were to enter this annular space then the pressure would increase, and this would be
detected. Detection of leakage from the storage reservoir itself (salt cavern, porous
reservoir) is much more difficult, in particular because rates of leakage are commonly
low, whereas the volume of stored product is very large. Although the pressure in the
storage system is monitored, the change in pressure due to such slow leakage would
not be measurable.
Potential consequences of leakage

In case of leakage, gas escapes into the environment (either surface or subsurface).
The potential consequences of leakage are expected to be different for leakage in
the subsurface vs. leakage at surface, as well as for leakage of hydrogen vs. leakage
of methane. In case of subsurface leakage of methane there are two main scenarios
that could occur. Firstly, the stored product migrates towards the surface in the
immediate vicinity of the storage facility, which could result in health, safety and
environmental hazards (heat radiation, explosion, suffocation, groundwater
contamination). Secondly, the stored product leaks into adjacent formations away
from the storage facility, where after the product contaminates the groundwater, and,
if released to the atmosphere at significantly high rate (not likely), similar health,
safety and environmental hazards. Furthermore, in both scenarios there is economic
risk in a sense that the gas becomes unrecoverable (Evans, 2008), as well as risk of
reputational damage and reduced public support.
If hydrogen is released in an unconfined open environment it will typically rise and

disperse more rapidly (several meters per second) compared to the heavier methane
(H2Tools, 2020). This rapid dispersion of hydrogen in open spaces (atmosphere)
makes it less likely that significant amounts of hydrogen could accumulate that could
cause an explosion in case of ignition, this in contrast to methane. Also, on ignition
methane radiates heat and creates a flame that is clearly visibly. Ignited hydrogen on
the other hand radiates little (infrared) heat (IR), but emits substantial UV (ultraviolet)
radiation. The lack of IR gives little sensation of heat but the exposure to a hydrogen
flame still causes severe burns because of the UV radiation. Because a burning
hydrogen flame is also not easily detectable, it increases the risks associated with
hydrogen when it ignites to form a flame. In case of leakage of hydrogen or methane
in confined spaces the risks of hydrogen accumulation are likely to be more severe
than those for methane accumulation. Although both gases, when mixed with air in a
combustible gas cloud, can explode when ignited, hydrogen is more prone to ignite
because of the lower ignition energy and wider flammability range (vol% of hydrogen
vs. air; Table 1). Furthermore, when a mixture of hydrogen and air explodes, the
higher flame propagation speed potentially generates high pressures that could result
in an explosion (a pressure shock wave) with massive burst damage, i.e., damage to
buildings or even collapse (H2Tools, 2020; Hyde & Ellis, 2019). In contrast, when a
mixture of methane and air explodes, the potential for burst damage is lower, but the
longer duration of the flame, in combination with the heat that it radiates, can
potentially lead to lasting harm (Li et al., 2015).
3.2.3 Risk theme 3: Uncontrolled outflow at the wellhead (blow-out)

The worst-case scenario is an incident that leads to rip-off of the well head with its
multiple safety installations, which could lead to an uncontrolled outflow of the stored
gas at the wellhead (also referred to as a blow-out). If the gas ignites, it could form a
gas flare (flash fire), of which the heat radiation effects are expected to be different
between hydrogen and methane (as discussed above). When ignition is not
immediate, a cloud of methane or hydrogen could potentially form which, upon
ignition, would cause an explosion.
A significant difference between hydrogen and methane in this context is the

flammability range (volumetric ratio gas to air) of hydrogen (between 4% and 75% in
air; Table 1), which is very wide compared to methane (between 5.3-15% in air; Table
1). Furthermore, although both hydrogen and methane can ignite when mixed even
in small amounts with ordinary air, hydrogen requires a much lower ignition energy,
which is the energy that is required to initiate hydrogen combustion (0.02 mJ for
hydrogen and 0.29 mJ for methane; Table 1). This property gives hydrogen a
significantly higher ignition potential than methane. Based on the combination of
these properties it can be assumed that ignition is almost certain to happen in case
of a hydrogen release, whereas this is not necessarily the case for methane. Although
this reduces the risks of explosion of hydrogen vs. methane in case of low outflow
rates (leakage), in case of a blow-out, outflow rates are expected to be very high (in
the order of tens to a hundred kg/s), and may effectively lead to congestion at the
location of release. In such a situation even a minor delay in ignition (e.g. tenths of
seconds to seconds) could result in an explosion because significant amounts of
hydrogen would have already been released. In the absence of confinement and
congestion though, no overpressures are generate, and the consequence of an
explosion is limited to a flash fire.
However, a significant outflow in case of a blow-out event is very unlikely if an SSSV

has been installed at the correct depth in the well, also see Figure 2. Such safety
valves are designed to be failsafe, i.e., to close off the well automatically in case of
an incident at surface, and are generally installed below the crater depth. According
to NOGEPA standard 41 section 3.5.1 all wells that are capable of sustained free flow
are required to have an SSSV installed (NOGEPA OPCOM, 2016). It is only accepted
as a functional barrier after having been installed in the well and tested under
operational conditions, and is active at all times during the storage operation. It must
be removed though when a workover of the well must be performed, in which case
an alternative blow-out prevention barrier should be installed, e.g., in the form of a
BOP (blow-out preventer) that is suitable to close on hydrogen gas.
3.2.4 Risk theme 4: Diffusion and dissolution

The hydrogen molecule is the smallest chemical particle known (see kinetic diameter
in Table 1). In its gaseous state this molecule has a high penetrability associated with
a high diffusion coefficient in solids, which is more than three times higher than
methane (see Table 1, Tarkowski, 2019). This gives a higher potential for diffusional
transport of hydrogen compared to methane when stored in (naturally dry) salt
structures. Alternatively, when the storage systems are water-enriched porous
structures, such as aquifers and depleted gas fields, dissolution processes play a
dominant role. In aquifers, the loss of hydrogen is expected to be lower compared to
methane, because of the much lower solubility of hydrogen in water, which is more
than 13 times lower than for methane (Table 1). However, in depleted fields, the
situation will be different, because the formation water is already saturated with
methane. As such, no loss is expected when these are re-used for UGS, while their
re-use for hydrogen can cause loss because some of the methane in the brine is
expected to be substituted with hydrogen.
3.2.5 Risk theme 5: Loss of H2 and/or contamination of the production stream

One of the major operational challenges in underground storage systems is
associated with the loss or contamination of the stored product through geo- and
biochemical reactions (Foh et al., 1979; Lord, 2009). Such reactions could result in:
− the formation of corrosive and toxic fluids (most notably H2S) that would enter
into the production stream, thereby contaminating it;
− dissolution and precipitation of minerals that could affect flow of fluids through
the reservoir, and
− accelerated growth of microbial populations that would clog in particular the
region of the reservoir in the direct vicinity of the wellbore (near-wellbore
region);
− loss of hydrogen from the storage system.
Ultimately, this could lead to deterioration of the cap- and reservoir rock, integrity loss
of wellbore materials and interfaces, alteration of crucial reservoir properties,
pressure loss, water cuts, and possible temperature changes in the reservoir (DBI-
GUT, 2017; Hemme & van Berk, 2017).
The kinetics of geo- and biochemical reactions and the number of reacting
elements/micro-organisms are the key influencers on the amount of generated and
released harmful products. In general, higher temperatures, pressures (with greater
depths), catalysts and salinity levels can increase the occurrence and rate of
chemical reactions and microbial activity (Tarkowski, 2019). The potential for
chemical and microbiological reactions are generally thought to be lower in salt
caverns (due to less water, microbiological activity and mineral concentrations) than
in porous depleted gas fields and aquifers (DBI-GUT, 2017; Hemme & van Berk,
2018). Hydrogen chemical reactivity, being restricted to redox reactions, is known to
be kinetically limited, likely because of the apolar nature of the molecule and the
strong H–H binding energy (436 kJ/mol) that requires the overstepping of a high
energetic barrier before an eventual electronic transfer can take place (Truche et al.,
2013). Therefore, most of the possible redox reactions induced by hydrogen remain
insignificant at low temperature, even on a geological time scale, provided that no
catalyst (bacteria, mineral surfaces, engineered materials) is present.
However, some hydrogen-induced redox reactions may be significant at low

temperatures. In particular H2S was demonstrated to form geochemically under
medium-hydrothermal condition (Truche et al., 2010) by an hydrogen-induced redox
reaction with the mineral pyrite (FeS2) as a potential oxidant for hydrogen, which
thereby transforms to pyrrhotite (FeS). This reaction exhibits all the characteristics of
a coupled dissolution-precipitation mechanism occurring at the pyrite–pyrrhotite
interface, and can modify the redox potential and pH of the formation water. It is
considered an important potential risk of underground hydrogen storage, in particular
in depleted fields, and requires further research to better understand under what
subsurface conditions and at which rates H2S forms, what the effects are, and to what
extent the associated risks can be mitigated.
A second mechanism by which H2S can form is bacterial sulfate reduction, i.e.,
whereby bacteria reduce SO42- to H2S in the presence of hydrogen. H2S is a highly
toxic gas even in small quantities, very aggressive towards storage facilities (weak
acid if dissolved in water and highly corrosive), can pose a threat to the environment
and can reduce the quality of the stored gas (Hemme & van Berk, 2017; SPCS, 2020;
DGI-BUT, 2017).There is ample experience with H2S in production streams from oil
and gas industry, and procedures for making borehole completion materials H2S
resistant are well established, which provides confidence that the risks of H2S to
wellbore integrity can be managed (DBIGUT, 2017; Lord, 2009), albeit at increased
cost (for H2S-resistant materials). In the Netherlands some fields produced so called
sour gas with considerable H2S concentrations, e.g. above 1 mol% (10,000 ppm). If
the presence of H2S is suspected additional safety measures, like using low alloy H2S
resistant steel materials and H2S detectors, are required. While stationary H2S
detectors generate an evacuation alarm at 5 ppm, personal (H2S) gas detectors
already generate an evacuation alarm at 1.6 ppm (this is the legal threshold value,
i.e., the max. concentration of H2S a person is allowed to work in for max. 8 hours),
after which the person involved will go to the muster location. Above 5 ppm the use
of breathing protection is required, either by using an independent breathing
apparatus or by a personal H2S detector in combination with an escape mask (SPCS,
2020).
Furthermore, for storage in depleted fields, pore clogging (i.e. filling/obstruction of the
pores) is an important risk because it could affect reservoir performance by
decreasing the permeability and porosity of the storage reservoir. This is particularly
relevant for the near-wellbore region (DBIGUT, 2017; Hemme & van Berk, 2018;
Panfilov, 2010). Clogging can either be the result of physical, biological or chemical
reactions (Maliva, 2020). Physical clogging can result from the mobilization of clay
particles, suspended matter or clay swelling (Konikow et al., 2001; Pavelic et al.,
2007). Biological clogging could take place during microbial growth and bacterial
accumulation. This process could accelerate when nutrient-rich water or organic
matter is present, which could stimulate microbially mediated redox reactions and
biomass growth (National Research Council, 2008). Chemical clogging can result
from hydrogeochemical reactions that lead to mineral precipitation (e.g. calcite,
gypsum, phosphates and oxides). Also, H2S can lead to the precipitation of
amorphous ferrous sulfide, which may cause plugging (Hemme & van Berk, 2018).
Before injecting hydrogen into any reservoir or cavern it is recommended to study the
mineralogical, chemical, physical and microbiological status of the storage reservoir
by measuring rock and fluid compositions, establishing the presence of bacterial
populations in the formation water, and by performing laboratory tests of reactions of

hydrogen with rocks and fluids in the reservoir under reservoir conditions. From these
tests the impact of geo- and biochemical reactions on reservoir properties
(performance), integrity and durability of rocks, well materials and interfaces between
them, and fluid flow in the reservoir and along the wells can be evaluated
(HyINTEGER, 2017). Additionally, the composition of the production stream must be
analyzed continuously, e.g. to be able to detect an increase in the H2S concentration
(Hemme & van Berk, 2017).
3.2.6 Risk theme 6: Subsidence and induced seismicity

Induced seismicity and subsidence have increasingly become important risks that the
wider public in the Netherlands is aware of, especially because of the issues with the
Groningen gas field. Over the past decade, several earthquakes have occurred in the
Groningen area, the largest one being the Huizinge earthquake in 2012 (3.6 on the
Richter scale), which are attributed to the production of natural gas from this field
(Vlek, 2019). Furthermore, it has been known for a long time that extraction of natural
gas and salt from the subsurface causes subsidence at the surface. In the case of
salt caverns, the amount and rate of subsidence depend on the rate of salt creep,
which itself is a function of salt type, pressure and temperature. In the case of porous
reservoirs, the amount and rate of subsidence area function of compaction, which in
turn is a function of the pressure inside the reservoir, the friction angle, the type of
reservoir rock, and he properties of surrounding formations (Wang et al., 2013).
In depleting gas fields, induced seismicity is generally caused by a pressure decline

in the reservoir. When operating pressures are too low, the reservoir rock is
progressively unable to support the overlying rock mass (overburden), and
compaction could take place. Because compaction does not occur at the same speed
at every location in the reservoir (it depends on the properties of the reservoir rock,
which vary in space), faults form that absorb the vertical movement brought about by
the (differential) compaction process. This movement along faults, which commonly
occurs abruptly (stress builds up and when too high it is released suddenly),
potentially causes earthquakes (induced seismicity). A similar effect could potentially
occur in porous reservoirs when used for storage. If storage pressures are allowed
to become lower than the pressure at time of cessation of production of gas and
conversion to a storage reservoir, compaction can continue, potentially leading to
induced seismic events. Furthermore, the pressure inside the storage reservoir
should not exceed a certain maximum pressure (lithostatic pressure) to avoid
fracturing of the rock. In practice, a safe operational pressure range is commonly
agreed with the regulator. It is key to remain within this operational pressure range to
reduce the magnitude and rate of surface subsidence and/or induced seismicity (Liu
et al., 2014). Under normal operating conditions (between the minimum and
maximum pressures) and assuming that a similar approach for UHS is used as for
UGS (with wide safety margins), no contrasting differences between the two storage
options are expected with respect to subsidence or induced seismicity.
In salt, stress build-up leading to faulting is highly unlikely because it behaves visco-
plastically, i.e., it bends rather than breaks, and therefore the risk of earthquakes
induced directly by the storage operations (as a consequence of cavern
convergence) must be considered negligible. Notwithstanding the above, prolonged
plastic deformation in salt may lead to movement along existing faults in brittle rock
layers above the salt, in particular above and around salt domes, which are often very
high but laterally confined. As an example, a very low magnitude (Mw=1) event was
recorded near the Zuidwending gas storage facility in January 2019. Although the
mechanism that caused it remains to be investigated, it is hypothesized that the brittle
rock overlying the salt dome or in its vicinity might have moved due to salt creep
(Ruigrok et al., Royal Netherlands Meteorological Institute, 2019). An alternative
cause could have been the breaking off of a (part of a) rock bench from the roof of
the cavern, which also generates a detectible seismic signal. To what extent the local
gas storage operations triggered the event or contributed to its occurrence is
unknown.
4 Discussion and conclusions
An important aspect of underground energy storage is the awareness of potential

risks associated with the use of the subsurface for this purpose. In this study the
potential risks associated with Underground Hydrogen Storage (UHS) and
Compressed Air Energy Storage (CAES) in salt caverns, and UHS in depleted gas
fields (porous media) were identified, and possible mitigation measures were
explored. Risks were inventoried by conducting a literature review, and
supplemented with expert knowledge. In total, 159 risks were identified, and included
in a risk inventory that categorizes the risks into their relevant project phase, system
component, and reservoir storage type (cavern vs. reservoir) and classifies them
according to the TEECOPS criteria1. Of the 159 risks, 17 are categorized as being
relevant during the pre-execution phase, 32 as relevant during the execute phase,
75 as relevant during the operational phase, 21 as relevant during the
decommissioning phase, and 12 as relevant during the post-abandonment phase.
The purpose of the risk inventory is to serve as a starting point for identifying and
managing risks in development projects, and to provide guidance on potential
mitigation measures to reduce the risks. A similar inventory was created for HT-ATES
and successfully applied in a preparatory study of a HT-ATES (high-temperature
aquifer thermal energy storage) demonstration project (Van Unen et al., 2020) to
select key risks and rank them. As such, this RI and approach is also recommended
in a preparatory study for CAES or UHS to select the most relevant risks and rank
them in a risk matrix, as a precursor to e.g. a more detailed bowtie analysis of causes
and consequences of specific undesired events.
For UHS to become an attractive solution the associated risks remain to be

thoroughly evaluated. As a step towards achieving this, here a selection of six key
risk themes associated with storage of hydrogen was made: material
integrity/durability, leakage of hydrogen, blow-out, diffusion and dissolution, loss
and/or contamination of hydrogen, and ground motion (subsidence, induced
seismicity). A qualitative non site-specific comparison was made for these risk
themes between UGS and UHS, primarily based on differences in gas properties
(with the properties of methane assumed representative for natural gas). Overall aim
of this comparison was to leverage the experience from UGS to provide useful
information to better understand and reduce risks and consequences, increase
control and inform stakeholders. Although in general, UGS and UHS have a similar
risk profile, there are also differences that were highlighted in this study.
One important difference between UHS and UGS is the way in which hydrogen and
methane affect material integrity and durability, and this has implications for re-use
of pipelines and surface facilities that are currently used to transport and store natural
gas. Hydrogen is a smaller and more diffusive molecule that can more easily
permeate through materials, especially when they contain defects and/or cracks.
Additionally, the embrittling nature of hydrogen can lead to progressive growth of
such defects and cracks when subjected to large, frequent pressure variations such
as are to be expected in UHS. The compatibility of hydrogen in mixtures with natural
gas, in particular the influence of hydrogen on the fatigue properties of relevant steel
grades and the resulting crack propagation was investigated by González Díez et al.
(2020). Although they concluded that no significant effects due to hydrogen-
enhanced fatigue crack growth are expected for the typical operating conditions (65-
80 bar) and material types (X42-X70) used in pipelines for natural gas, they stress
the importance of assessing the current condition of the integrity of the pipelines prior
to transporting hydrogen through it. Furthermore, re-use of existing UGS surface
facilities for (near) pure hydrogen storage is not straightforward. For example, to
compress a (near) pure hydrogen stream with existing (mechanical) compressors
such as are currently used will require extensive changes because many more
impellors will be required. In fact, to compress a pure hydrogen stream, a
reciprocating compressor is a more suitable compressor type than a mechanical
compressor. As such, replacement of existing gas processing units by new ones is
likely to be required10.
Leakage of hydrogen below-ground caused by integrity failure of wellbore materials

(steels/joints, cement, elastomers) and interfaces is an important potential risk of
UHS. UHS wells will be very similar to UGS wells, which in turn are very similar to
gas production wells. As such, standard practice would be to use similar well
materials that are specified by mature standards and guidelines based on decades
of experience with (production and) storage of natural gas. However, current state-
of-the-art well completion materials for gas storage wells are not proven to be fully
hydrogen resistant. Therefore, UHS wells must be designed with completion
materials that are proven compatible with hydrogen, i.e., all materials (steel alloys,
cement, elastomers and seals) used must have been subjected to a specific technical
integrity evaluation for hydrogen. Additionally, they would have to be compatible with
the products that could be generated from chemical and microbiological reactions
with hydrogen during storage operations, in particular H2S.
If leakage above-ground were to happen, in an unconfined open environment, then

hydrogen is expected to flow out faster and rise and disperse more rapidly than
methane. This rapid dispersion of hydrogen in open spaces (atmosphere) makes it
less likely that significant amounts of hydrogen could accumulate that could lead to
suffocation, or an explosion in case of ignition, this in contrast to methane. Both gases
are flammable, but hydrogen has a much wider flammability range and a much lower
ignition energy compared to methane, which gives hydrogen a higher ignition
potential compared to methane. In fact, based on the combination of these properties
it can be assumed that ignition is almost certain to happen in case of a hydrogen
release, whereas this is not necessarily the case for methane. On ignition methane
radiates heat and creates a flame that is clearly visibly. Ignited hydrogen on the other
hand radiates little (infrared) heat (IR), but emits substantial UV (ultraviolet) radiation.
The lack of IR gives little sensation of heat but the exposure to a hydrogen flame still
causes severe burns because of the UV radiation. Because a burning hydrogen flame
is also not easily detectable (contrary to methane), it increases the risks associated
with hydrogen when it ignites to form a flame. Detection sensors validated for
hydrogen should be used to detect possible hydrogen releases.
In case of leakage of hydrogen or methane in confined spaces, there is an elevated

risk of explosion for both hydrogen and methane, however, the effects of a hydrogen
explosion are different compared to methane. When a mixture of hydrogen and air
explodes, the higher flame propagation speed potentially generates high pressures
that could result in an explosion (a pressure shock wave) with massive burst damage,
i.e., damage to buildings or even collapse (H2Tools, 2020; Hyde & Ellis, 2019). In
contrast, when a mixture of methane and air explodes, the potential for burst damage
is lower, but the longer duration of the flame, in combination with the heat that it
radiates, can potentially lead to lasting harm (Li et al., 2015). In the absence of
confinement and congestion though, no overpressures are generated, and the
consequence of an explosion is limited to a flash fire.
A catastrophic event on the wellpad (e.g. an accident with a heavy truck, or a dropped
object) could lead to complete or partial removal the wellhead and/or Xmas tree with
all valves, which could lead to uncontrolled outflow of gas (also referred to as a blow-
out). A properly installed and operationally tested SSSV, which is mandatory for gas
(production and) storage wells, must prevent significant outflow in case of such
catastrophic event. Although SSSV’s are extensively used in oil and gas industry,
their effectiveness in shutting in a flowing hydrogen storage well is yet to be
confirmed. In fact, it is likely that acceptable leakage rates for hydrogen and hence
verification criteria will be different, in which case an SSSV may have to be designed
specifically for the purpose of hydrogen storage.
The retrievability of the gas from the subsurface reservoir is influenced by diffusion
and dissolution, by geochemical reactions, and by microbiological activity. Because
hydrogen is a lighter gas with a smaller molecular size and higher diffusivity
compared to methane gas, it is more prone to migrate through caprocks, wellbore
materials and materials used in surface facilities. Additionally, hydrogen has more
potential for reacting with rocks, reservoir fluids and interacting with microbes in the
reservoir compared to methane. This might affect reservoir performance (e.g. by pore
clogging due to precipitation of minerals or rapid bacterial growth in the near-wellbore
region) and/or could result in loss of hydrogen and/or contamination of the production
stream due to the formation of H2S, a toxic, corrosive gas that degrades wellbore
materials and poses a threat to human health when released to the atmosphere.
In UGS operations wide safety margins are applied to minimize the risks associated
with too high pressures that might either fracture the rock and/or re-activate faults
(and induce seismic events), and too low pressures that would cause further
(differential) compaction (and subsidence or induced seismic events). In this way, the
risks of subsidence and induced seismicity are minimized.
To conclude, although the risks associated with UHS are generally known, further
research is required in particular on the long-term durability of materials subjected to
hydrogen, and interactions of hydrogen with rocks, fluids and microbes in reservoirs.
Furthermore, the availability of specific standards, guidelines and perhaps even a
regulatory framework (laws) would be beneficial for the application of hydrogen
storage, but currently (to our knowledge) this does not exist, although the
development of a separate policy framework aimed at mitigating the risks of hydrogen
is currently ongoing. To analyse and demonstrate the causal relations between
potential threats, failing barriers, and consequences, it is recommended to perform a
bow-tie analysis for selected risk themes relevant for a specific use case of UHS. As
a pre-cursor to such a bow-tie analysis, a workshop with experts in the relevant field
could be organized to rank the risks (quantitively) associated with the specific use
case of UHS in a risk matrix based on their consequence and probability rating, and
leveraging the risk inventory developed here.
References
Bérest, P., Reveillere, A., Evans, D., & Stower, M., 2019. Review and analysis of
historical leakages from storage salt caverns well. Oil & Gas Science and Technology
- Rev. IFP Energies nouvelles 74, 27. doi: 10.2516/ogst/2018093
CMEO, 1993. MiIieu-effectrapport ondergrondse gasopslag te Norg.
DAGO, 2019. 20190903 DAGO Risico Matrix (QHSEP).
DBIGUT, 2017. The effects of hydrogen injection in natural gas networks for the
Dutch underground storages. Publication number: RVO-079-1701/RP-DUZA
DNV-GL, 2017. Verkenning waterstofinfrastructuur. In opdracht van het Ministerie

van Economische Zaken. Report no. OGNL.151886, Rev. 2.
Energystock, “Aanvraag Instemming Wijziging Opslagplan Zuidwending”, 2017.
Evans, D.J., 2008. An appraisal of underground gas storage technologies and

incidents, for the development of risk assessment methodology. Health and Safety
Executive. RR605 Research Report
Foh, S., Novil, M., & Rockar, P., 1979. Underground Hydrogen Storage Final
Report. Brookhaven National Laboratories, Upton, NY.
Gilette, L., & Kolpa, R.L., 2007. Overview of interstate hydrogen pipeline systems.
Argonne National Laboratory Environmental Science Division, Argonne (IL), p. 52
Gonzalez-Diez, N., van der Meer, S., Bonetto, J., & Herwijn, A., 2020. North Sea
Energy D3.1 Technical assessment of Hydrogen transport, compression, processing
offshore. As part of Topsector Energy: TKI Offshore Wind & TKI New Gas, NSE3-
D3.1
Hemme, C., & van Berk, W., 2017. Potential risk of H2S generation and release in
salt cavern gas storage. Journal of Natural Gas Science and Engineering, 47, 114-
123. doi: 10.1016/j.jngse.2017.09.007
Hemme, C., & van Berk, W., 2018. Hydrogeochemical Modeling to Identify
Potential Risks of Underground Hydrogen Storage in Depleted Gas Fields. Applied
Sciences, 8, 2282. doi: 10.3390/app8112282
Hyde, K., & Ellis, A., 2019. Feasibility of hydrogen bunkering. Interreg North Sea
Region Dual Ports.
H2Tools, 2020. Link
HyINTEGER, 2017. Long-term underground storage of hydrogen.

Klebanoff, L.E., Pratt, J.W., & La Fleur, C.B., 2016. Comparison of the Safety-
related Physical and Combustion Properties of Liquid Hydrogen and Liquid Natural
Gas in the Context of the SF-BREEZE High-Speed Fuel-Cell Ferry
Koelewijn, R., van Dam, M., & Hulsbosch-Dam, C., 2020. North Sea Energy D4.2
Report on offshore structural integrity and safety performance of H2 production,
processing, storage and transport. As part of Topsector Energy: TKI Offshore Wind
& TKI New Gas, NSE3-D4.2
Konikow, L. F., August, L. L., & Voss, C. I., 2001. Effects of clay dispersion on
aquifer storage and recovery in coastal aquifers. Transport in Porous Media, 43, 45–
64.
Kruck, O., Crotogino, F., Prelicz, R., & Rudolp, T., 2013. D(4) – Overview on all
known underground storage technologies for Hydrogen. HyUnder, D no. 3.1
Li, D., Zhang, Q., Ma, Q., Shen, S. Comparison of explosion characteristics
between hydrogen/air and methane/air at the stoichiometric concentrations,
International Journal of Hydrogen Energy, Volume 40, Issue 28, 2015, pages 8761-
8768
Liu JF, Xie HP, Hou Z, Yang CH, & Chen L., 2014 Fatigue damage evolution of
rock salt under cyclic loading. Acta Geotech 9(1):153–160
Lord, A.S., 2009. Overview of Geologic Storage of Natural Gas with an Emphasis
on Assessing the Feasibility of Storing Hydrogen. Sandia National Laboratories,
Albuquerque NM (SAND2009–5878).
Maliva R.G., 2020. Clogging. In: Anthropogenic Aquifer Recharge. Springer

Hydrogeology. Springer, Cham
Maytal BZ., Pfotenhauer J.M. (2013) The Joule-Thomson Effect, Its Inversion and
Other Expansions. In: Miniature Joule-Thomson Cryocooling. International
Cryogenics Monograph Series. Springer, New York, NY
Melaina, M.W., Antonia, O., & Penev, M., 2013. Blending Hydrogen into Natural
Gas Pipeline Networks: A Review of Key Issues. NREL.
National Research Council, 2008. Prospects for Managed Underground Storage

of Recoverable Water. Division on Earth and Life Studies, Water Science and
Technology Board, Committee on Sustainable Underground Storage of Recoverable
Water
NOGEPA OPCOM, 2016. Industry Standard No. 41 – Well Engineering and

Construction Process.
Opedal, N., Greenhalgh, E., Van der Valk, K., 2020. Current state-of-the-art
assessments and technical approach for assessment of well re-use potential and
CO2/brine leakage risk, Deliverable D2.1 of the REX-CO2 project.
Panfilov, M. (2010). Underground storage of hydrogen: In situ self-organisation

and methane generation. Transport in Porous Media, 85(3), 841-865.
doi:10.1007/s11242-010-9595-7
Pavelic, P., Dillon, P. J., Barry, K. E., Vanderzalm, J. L., Correl, R. L., & Rinck-
Pfeiffer, S. M., 2007. Water quality effects on clogging rates during reclaimed water
ASR in a carbonate aquifer. Journal of Hydrology, 334, 1–16.
Peterhead CCS project, 2016. Risk management plan & risk register. Doc No:
PCCS00-PT-AA-5768-00001, Date of issue: 19/01/2016, DECC Ref No: 11.023
Pudlo, D., Ganzer, L., Henkel, S., Kuhn, M., Liebscher, A., De Lucia, M., Panfilov,
M., Pilz, P., Reitenbach, V., Albrecht, D., Wurdemann, H., & Gaupp, R., 2013. The
H2STORE Project: Hydrogen Underground Storage – A Feasible Way in Storing
Electrical Power in Geological Media? In: Hou M., Xie H., Were P. (eds) Clean Energy
Systems in the Subsurface: Production, Storage and Conversion. Springer Series in
Geomechanics and Geoengineering. Springer, Berlin, Heidelberg. doi: 10.1007/978-
3-642-37849-2_31
Reitenbach, V., Albrecht, D., & Ganzer, L., 2014. Einfluss von Wasserstoff auf
Untertagegasspeicher; DGMK Research Report 752. Hamburg: DGMK
E. Ruigrok, J. Spetzler, B. Dost, and L, Evers, 2019. Veendam event, 09-01-02019.

Royal Netherlands Meteorological Institute (KNMI), Technical Report 373.
SODM, 2019. De integriteit van onshore putten in Nederland. https://www.sodm.nl
SPCS, SAFETY PLATFORM CARGO SURVEYORS (SPCS, 2020) – a Bureau

Veritas, Intertek, Saybolt and SGS cooperation and safety initiative. Document
reference: SPCS_004_2020_03_V4
Tarkowski, R., 2019. Underground hydrogen storage: Characteristics and

prospects. Renewable and Sustainable Energy Reviews, 105, 86-94. doi:
10.1016/j.rser.2019.01.051
Truche, L., Berger, G., Destrigneville, C., Guillaume, D., Giffaut, E., 2010. Kinetics
of pyrite to pyrrhotite reduction by hydrogen in calcite buffered solutions between 90
and 180 °C: implications for nuclear waste disposal. Geochimica et Cosmochimica
Acta 74, 2894–2914.
Truche L., Jodin-Caumon M., Lerouge C., Berger G., Mosser-Ruck R., Giffaut E.,
Michau N., (2013) Sulphide mineral reactions in clay-rich rock induced by high
hydrogen pressure. Application to disturbed or natural settings up to 250 °C and 30
bar. Chemical Geology 351, 217–228.
Uehera, I., 2013. Handling and Safety of Hydrogen, Energy Carriers and
Conversion Systems, vol. I
Van Unen, M., van der Valk, K., Koornneef, J., Brunner, L., & Koenen, M., 2020.
HEATstore risk assessment approach ECW case Middenmeer. TNO report.
Vlek, C, 2019. Rise and reduction of induced earthquakes in the Groningen gas
field, 1991–2018: statistical trends, social impacts, and policy change. Environmental
Earth Sciences (2019) 78:59. https://doi.org/10.1007/s12665-019-8051-4
Wang, M., D. Jia, A. Lin, L. Shen, G. Gang Rao, and Y. Li (2013), Late holocene
activity and historical earthquakes of the Qiongxi thrust fault system in the southern
Longmen Shan fold‐and‐thrust belt, eastern Tibetan Plateau, Tectonophysics, 584,
102– 113
Weidner, E., Honselaar, M., Ortiz Cebolla, R., Gindroz, B., & de Jong, F., 2016.
Sector Forum Energy Management / Working Group Hydrogen. CEN – CENELEC.
doi: 10.2790/66386
Appendix 1 – Risk Inventory
Please find below the full risk inventory. The Excel file of the risk inventory can be
shared on request by sending an email to:
The following pages will show the inventory in the order of the tabs that are in the
Excel file, and which also include how the inventory is set-up and could be used.
Tabs:
Risk Inventory
a. Readme
b. Input
1. Pre-Execute
2. Execute
3. Operate
4. Decommission
5. Post Abandonment
6. All Phases
Review sheet
References
Revision control
TAB: Risk Inventory LSES
TNO Risk Inventory for Subsurface Energy Storage in Salt Caverns and Depleted Fields Version 1.0
General description
This risk inventory for subsurface energy storage projects has been produced by TNO in the context of the TKI LSES project. It details risks associated with the
subsurface energy storage technologies CAES, hydrogen in salt caverns and hydrogen in depleted fields.
It is compiled from risks found in literature, supplemented with expertise from partners in the project consortium. References used can be found listed in the
'References' tab, the reference is numbered to be able to trace back the risks in this sheet to the literature. It is suggested to use this as an inventory from which the
most relevant risks for a particular project can be identified. This procedure has been successfully used for a similar risk inventory as part of the Dutch Heatstore
demonstration case, the method is described in Van Unen et al., 2020, HEATSTORE risk assessment approach for HT-ATES applied to demonstration case Middenmeer,
The Netherlands. 15 pp. (reference 39).
Authors: M. van Unen, K. van der Valk and R. Groenenberg

Contributors: L. Brunner
Project Manager: M. Halter
DISCLAIMER: This risk inventory is based on risks and mitigations that are found in literature. Some of the risks are a combination of multiple references or
interpretations of risks that are found in literature. The mitigations in this inventory are found in literature and are supplemented by the team. Please refer back to
the references if anything is unclear. The inventory of risks and associated mitigations is not necessarily complete and can be used as a starting point in identifying
the most relevant risks for a project. Using this risk inventory does not replace a dedicated risk assessment workshop with the required expertise.
All rights reserved.

No part of this publication may be reproduced and/or published by print, photoprint, microfilm or any other means without the previous written consent of TNO.
In case this inventory was drafted on instructions, the rights and obligations of contracting parties are subject to either the General Terms and Conditions for
commissions to TNO, or the relevant agreement concluded between the contracting parties. Submitting the report for inspection to parties who have a direct interest
is permitted.
© 2020 TNO
Please cite this inventory as: van Unen, M et al., 2020b: Risk Inventory for large scale subsurface energy storage
Start using this Risk Register by making a separate copy of the file before adjusting it, then please go to sheet a. 'Readme' to understand how the sheet works.
Go to sheet a. Readme
a TNO – Applied Geosciences, Princetonlaan 6, Utrecht 3584 CB, The Netherlands

TAB a: Readme
Readme
This Readme is prepared to make it easier to understand how this Risk Register is set-up. Below definitions for the structure of the risks has been defined (TEECOPS, project
phases, risk ratings, system components and storage types). Tab b. 'Input' gives the option to define the project. Tabs 1. to 6. are the core of the risk register; they contain the risks
and allow for ranking of the risks (both unmitigated and mitigated). The risk ranking (color code) will automatically follow from what is chosen as likelihood and as consequence
rating.
Filtering:
In the risk register it is possible to filter on application; storage facility. This can be done by clicking the drop-down button in the 'Storage type' cell and selecting only the relevant
storage types (and blanks). If you are interested in any storage (CAES and hydrogen) in salt caverns, please select: blanks + CAES in salt caverns + Hydrogen in salt caverns +
Hydrogen storage + Storage in salt caverns + All subsurface storafe reservoirs. If you want to only have risks relevant for Hydrogen in depleted reservoirs, please select: blanks +
Hydrogen in depleted fields + Hydrogen storage + All subsurface storage reservoirs.
Alternatively, one could filter on the risks earmarked with relevant TEECOPS category by clocking the dropdown button in any of the blue coloured TEECOPS cells and only select
the category.
TEECOPS Definitions
TEECOPS Definitions1
T Technical (Sub)surface, Infrastructure, Technology, Operability, Availability, Integrity, Sustainability, Maintenance
Ec Economical Life-Cycle Cost, Phasing, Valuation method, Capacity, Economic model, Regret costs
En Environmental Surface exposure, Subsurface environment
C Commercial Contracting & Procurement, Financing, Business controls, Legal, Terms & Conditions, Competition, Marketing, Liabilities, Collaboration Agreement
O Organisational Structure, Resources, Procedures, Project Controls, Knowledge Management, Systems & IT, Interfaces, Partners, Governance
P Political Government, Stakeholders, Employment, Regulation, Security, Reputation, NGOs, Export Control, Localisation
S Societal Community, Public opinion, Social License to Operate
Project phase definitions
Project workflow
5. Post abandonment
1. Pre-Execute 2.Execute 3. Operate 4. Decommisioning
monitoring
Analysis Analysis Detailed design Execute

(incl. concept Detailed design
selection)
Project work flow phases; Risk associated with the underground storage of hydrogen and CAES (excluding UGS and CO2) during
1. Pre-execute All work done prior to the start of the execution phase; including analysis and design
2. Execute The Execution phase; in this phase the facility is built (or updated) for energy storage
3. Operate The operational phase; the actual phase where energy is stored and produced
4. Decommission The Decommissioning phase; this includes the abandonment of wells, removal of the surface facilities and clearing the site for future use
5. Post-abandonment The post decommisioning phase; these include risks that could come to light by monitorring of the abandoned site
6. General All of the above defined project phases (to prevent having them in all phases)
Risk rating
Risk rating
Probability
Low Medium High
Consequence
Low L L M
Medium L M H
High M H H
System components
System component definition

General Risks that are relevant for all (or multiple) of the system components
Surface Facilities These include compressors, piping, instrumentation, process facilities
Well This includes the X-mas tree, wellhead, well (completion and cemented casings), sand-face completion
Subsurface (reservoir) The target storage reservoir, the caprock and overburden
Project specific Any risks that are project specific and probably not relevant for (most) other projects
Storage types
Hydrogen in salt cavern

Hydrogen in depleted gas fields
CAES in salt caverns
Storage in salt caverns
Hydrogen storage
All subsurface storage reservoirs
1
These definitions are based on reference 38 from the reference list; Risk management plan for the Peterhead project
TAB: b. Input
Input (project definition)

Date: 02 July 2020
Risk assessor: LSES consortium
Project name: TKI LSES
Project type: TKI
Type of Energy Storage:
TAB: 1. Pre-Execute
Date last modified: Risk assessor: Project: Type of energy storage Version
Risk category
14 September 2020 LSES consortium TKI LSES 1
1. Pre-execute (scoping, analysis, concept select, detailed design)
Probability of Unmitigated Mitigated Mitigated prob. Mitigated
Risk ID Risk description Storage type Reference T Ec En C O P S Consequence Mitigations Comments
consequence Risk rating consequence of consequence risk rating
T Ec En C O P S General
Unsuitable contracts (roles and responsibility
All subsurface storage
PE-G1 not clearly defined) leading to suboptimal 28 T Ec O - Select experienced and suitable management
reservoirs
performance or exploding costs
PE-G2 Demand analysis and forecast are inaccurate 28 T Ec C S - Get a good overview on the demand and forecasts and estimate uncertainties
reservoirs
- Thorough feasibility study including risks
PE-G3 Lack of financing for the executing phase 28 Ec O - Thorough cost management
reservoirs
- Thorough analysis of funding opportunities
- Prepare and execute communication and participation plans
Low social acceptance for hydrogen storage
PE-G4 Hydrogen storage 28 Ec C O P S - Early inclusion of stakeholders in decision making
stops project
- Stakeholder analysis/mapping
- Prepare and execute communication and participation plans
Low social acceptance for CAES storage stops
PE-G5 CAES in salt caverns 28 Ec C O P S - Early inclusion of stakeholders in decision making
project
- Stakeholder analysis/mapping
Organization is not experienced / financially All subsurface storage - contactor / investor shall hire additional proper external experts (domestic, foreign)
PE-G6 2 O
robust enough for the challenge reservoirs the projects
PE-G7
T Ec En C O P S Surface facilities
Inappropiate/inadequate surface All subsurface storage - Design with flexibility
PE-S1 1 T
technologies design reservoirs - Detailed design surface facilities after well test (also postponing start date)
PE-S2
T Ec En C O P S Well
- material selection (suitable for gas with anticipated composition),
- casing/ tubing connection
PE-W1 Inappropiate/inadequate well design 1 T - loading applied on the well during operations (P, T, stress)
reservoirs
- well verification (how the existing well envelope was tested? what T-range? are
tests applicable to H2?)
PE-W2 Inappropriate packer selection / installation 1 T - material, elastomer selection, well loading
reservoirs
Inappropriate wellhead / x-mas tree selection All subsurface storage
PE-W3 1 T - elastomer selection, steel selection (esp. if H2S present), pressure rating
/ installation reservoirs
PE-W4 Inappropriate SSSV design 1 T - material, tightness against H2
reservoirs
All subsurface storage - “hydrogen” tightness, potential (corrosive) reaction to cement, other chemical
PE-W5 Inappropriate cement design/specs 1 T
reservoirs reaction that could reduce sealing performance, cycling pressure
PE-W6 Inappropriate completion design 1 T - material (especially if H2S present), suspension fluid
reservoirs
- robustness of primary and secondary envelope e.g. if tubing is leaking
PE-W7 Integrity control not failsafe 1 T - functioning of primary and secondary barrier during operations vs. during
reservoirs
drilling/workover, well control barrier
T Ec En C O P S Reservoir (subsurface)
Not able to find a suitable storage site in the All subsurface storage - Careful site characterization
PE-R1 26 T Ec
area of interest reservoirs - Drill extra wells
- Gather more data (2D or 3D seismic)
Insufficient knowledge on subsurface site All subsurface storage
PE-R2 10 T Ec En C O P S - Look at offset wells (if available)
characteristics, leading to higher risks reservoirs
- Drill additional exploration wells
- Good reservoir thickness and petrophysical properties are required.
- Reservoirs need good connected porosities and high permeabilities
- Reservoir should be positioned at a depth allowing for a wide pressure range for the
Unfavourable subsurface conditions lowering All subsurface storage
PE-R3 7 Ec applicable and approved minimum and maximum injection and withdrawal pressures.
the economic value of the storage site reservoirs
- Extensive reservoir characterisation and reservoir modelling needs to be performed
- Host rock must be strong enough to enable the construction of a self-supporting
cavern
PE-R4
T Ec En C O P S Project specific
PE-P1
TAB: 2. Execute
Risk category
2. Execute (incl. well test and injectivity test)
Unanticipated delays and costs (materials, services, All subsurface storage
E-G1 28 T Ec O - Include time/cost buffer in the planning
maintenance) reservoirs
Roles and responsibility not clearly defined leading to All subsurface storage
E-G2 28 T Ec O - Select experienced and suitable management
suboptimal performance or exploding costs reservoirs
- Design fit for urban environment
Over expenditure on CAPEX (CAPEX overrun) because All subsurface storage
E-G3 29 Ec C O - Tender strategy fit for market supply
of unforeseen costs or unfavourable tender reservoirs
- Detailed design narrow down the uncertainty range
- In order to have a continuously active data transfer, two communication connections will
Interruptions in signal transfers due to failures or All subsurface storage
E-G4 18 T O be needed. One of the two connections functions as a backup, with functionality to switch
maintenance reservoirs
over automatically if the primary connection is interrupted.
E-G5 Lack of financing for next phases 28 Ec C O P S - Thorough cost management
reservoirs
- Evaluate the cap rock layers and sealing capacity.
- The spill point of the targeted structure and any flow must be determined
Contamination of groundwater due to any type of All subsurface storage
E-G6 2; 7; 12 T Ec En C P S - Leakage along fractures must be excluded
leakages or emissions reservoirs
- Monitoring water levels and water chemistry in observation wells completed above the
cap rock.
E-G7
Failure of above ground infrastructure (valve, pipes, All subsurface storage - Check and repair above ground infrastructure
E-S1 1; 9 T
wellhead or compressor units) reservoirs - Fit for purpose design
- Early involvement of contractors/experts
- Follow standards and guideline
E-S2 General surface facility construction risks 1 T - Best Practices
reservoirs
- QA/QC
- Collapse of infrastructure
Contribute to the instability, breaching and collapse of
solution mined salt caverns
Might lead to problems with cavern construction,
producing unstable or poorly shaped and inefficient voids
Failure of facilities due to uncontrolled leaching During leaching the rock mechanical limitations must not be exceeded, otherwise the
E-S3 Storage in salt caverns 9 T for gas storage. Problems might occur if, for example,
during cavern construction strategy must be modified which will also lead to a lower cavern volume.
more soluble evaporitic horizons (e.g. potash) are
present within the bedded salt, unexpectedly thick non-
halite interbeds are present or wet rockhead is
developed
Malfunction of the control panel that is connected to
All subsurface storage - Additional wires for most risky connections
E-S4 the transformer facility leading to an interruption in 25 T
reservoirs - Additional transformers that can step in when needed
the electricity cycle
E-S5
E-W1 Not able to lower the casing string 28 T - Ensure safe clearance and drift diameter of the well
reservoirs
E-W2 Trajectory issues (deviation from target) 28 T En - Thorough Drill Plan/Program and its execution
reservoirs
Drilling is more complicated/more expensive than All subsurface storage
E-W3 28 T Ec - Thorough Drill Plan/Program and its execution
anticipated reservoirs
Issues in transporting/handling radioactive sources All subsurface storage - Radioactive waste management plan
E-W4 28 T En
for logging reservoirs - Applying radiation safety protocols
Standard drilling risks also common to O&G
- Drill according to newest lessons learned
operations (e.g. rig issues, failure of the well or All subsurface storage
E-W5 1 T - Early involvement of contractors and experts
casing, hitting over pressured layer, shallow gas reservoirs
- state of the art drilling program
pockets, getting stuck, losses, losing circulation)
Geostatic (isostatic) stress increase in salt during
injection of water for creating the cavern (which can
E-W6 Storage in salt caverns 5;7 T - Pressure monitoring of the subsurface
lead to fracturing and loss of cavern integrity or
induced seismicity)
The injection of the brine during each cavern - Good monitoring and control of pressure changes in the cavern
E-W7 emptying-filling cycle could dissolve the salt wall, Storage in salt caverns 9; 25 T - Careful site characterization (subsurface mapping etc. using high resolution seismic
causing structural weakness and ultimately failure reflection data for example)
Damage of the wells due to seismicity and collapse of All subsurface storage - Pressure monitoring of the subsurface
E-W8 5; 7 T Ec En
the subsurface structures reservoirs - Seismic monitoring
E-W9
TAB: 2. Execute (continued)

Reservoir (subsurface)
- Training and certifying of the personnel
Wrong choice of fluids or techniques damaging the All subsurface storage
E-R1 28 T En - Select Experienced and suitable Management
reservoir reservoirs
- Thorough geological survey/core sample analysis
- Adaptation of the drill path to reach multiple targets
- Avoid excessive contamination of the well
- Use of clay-mineral free drilling mud
- Avoid the use of loss control material during drilling of the production section
- Avoid the cementing of previous casing string in the production section
- Try to drill long enough production section for securing the expected yield
- Use of external casing packer between the loose formation and the productive layer
Flow rate lower than expected during flow test (e.g.
Hydrogen in depleted - In case of porous reservoir use of under reaming and gravel pack in the production
E-R2 because of lower permeability, heterogeneity of the 28 T
gas fields section
reservoir)
- Accurate collection and interpretation of productivity data of wells for securing
information for the expected yields
- Doing new measurements in existing wells for securing information for the expected
yield
- Dedicated exploration well
- Update design and include more sources to increase supply
- Include potential extra wells in risk margin for project
Fluid chemistry / gas content / physical properties are - Adapt the material selection to the chemical/physical properties of the fluid
E-R3 different than expected, which can alter the well or 28 T En - Additional chemical sampling and hydrogeological analyses
reservoirs
change the composition of the injected energy - Re-evaluate hydrogeological model

- Accurate collection an interpretation of expected geology to provide information on the
Well ends up in a non-suitable part of the salt layer target reservoir
E-R4 Storage in salt caverns 28 T Ec
during the cavern build-up - Doing new surface geophysical measurements for the better understanding of expected
geology
- Drilling further, side-tracking
- Accurate collection an interpretation of expected geology for securing information on
Geological lithology or stratigraphy is different than All subsurface storage
E-R5 28 T the target reservoir
expected reservoirs
- Doing new surface geophysical measurements for the better understanding of expected
geology for securing information on the target reservoir
- Adapt the power plant design under given temperature/pressure
(Re-)injection of the stored product is more difficult All subsurface storage
E-R6 28 T - Adaptation of the drill path to reach multiple targets
than expected reservoirs
- In case of porous aquifers use of under reaming and gravel pack in the production
section
- Avoid extreme overpressure drilling

E-R7 Mud losses leading to severe technical issues 28 T - Proper composition and parameters of drilling fluid /mud program
reservoirs
- Detailed subsurface analysis to identify potential loss zones and adapt drilling program

- Accurate collection an interpretation of expected geology to provide information on the
Hydrogen in depleted target reservoir
E-R8 Well misses target formation 28 T Ec
gas fields - Doing new surface geophysical measurements for the better understanding of expected
geology
- Drilling further, side-tracking
E-R9 Induced seismicity (e.g. during drilling or stimulation) 28 T Ec En S - Installation of seismic monitoring system
reservoirs
Presence of faults or fractures in the storing All subsurface storage - Geological explorations, determine the presence of faults/fractures, the sealing capacity
E-R10 9 T Ec
sequences, which can lead to leakages or losses reservoirs of the faults and the presence of halite beds.
E-R11 Leakage of subsurface fluids due to seismicity 2; 7 T Ec En C O P S - Find locations which benefit from a second sealing formation above the storage
reservoirs
Subsidence due to seismicity and collapse of the All subsurface storage
E-R12 5; 7 T Ec En - Pressure monitoring of the subsurface
subsurface structures reservoirs
Brine pressure increase will result in a figure larger

than the geostatic pressure, leading to
E-R13 Hydrogen storage 11 T Ec En - Careful pressure monitoring
hydrofracturing, which can lead to leaking (water
pollution), cavern / reservoir collapse and subsidence
Once a salt cavern is penetrated there is a risk for fast - Careful penetration during drilling together with subsurface monitoring
E-R14 deterioration of the salt cavern, because Storage in salt caverns 26 T O - It is recommended to limit the cavern height to diameter ratio so that it would not
of their large widths and thin heights. exceed 5.0
E-R15
E-P1
TAB: 3. Operate
Risk category
3. Operate

- Evaluate the cap rock layers and sealing capacity.

Contamination of groundwater due to any All subsurface storage - The spill point of the targeted structure and any flow must be determined
O-G1 2; 7; 12 T Ec En C P S
types of leakages or emissions reservoirs - Leakage along fractures must be excluded
- Monitoring water levels and water chemistry in observation wells completed above the cap rock.

O-G2 Malfunction of technical operating system 6 T - Measurement of mass flow and volume, pressure, temperature and pumping rate
reservoirs
- Running down-hole logs in the injection/withdrawal wells (e.g., temperature logs to identify
thermal anomalies, neutron logs to detect gas behind the casings above the storage formation,
Leakage of gas from well or wellhead leading to cement bond logs to determine the integrity of the cement behind the casings, calliper logs to find
O-G3 Hydrogen storage 11; 12 Ec En
explosions, fires or suffocation mechanical breaks in the casings, noise logs to “listen” to the flow of gas behind the casings, and a
variety of casing inspection logs to determine the thickness and general condition of well casings
and joints)
Unanticipated delays and costs during
O-G4 operations and maintenance (reparation of 28 T Ec O - Include time/cost buffer in the planning
reservoirs
materials, services)
- Carefully monitor the well head
Incident that leads to rip off of the very robust All subsurface storage
O-G6 7 T Ec En - Blow-out can be prevented by an automatically closing subsurface safety valve, installed some
well head with its multiple safety installations reservoirs
meters below the well head
Too high heat generation when compressing air - Careful temperature monitoring
O-G7 CAES in salt caverns 9 T O
due to a malfunction in heat exchangers - Emergency stop system
- Significantly reducing uncertainty around encountering less favourable hydrologic properties than
Too little income generated to pay back the All subsurface storage expected at the site
O-G8 16 Ec O
loan reservoirs - Reducing the pressure required to achieve the targeted injection rate and increasing the air
storage volume efficiency in the structure.
Too little income generated to pay back the All subsurface storage
O-G9 28 Ec O - Thorough cost management
loan reservoirs
- In order to have a continuously active data transfer, two communication connections will be
Interruptions in signal transfers due to failures All subsurface storage
O-G10 18 T needed. One of the two connections functions as a backup, with functionality to switch over
or maintenance reservoirs
automatically if the primary connection is interrupted.
- Careful site management
- Determine potential flow paths
Economic risks (e.g. larger amounts of hydrogen All subsurface storage
O-G11 20 Ec O - Do injection tests when exploration wells are drilled.
gets trapped, leakage of the energy source) reservoirs
- Pipe storage should be located at an appropriately safe distance from surrounding buildings and
settlements
- Keep continuous monitoring of standards, technologies and political situation
- LPG storage withdrawal operations have optimum saturated brine injection rates to prevent
All subsurface storage excessive tubing vibration.
O-G12 Nuisance / emissions such as noise, dust, light 7 Ec O P S
reservoirs - Maximum noise levels and noise plan (day night rhythm)
- Insulation
- Early involvement of neighbourhood
O-G13
TAB: 3. Operate (continued)

Geostatic (isostatic) stress increase in salt
O-W1 during injection of H2 or air (this can lead to Storage in salt caverns 5;7 T - Pressure monitoring of the subsurface
fracturing or seismicity)
Excessive tubing vibrations can lead to well
failure, but can be mitigated by a good design All subsurface storage
O-W2 13 T En S - a good design and keeping a safety margin on the speed at which air or H2 is injected/withdrawn.
and keeping a safety margin on the speed at reservoirs
which air or H2 is injected/withdrawn.
Corrosion of pipelines and components
- Material selection and design principles fit for expected potential corrosion mechanisms
(buried), flanges, valves, fittings, pressure All subsurface storage
O-W3 9 T - Corrosion avoidance by the injection of dry air between steel and the casings
vessels, pumps, compressors and injectors, reservoirs
- Strict monitoring of minimum pressures
wells and their casings/cements
Erosion of pipelines and components during All subsurface storage - Well checks and maintenance
O-W4 1 T
hydrogen of CAES storage reservoirs - Check erosional velocities acceptable for the equipment and base operating strategy on it
- Following analysis and inspections (mechanical integrity test, cement logs, mechanical
assessment of the stress/strain experienced when at atmospheric pressure).
O-W5 Damage to production wells 3; 7; 9; 20; 22 T Ec En - Careful check of the blowout-prevention system
reservoirs
- Installing subsurface safety valves. Safety shut-off valves are installed in gas caverns about 50 m
below the surface.
Thermal stress of the salt during high - Careful monitoring of the pressure and temperature changes in the subsurface.
O-W6 CAES in salt caverns 15 T
withdrawal rates of compressed air - Careful monitoring of the withdrawal velocities.
Failure in cementation leading to permeable All subsurface storage - During operation the rock mechanical limitations must not be exceeded, otherwise the strategy
O-W7 20 T En
cements reservoirs must be modified which will also lead to a lower cavern volume.
The energy released versus time by a blow-out during storage in
depleted gas fields is smaller than for a cavern blow out because
common well diameters for porous storages are smaller. Because of
Failure of the Subsurface Safety Valve during - Make sure the safety valve is working well, e.g. preventative maintenance the commonly much larger inventory of aquifer storages the duration
O-W8 Storage in salt caverns 7 T En
storage in salt caverns (leading to a blow-out) - Make a safety assessment in the case of a blow-out of the blow-out could last longer. It is impossible for the gas stored in
the aquifer to ignite within the formation itself because of the
absence of oxygen which completely prevents the formation of a
combustible mixture
The energy released versus time by a blow-out during storage in
depleted gas fields is smaller than for a cavern blow out because
common well diameters for porous storages are smaller. Because of
Failure of the Subsurface Safety Valve during
Hydrogen in depleted - Make sure the safety valve is working well, e.g. preventative maintenance the commonly much larger inventory of aquifer storages the duration
O-W9 the storage in depleted gas fields (leading to a 7 T En
gas fields - Make a safety assessment in the case of a blow-out of the blow-out could last longer. It is impossible for the gas stored in
blow-out)
the aquifer to ignite within the formation itself because of the
absence of oxygen which completely prevents the formation of a
combustible mixture
O-W10 Leakage along the well through the cements 9 T Ec En - Inject/squeeze cements to fill up the leakage paths
reservoirs
Cyclic loading of the wells used for both
injecting as "producing" introduces risk of All subsurface storage
O-W11 1 T - Well design and material selection fit for purpose with respect to cyclic loads
fatigue loads for the steel and cement (e.g. reservoirs
cycles: inject - idle - produce - idle - inject)
- Following analysis and inspections (mechanical integrity test, cement logs, mechanical
O-W12 Uncontrolled gas release (blowout) 3; 7; 9 T Ec En C O P S assessment of the stress/strain experienced when at atmospheric pressure).
reservoirs
- Install an automatically closing subsurface safety valve some meters below the well head (SSSV)
O-W13
Subsidence associated with pressure loss in the
O-R1 Storage in salt caverns 9 T - Careful monitoring of pressure and changes in surface characteristics
cavern
- Monitor the pressure and temperature in the reservoir
- Ensure that the composition of natural gas and air remains outside the ignition envelope
- Determine the flammability limits
- Determine the heat provided by the reservoir
O-R2 Underground fire or explosion CAES in salt caverns 4 T Ec En
- Determine local failures of the reservoir, which could produce rapid pressurization
- The geologic conditions and geometry of the underground storage facility must be investigated
(e.g. density differences between the natural gas and air, and permeability contrasts could
influence the mixture between the natural gas and air.
- Passive seismic Monitoring
- Cavern stability assessment
- Determine and monitor operating pressures
- Estimate geostatic stress in salt at the depth of the cavern
Rock salt creeps and begins to deform when
O-R3 Storage in salt caverns 5; 7; 9 T En - Long term stability of the cavern is ensured by many years of experience and by site-specific lab
affected by high formation pressures.
testing in combination with rock-mechanical models
- Groundwater control method, which maintains the internal cavern pressure below the natural
prevailed water pressure within the rock
- Pressure and mechanical integrity testing wells and cavern

- Passive seismic Monitoring
- Cavern stability assessment
- Determine and monitor operating pressures
- Estimate geostatic stress in salt at the depth of the cavern
Geostatic (isostatic) stress increase and cavern
O-R4 Storage in salt caverns 5; 7; 9 T Ec En - long term stability of the cavern is ensured by many years of experience and by site-specific lab
volume loss
testing in combination with rock-mechanical models
- groundwater control method, which maintains the internal cavern pressure below the natural
- Pressure and mechanical integrity testing wells and cavern
- Lowering storage pressures within the storage reservoir or cavern
- Accumulated gas in shallow traps can be pumped out to prevent further migration and surface
releases
O-R5 Leakage during hydrogen storage in salt caverns Hydrogen storage 5; 9 T Ec En P S
- Monitoring of injection/production/abandoned wells to detect damage or leakage
- Repair of leaking injection/production/abandoned wells
- Quantify health and ecological risk
Daily temperature variations can lead to

compression and decompression of - Careful with withdrawal rates
O-R6 CAES in salt caverns 5; 7 T
compressed air (possibly leading to generation - Commissioning of cavers/pressure testing
of tensile fractures in the reservoir).
All subsurface storage - Usage of 5D QM Storage Monitor, which is permanently recording the seismo-acoustic background
O-R7 Reservoir seismicity 6 T En
reservoirs and any signal and information contained therein.
Occurrence of non-halite interbeds leading to
O-R8 Storage in salt caverns 7 T Ec - Increase geological knowledge by exploration
reduced cavern size
Occurrence of faults leading to different All subsurface storage
O-R9 7 Ec - Increase geological knowledge by exploration
operational pressure range reservoirs
- Exploration of geological structures
Slow uncontrolled leakage of gas through the All subsurface storage - The cover rock needs to separate the reservoir from shallow structures to prevent gas escape
O-R10 7 Ec En
subsurface reservoirs - Groundwater control method, which maintains the internal cavern pressure below the natural
- In a brittle collapse zone, water ingress could lead to ongoing dissolution but facilitate relatively
Subsidence characterised by predominantly rapid creep of adjacent salts (as a result of the presence of water), back into the dissolution cavity.
O-R11 Storage in salt caverns 9 T Ec En
brittle (collapse) deformation This could conceivably help to prevent the formation of large cavities and subsequent catastrophic
collapse.
- Good monitoring and control, so that undetected enlargement of the cavern and inaccuracies in
storage volumes can be indicated.
Overfilling of caverns resulting in cavern
O-R12 Storage in salt caverns 9 T - Accurate inventories of stored or injected product
instability and collapse
- the salt cavern operations are mainly restricted by the maximum pressure change rate per unit
time to ensure stability, and also by the maximum flow velocities inside the well
Release of stored product through cracks in the - Careful site characterization (subsurface mapping etc. using high resolution seismic reflection
cavern wall data for example)
(Partial) cavern roof collapses (loss of cavern
integrity) leading to thinning of the ‘protective’ - Careful site characterization (subsurface mapping etc. using high resolution seismic reflection
cavern roof salt associated with failure of the data for example)
cavern seal
- Good monitoring and control of pressure changes in the cavern
Collapse of internal ledges or benches formed
O-R15 Storage in salt caverns 9 T Ec En - Careful site characterization (subsurface mapping etc. using high resolution seismic reflection
by non salt interbeds
data for example)
Microorganisms could speed up geo-chemical
reactions
leading to corrosion and acidification of the
- Monitor the quality and the amounts of the product
reservoir fluids through accumulation of Hydrogen in depleted
O-R16 23 T Ec En - Performing basic microbiological analyses of the reservoir environment followed by laboratory
hydrogen sulphide (H2S) and organic acids. gas fields
experiments to estimate the kinetics of nutrient/energy usage and by-product formation.
Enhanced microbial activity can also lead to a
potential decrease in storage capacity and
production of CO2
Decrease of permeability through iron sulphide
Hydrogen in depleted
O-R17 (FeS) precipitation, biofilms and extra-cellular 23 T Ec En - Monitor the quality and the amounts of the product
gas fields
substances
Reduction of gas quality through hydrogen
Hydrogen in depleted
O-R18 consumption and hydrogen sulphide (H2S) 23 T Ec En - Monitor the quality and the amounts of the product
gas fields
formation
Trigger geo-chemical reactions with rock
minerals and reservoir fluids during the storage
of hydrogen (could lead to energy loss and - Porosity and Permeability monitoring of the reservoir
O-R19 Hydrogen storage 23 T Ec En
other undesirable effects such as damage in the - Subsurface monitoring of the pressure and temperature
rock and mineral structure leading to alteration
of reservoir properties).
Undesired water suction during large well - To investigate the risk of water suction in CAES reservoirs, the critical flow rate and interface
O-R20 CAES in salt caverns 26 T
penetrations height can be determined

- The influence of other gas components like carbon monoxide and oxygen must be considered
Hydrogen admixture leading to undesired
O-R21 Hydrogen storage 23 T Ec - Must consider all boundary conditions like pressure, partial pressure, temperature, water or
energy losses
condensate content, geo-chemical context, etc.
Loss of cavern capacity as a consequence of
O-R22 having operated at pressures too low to Storage in salt caverns 9 T Ec - Careful monitoring of pressure
maintain cavern walls
The stored product is dissolved into the water
Hydrogen in depleted - Risk analysis of geological storage facilities
O-R23 of the caprock, whereupon it is not recoverable 9; 23 Ec En
gas fields - Determine significance of risk
and a valuable commodity is lost.
- Perform regular sonar scans to accurately map the cavern walls and monitor closure
Leakage from the cavern (through the well)
O-R24 Storage in salt caverns 9 T Ec En C P S - The installation, during cavern construction, of a remote second well system that would allow
during salt cavern storage
safe draw down of the stored product in the event of an incident
All subsurface storage - Careful site management (e.g. land ownership, proximity to critical infrastructure (natural gas
O-R25 Unexpected hydrogeologic conditions 16 T
reservoirs pipeline and transmission), and nearby exploration wells)
- Lowering storage pressures within the storage reservoir or cavern
- Accumulated gas in shallow traps can be pumped out to prevent further migration and surface
Leakage during hydrogen storage in depleted Hydrogen in depleted releases
O-R26 5; 9 T Ec En C P S
gas fields gas fields - Monitoring of injection/production/abandoned wells to detect damage or leakage
- Repair of leaking injection/production/abandoned wells
- Quantify health and ecological risk
Presence of anomalous zones (higher solubility
or porosity) in what has been assumed to be
O-R27 Storage in salt caverns 9 T Ec En - Adequate subsurface characterization by wells, seismic etc.
homogenous salt, which includes leaky
interbeds
The stored product migrates away from the
All subsurface storage - Risk analysis of geological storage facilities
O-R28 storage area, whereupon it is not recoverable 9 Ec En
reservoirs - Determine significance of risk
and a valuable commodity is lost.
O-R29 Leakage through the caprock/ overburden 9 Ec En - Adequate cap rock characterization
reservoirs
Isostacy can change the pressure conditions of All subsurface storage
O-R30 14 T Ec - Careful pressure monitoring and changes in subsidence
the reservoir reservoirs
Cooling, contraction and thermal stressing of
- Careful with withdrawal rates
O-R31 salt at cavern walls (can lead to generation of CAES in salt caverns 5; 7 T
- Commissioning of cavers/pressure testing
tensile fractures).
Small leaks developing along the contact All subsurface storage
O-R32 20 T Ec En - Monitoring can continuously check the stability and integrity of the sealing structure
surfaces of the sealing structure reservoirs
Piezoelectricity generating an electric potential
when specific stress/strain conditions are - Monitor the cycles stress conditions near the borehole
O-R33 applied (cycled stress conditions especially 4 T - Purge the reservoir of natural gas before use (site-specific analyses).
reservoirs
near boreholes, may facilitate this - Install an in-situ gas monitor down hole to provide near source measure of natural gas presence
phenomenon).
Static electricity caused by the proximal
position of two materials with an imbalance of - Monitor the build up of charge by particles
O-R34 positive and negative charges. This could CAES in salt caverns 4 T - Monitor the natural gas content to adjust the air-fuel ratio in the gas turbine
provide an ignition source for the fuel-air - Grounding of all piping
mixture.
- Drilling of monitoring wells in the perimeter of the formation and the overlying cap rock. These
Gas leakage through unknown fractures or the Hydrogen in depleted
O-R35 20 T Ec En are measures that are done additionally to wells which monitor the gas to liquid interface within
spill point or other potential leak paths. gas fields
the storage range of the formation.
Thermal expansion of the cavern fluid, leading
O-R36 Storage in salt caverns 9, 11 T Ec - Continuous temperature and pressure monitoring of the cavern
to pressure build-up in the cavern
O-R37
O-P1
TAB: 4. Decommission
Risk category
4. Decommission (including post decommission phase)
Contamination of groundwater due to any All subsurface storage
D-G1 2 T Ec En P S - Modelling of leakage and monitoring of the well
type of leakages or emissions reservoirs
Leakage of the stored product in salt - On abandonment, closure and monitoring of salt cavern stability and
D-G2 caverns during the decommissioning Storage in salt caverns 9 T Ec En internal pressure to prevent over pressurization and possible failure of he
phase walls or roof rock and the wellhead/valves
Leakage of the stored product in a
Hydrogen in depleted gas - On abandonment of depleted fields, withdrawal of injected stored gas to
D-G3 depleted gas fields during the 9 T Ec En
fields below cushion gas levels, as undertaken in some decommissioned facilities
decommissioning phase
- Running down-hole logs in the injection/withdrawal wells (e.g.,
temperature logs to identify thermal anomalies, neutron logs to detect gas
behind the casings above the storage formation, cement bond logs to
Leakage of gas from well or wellhead
D-G4 Hydrogen storage 11; 12 Ec En determine the integrity of the cement behind the casings, calliper logs to find
leading to explosions, fires or suffocation
mechanical breaks in the casings, noise logs to “listen” to the flow of gas
behind the casings, and a variety of casing inspection logs to determine the
thickness and general condition of well casings and joints)
Unanticipated delays and costs in
D-G5 decommission operations (materials, 28 T Ec O - Include time/cost buffer in the planning
reservoirs
services, maintenance)
Postponing decommissioning because of
All subsurface storage - Have clear cut-off point and decommissioning moment based on equipment
D-G6 economical attractiveness increasing 1 T Ec O S
reservoirs properties
difficulty of decommissioning
- In order to have a continuously active data transfer, two communication
Interruptions in signal transfers due to All subsurface storage connections will be needed. One of the two connections functions as a
D-G7 18 T O
failures or maintenance reservoirs backup, with functionality to switch over automatically if the primary
connection is interrupted.
D-G8 Lack of financing for next phases 28 Ec C O S - Thorough cost management
reservoirs
- Significantly reducing uncertainty around encountering less favourable
Financial risk during the decommissioning All subsurface storage hydrologic properties than expected at the site
D-G9 16 Ec O S
phase reservoirs - Reducing the pressure required to achieve the targeted injection rate and
increasing the air storage volume efficiency in the structure.
D-G10 Subsidence 9 T - Careful pressure monitoring
reservoirs
- As part of the decommissioning all the gas must be withdrawn and caverns
D-G11 Availability of brine for refill of caverns Storage in salt caverns T Ec En
refilled with saturated brine - in the order of billions of liters
Failure of above ground infrastructure
D-S1 (valve, pipes, wellhead or compressor 9 T - Technical status survey before decommissioning
reservoirs
units)
- In order to have a continuously active data transfer, two communication
Interruptions in signal transfers due to All subsurface storage connections will be needed. One of the two connections functions as a
D-S2 18 T O
failures or maintenance reservoirs backup, with functionality to switch over automatically if the primary
connection is interrupted.
D-S3 Obstruction of pump turbine 25 T - If necessary double pump turbines or extra maintenance
reservoirs
D-S4 Control panel connection malfunction 25 T - Additional wires for most risky connections
reservoirs
- Determine the radioactive characteristic of the facility
- Determine the extent and detail of the characterization of the radioactive
content of the facility should be influenced by the expected risk level,
D-S5 Radioactive contamination 37 T En concerning, for example, the measurement of dose rates, determination of
reservoirs
the contamination levels, determination of activation in reactor facilities, the
use of scaling factors between key radionuclides and hard to detect
radionuclides
D-S6
TAB: 4. Decommission (continued)

- Use lessons learned from industry
General well decommissioning risks from
All subsurface storage - Design for decommissioning already at start
D-W1 O&G (stuck items, phishes in hole, unable 1 T Ec
reservoirs - include decommission in all decisions
to create barrier)
- Monitoring
All subsurface storage - Make a safety and performance assessment in order to mitigate the risk,
D-W2 Uncontrolled gas release (blowout) 3 T Ec En C O P S
reservoirs allowing for low safety, health and economical risks
D-W3
D-R1 Micro seismicity 9 T En S - Seismic monitoring
reservoirs
D-R2 Seismicity 9 T En S - Seismic monitoring
reservoirs
Collapse of the subsurface stratigraphy / All subsurface storage
D-R3 9 T Ec En C S - Careful pressure monitoring
reservoirs (can lead to subsidence) reservoirs
D-R4
D-P1
TAB: 5. Post abandonment

Risk category
5. Post-abandonment
Subsidence and sinkhole formation (can be All subsurface - Renewed injection of other substituents
PA-G1 9 T Ec En S
associated with damage to infrastructure) storage reservoirs - Pressure monitoring
All subsurface
PA-G2 Uncertainty on future utilisation 24 Ec - Make a clear time schedule on the operation time of the plants
storage reservoirs
All subsurface
PA-G3 Seismicity 9 T - Seismic monitoring
storage reservoirs
PA-G4
Damage/changes to buildings and agriculture, All subsurface - Monitor the areas (buildings, agriculture) that are making use of the energy
PA-S1 31 Ec En P S
which will change the energy demand storage reservoirs -Monitor the adjacent and overlying areas for damage/problems
Induced seismicity, which can result in alteration All subsurface
PA-S2 1 En O P S - Careful post-abandonment monitoring of seismic activity
of the storage site (surface and subsurface) storage reservoirs
PA-S3
All subsurface
PA-W1 Abandonment plug deteriorating over time 1 T En P S - Monitoring if possible
storage reservoirs
PA-W2
In a brine-filled cavern of great vertical extent, slow downward growth can take place because
Long term pressure changes due to thermal
salt solubility is fractionally higher at the base; the thermal gradients generate slow convection
expansion of the cavern fluid. Differences in
Storage in salt currents causing salt deposition at the top of the cavern. However, the effects and potential
PA-R1 thermal gradients can also lead to additional 9, 11 T Ec - Post-abandonment temperature and pressure monitoring of the cavern
caverns problems might be less in a bedded salt deposit where the brine is saturated and the vertical
dissolution and precipitation of the salt in the
cavern extent is perhaps only 50-100 m, with only a very small temperature difference likely
cavern
across the cavern height.
- Filling the cavern to a pressure just below that of lithostatic pressure before the
Asymptotic pressurization, which is the gradual cavern is sealed and abandoned is a means of restricting excessive strains, perhaps Asymptotic pressurization of a cavern relates more to when storage operations have ceased and
closure of large caverns due to salt creep. This can indefinitely. This requires that the cavern and access well(s) do not leak, which is the cavern has been abandoned following operations
Storage in salt
PA-R2 lead to substantial flexure and high strains in the 9, 11 T Ec not always the case and should this occur, brings with it problems that require The analysis of creep effects alone shows that cavern pressure approaches that of the lithostatic
caverns
overburden strata, which in turn can substantially previous planning. Thus post-gas storage cavern only asymptotically if creep is the only process (i.e. no leakage). Over pressuring and possible
increase their permeability. decommissioning/abandonment/mitigation planning strategies are worthy of note hydraulic fracturing as a result is a possibility, but is thought unlikely.
at this stage.
Leakage/transport of the fluid out of the cavern - water curtain represents an effective way of increasing hydrostatic pressure and
Storage in salt
PA-R3 into surrounding porous strata via porous non-salt 9 T preventing gas leakage
caverns
interbeds - The gas tightness should have been studied for the operational phase
All subsurface
PA-R4 Micro seismicity 9 T - Seismic monitoring
storage reservoirs
All subsurface
PA-R5 Seismicity 9 T - Seismic monitoring
storage reservoirs
Uncertainty on the lon-term creep- and
Storage in salt
PA-R6 permeation behavior of abandoned caverns that
caverns
were used for storing gas
PA-P1
TAB: 6.All phases

Risk category
02 July 2020 LSES consortium TKI LSES 1
6. General (for all subsurface storage types/applications)
All subsurface storage - Start with a low maximum energy production value, to reduce the occurrence of
G-G1 Market distortions 19 Ec C P S
reservoirs unforeseen rejected offers.
Best practices not applied leading to incidents or All subsurface storage - Detailed safety and health assessment
G-G2 28 Ec O S
decreased performance reservoirs - Assess the possible risks for each step
Changes in policies, laws, taxes and regulations All subsurface storage
G-G3 28 Ec C O P S - Keep continuous monitoring of standards, technologies and political situation
put development / economy in jeopardy reservoirs
All subsurface storage - Good bonding with the clients

G-G4 Lack or loss of clients 28 C O S
reservoirs - Make the clients feel comfortable and keep them posted at all steps
Low financing for work leading to low safety All subsurface storage
G-G5 28 Ec O S - Preparation of cash reserves
standards reservoirs
Unsuitable contracts (roles and responsibility not

G-G6 clearly defined) leading to suboptimal 28 T Ec O S - Select experienced and suitable management
reservoirs
performance or exploding costs

G-G7 Demand analysis and forecast are inaccurate 28 T Ec C O S - Get a good overview on the demand and forecasts
reservoirs
- Significantly reducing uncertainty around encountering less favourable hydrologic

All subsurface storage properties than expected at the site
G-G8 Financial risk prior to the start of the project 16 Ec O
reservoirs - Reducing the pressure required to achieve the targeted injection rate and increasing
the air storage volume efficiency in the structure.
Unanticipated delays and costs (materials, All subsurface storage
G-G9 28 Ec O - Include time/cost buffer in the planning
services, maintenance) reservoirs
External natural hazard damaging surface All subsurface storage - Thorough emergency planning (ERP)
G-S1 28 T Ec En S
infrastructure reservoirs - Include adequate specifications for possible emergency scenarios
Anthropogenic hazard damaging surface All subsurface storage - Thorough emergency planning (ERP)
G-S2 28 T Ec En S
infrastructure reservoirs - Include adequate specifications for possible emergency scenarios
- Thorough emergency planning (ERP)
G-W1 External natural hazard damaging well 28 T Ec En S - Include adequate specifications for possible emergency scenarios
reservoirs
- Well design has safety measures (e.g. SSSV) if well can flow by itself
- Thorough emergency planning (ERP)
G-W2 Anthropogenic hazard damaging well 28 T Ec En S - Include adequate specifications for possible emergency scenarios
reservoirs
- Well design has safety measures (e.g. SSSV) if well can flow by itself
All subsurface storage - Start procedure for international standards
G-W3 No (international) (design) standards available 1 T P
reservoirs - Determine fit for purpose design considerations
G-R1
G-P1
TAB: 6.References
Reference ID Reference Link
1 TNO Risk register team
Thienen-Visser et al. (2014) Bow-tie risk assessment combining causes and effects
2
applied to gas oil storage in an abandoned salt cavern. Engineering Geology, 168, 149-166
Geostock Entrepose (2018). Well and storage integrity a changing landscape. CBE,18th
3
Annual Global Energy Village, Barcelona, April 17-18
Mark C. Grubelich, Stephen J. Bauer, & Paul W. Cooper (2011). Potential hazards of
4
compressed air energy storage in depleted natural gas reservoirs.
5 TNO 2017 Eneco Pre-feasibility study for energy storage in Epe salt caverns
6 S. Dewald (2018). 5D Quantum monitor storage performance.
Olaf Kruck, Fritz Crotogino, Ruth Prelicz, Tobias Rudolph (2013). Overview on all Known
7
underground storage technologies for hydrogen.
8 Gordon E. Andrews (2019). The Hy4Heat Programme
D. J. Evans (2008). An appraisal of underground gas storage technologies and incidents, for
9
the development of risk assessment methodology. British geological survey
Deborah Keeley (2008). Failure rates for underground gas storage: Significance for land
10
use planning assessments. RR671 Research Report
11 Berest, P., & Brouard, B. Safety underground storage
Benson. Lessons Learned from Natural and Industrial Analogues for Storage of
12
Carbon Dioxide in Deep Geological Formations
Richard Valenti (2010). KCC open meeting 2010: Compressed Air Energy Storage (CAES)
13
Act KCC Rule Development
Allen (2013). Anthropogenic Earth-Change: We are on a Slippery Slope, Breaking New
14 Ground and It’s Our Fault—A Multi-Disciplinary Review and New Unified Earth-System https://www.researchgate.net/profile/C_Allen2/
Hypothesis. Journal of Earth Science and Engineering 4 (2013) 1-53
Gessel et al., 2017. Eneco pre-feasibility study for energy storage in Epe salt caverns S-
15
102, S-105 and S-106
McGrail et al., 2013. Techno-economic Performance Evaluation of Compressed Air Energy
16
Storage in the Pacific Northwest
Succar & Williams (2008). Compressed Air Energy Storage: Theory, Resources,
17
And Applications For Wind Power. Princeton university
18 Tennet (2017). Productspecification FCR
Herts and partners (2017). Public consultation on “FCR cooperation” potential market
19
design evolutions
HyUnder, 2013. Assessment of the potential, the actors and relevant business cases for
20 large scale and seasonal storage of renewable electricity by hydrogen underground
storage in Europe
Matos et al., 2016. LARGE SCALE UNDERGROUND ENERGY STORAGE FOR RENEWABLES
21 INTEGRATION: GENERAL CRITERIA FOR RESERVOIR IDENTIFICATION AND VIABLE
TECHNOLOGIES
F. Crotogino, S. Donadei, U. Bünger, H. Landinger, (2010). Large-Scale Hydrogen
22
Underground Storage for Securing Future Energy Supplies
Netherlands Enterprise Agency (2017). The effects of hydrogen injection in natural gas
23
networks for the Dutch underground storages.
24 Alexander Körner (2015). Technology Roadmap Hydrogen and Fuel Cells
van Berchum, E., (2014). Pumped hydro storage: pressure cavern - Large-scale energy
25
storage in underground salt cavern
Ullmann & Dayan (2012). Thermodynamic and hydrodynamic response of compressed air
26 DOI: 10.1515/revce-2012-0006
energy storage reservoirs: A review
Nordin (2016). Balancing the energy and transport system of a fully renewable hospital
27
and waste water treatment plant using Fuel Cell Electric Vehicles and hydrogen
28 GeoRISK D2.1 Risk Register
29 Drijver, Struijk and Koornneef (2018) - Hoge temperatuur opslag warmtenet Zuid-Holland
30 Cabeza (2016). Advances in Thermal Energy Storage Systems

Zaadnoordijk, Hornstra and Bonte (2013). Grondwaterbescherming en hoge- temperatuur
31
opslagsystemen
Wassenaar (2017). Projectplan Hernieuwbare Energie: HTO: Hoge Temperatuur Opslag
32
van restwarmte van AVR Duiven
Koornneef, Pluymaekers, Loeve, Goldberg and Wassenaar (2016). AVRFeasibility study of
33
a High Temperature Aquifer Thermal Energy Storage at AVR Duiven
https://medium.com/@cH2ange/louis-londe-technical-director-at-
34 cH2ange. Hydrogen caverns are a proven, inexpensive and reliable technology. geostock-hydrogen-caverns-are-a-proven-inexpensive-and-
346dde79c460
Christina Hemme & Wolfgangvan Berk, 2017. Potential risk of H2S generation and release
35 https://doi.org/10.1016/j.jngse.2017.09.007
in salt cavern gas storage. Journal of Natural Gas Science and Engineering, 47, 114-123
36 Compressed Air Energy Storage. Storing Energy, 2016

37 IAEA, 2013. Safety assessment for deommissioning. Safety reports series no. 77
https://assets.publishing.service.gov.uk/government/uploads/syste
38 Peterhead CCS project - Risk management plan & risk register m/uploads/attachment_data/file/531405/11.023_-
_Risk_Management_Plan_and_Risk_Register.pdf
Van Unen et al (2020), HEATSTORE risk assessment approach for HT-ATES applied to https://www.heatstore.eu/documents/TNO%20report%202020%20R
39
demonstration case Middenmeer, The Netherlands. 15 pp. 10192_HEATSTORE_Final_2020.03.18.pdf
Gevolgen Waarschijnlijkheid (kans)
1 2 3 4 7
Zeldzaam Onw aarschijnlijk Geloofw aardig Waarschijnlijk Zeer w aarschijnlijk
Nog nooit voorgekomen Kan w el eens voorkomen Heeft zich voorgedaan Komt enkele keren per Komt meerdere keren per
in de industrie in de industrie in de industrie jaar voor in de industrie jaar voor in de industrie
ERNST
(effect)
Ernst label
Project
20190903 DAGO Risico Matrix (QHSEP).

Streven naar continue verbetering
Geringe
1 A gevolgen
1 2 3 4 7
Kleine
2 B gevolgen
2 4 6 8 14
TNO report | TNO 2020 R12005 | Project LSES - WP4
Niet acceptabel:
Maatregelen nemen tot
verlaging van het risico
Aanzienlijke
3 C gevolgen
3 6 9 12 21
Grote
4 D gevolgen
4 8 12 16 28
Niet acceptabel: project
stoppen!
Appendix 2 – Consequence & Probability matrix
Uitgebreide
7 E gevolgen
7 14 21 28 49
Figure 3: Consequence – Probability ranking matrix for identifying whether the effect of the risk is acceptable or not
46 / 47
acceptable, and whether mitigations should be taken or the project should stop. The matrix is based on DAGO, 2019.
Gevolgen
Q = Kwaliteit H = Gezondheid S = Veiligheid E = Milieu P = Publieke Acceptatie
ERNST
(effect)
20190903 DAGO Risico Matrix (QHSEP).

Gering Gering lichamelijke of psychische schade aan
Geringe Geen storing in het proces, geschatte gezondheid Niet schadelijk voor de individuele inzetbaarheid of Gering personen. Gering Verw aarloosbare financiële gevolgen. Lokaal
1 Geringe invloed Geringe invloed op de publieke acceptatie.
schade reparatiekosten lager dan EUR 5.000. s voor de uitvoering van het w erk. risico Gering verlies / schade aan installatie(delen). effect milieurisico, binnen de installatie en/of systeem.
effect/letsel Geringe verstoring van de productie.
Schadelijk voor de uitvoering van het w erk,

Klein Gew onden hebben lichte medische zorg nodig en Verontreiniging; schade zodanig dat er gevolgen
beperking van de activiteit, maximaal één w eek
Kleine Mogelijk korte verstoring van het proces, geschatte gezondheid kunnen het w erk direct hervatten. Beperkt verlies / zijn voor het milieu. Kleine incidentele overschrijding Lichte lokale media en/of lokale politieke aandacht,
2 nodig voor volledig herstel. Gebruik chemische Klein risico Klein effect Kleine invloed
schade reparatiekosten lager dan EUR 50.000. s schade aan installatie(delen). Beperkte verstoring van w ettelijke criteria. Geen permanent effect op met potentieel negatieve aspecten voor de operator.
middelen die in beperkte mate op de gezondheid van
effect/letsel van de productie. het milieu.
invloed zijn, zoals bijvoorbeeld irriterende stoffen.
Leidt tot blijvende of gedeeltelijke

TNO report | TNO 2020 R12005 | Project LSES - WP4
arbeidsongeschiktheid. Of ongeschikt voor het Gew onden hebben medische zorg nodig en kunnen
Groot Regionale publieke bezorgdheid. Uitgebreide
verrichten van w erk over een langere periode, het w erk niet hervatten. Vervanging / reparatie van Beperkte lozing op de omgeving van een bekende
Lokale Langdurige verstoring van het proces, geschatte gezondheid Lokaal Lokaal Aanzienlijke negatieve aandacht in de lokale media en/of politiek.
3 langdurige afw ezigheid. Gebruik chemische installatie(delen) beperkt zich tot een paar dagen stof met geringe toxiciteit. Herhaalde overschrijding
schade reparatiekosten lager dan EUR 500.000. s risico effect invloed Met als gevolg een mogelijk negatieve houding bij de
middelen die onomkeerbare schade veroorzaken - stilstand. van w ettelijke criteria.
effect/letsel lokale overheid en vorming van actiegroepen.
zonder enrstige handicap, bijvoorbeeld law aai, Berichtgeving door lokale media.
slechte arbeidsomstandigheden.
Permanente invaliditeit of de mogelijkheid tot één

Nationale publieke bezorgdheid. Uitgebreide
Permanent dode als gevolg van een incident, bijvoorbeeld een Een ernstig gew onde of zelfs een enkel sterfgeval. Ernstige milieuschade; het bedrijf moet uitgebreide
Installatie voor maximaal zes maanden buiten bedrijf negatieve aandacht in de nationale media en/of
Grote arbeidsong explosie. Gebruik chemische middelen die Herstel van installatie(delen) leidt tot een verstoring maatregelen treffen om de vervuilde omgeving
4 en/of geschatte reparatiekosten lager dan EUR Groot risico Groot effect Nationale invloed politiek. Met als gevolg een mogelijk negatieve
schade eschikt onomkeerbare schade veroorzaken met ernstige van de productie van meer dan een maand. w eer in de oorspronkelijke staat te herstellen.
5.000,000. houding bij de nationale overheid en vorming van
tot 1 dode handicap of overlijden, bijvoorbeeld corrosieve Berichtgeving door nationale media. Uitgebreide overschrijding van w ettelijke criteria.
landelijke actiegroepen.
stoffen of bekende carcinogene stoffen.
Mogelijk meerdere doden als gevolg van een Internationale publieke bezorgdheid. Uitgebreide
Meerdere ernstig gew onden of doden. Significant Aanhoudende ernstige milieuschade of overlast die
incident, bijvoorbeeld een explosie. Gebruik van negatieve aandacht in de internationale media en/of
Uitgebreide Uitvallen van delen van de installatie, geschatte Meer dan 1 verlies / schade van installatie(delen), met enkele Enorm zich uitstrekt over een groot gebied. Een groot Internationale
7 chemicaliën met acute toxiteitseffecten Enorm risico politiek, met potentieel ernstige gevolgen voor
schade reparatiekosten meer dan EUR 10.000.000 dode maanden stilstand tot gevolg. effect verlies van natuurw aarde. Constante hoge invloed
(w aterstofsulfide, koolmonoxide) of bekende toegang tot nieuw e w ingebieden. Vorming van
Berichtgeving door internationale media. overschrijding van w ettelijke criteria.
carcinogene stoffen. internationale actiegroepen.
Figure 4: Matrix for interpreting the consequence – probability relationship of a risk. The matrix is based on DAGO, 2019.
47 / 47

Tno 2020 R12005

Uploaded by

Copyright:

Available Formats

Tno 2020 R12005

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Tno 2020 R12005

Uploaded by

Copyright:

Available Formats

Princetonlaan 6

storage of compressed air (CAES) and hydrogen

Date 30 August 2020

All rights reserved.

• In case of leakage of hydrogen or methane in confined spaces, where leakages

• A catastrophic event on the wellpad (e.g. an accident with a heavy truck, or a

• Contrary to methane, which is an inert gas, hydrogen is a reactive gas4. It can

2 Risk Inventory .......................................................................................................... 7

3 Qualitative risk comparison - UHS vs. UGS ........................................................ 10

4 Discussion and conclusions ................................................................................ 23

Appendix 1 – Risk Inventory ................................................................................ 30

Appendix 2 – Consequence & Probability matrix .............................................. 46

1) “Surface Facilities” group: includes compressors, piping, instrumentation,

− Technical: (Sub)surface, Infrastructure, Technology, Operability, Availability,

3 Qualitative risk comparison - UHS vs. UGS

The qualitative risk assessment presented in this section focuses on hydrogen

3.1 Gas properties

3.2 Qualitative comparison of selected risks for H2-storage vs. CH4-storage

3.2.1 Risk theme 1: Material integrity and durability

Pipelines and surface facilities

3.2.2 Risk theme 2: Leakage of hydrogen

Leakage from salt cavern or porous reservoir

In natural gas reservoirs, trapping of gas requires the presence of an impermeable

Furthermore, geological uncertainties could potentially increase the probability of

Leakage could also happen in the subsurface, either because of unforeseen

Potential consequences of leakage

If hydrogen is released in an unconfined open environment it will typically rise and

3.2.3 Risk theme 3: Uncontrolled outflow at the wellhead (blow-out)

A significant difference between hydrogen and methane in this context is the

However, a significant outflow in case of a blow-out event is very unlikely if an SSSV

3.2.4 Risk theme 4: Diffusion and dissolution

3.2.5 Risk theme 5: Loss of H2 and/or contamination of the production stream

However, some hydrogen-induced redox reactions may be significant at low

populations in the formation water, and by performing laboratory tests of reactions of

3.2.6 Risk theme 6: Subsidence and induced seismicity

In depleting gas fields, induced seismicity is generally caused by a pressure decline

4 Discussion and conclusions

An important aspect of underground energy storage is the awareness of potential

For UHS to become an attractive solution the associated risks remain to be

Leakage of hydrogen below-ground caused by integrity failure of wellbore materials

If leakage above-ground were to happen, in an unconfined open environment, then

In case of leakage of hydrogen or methane in confined spaces, there is an elevated

CMEO, 1993. MiIieu-effectrapport ondergrondse gasopslag te Norg.

DAGO, 2019. 20190903 DAGO Risico Matrix (QHSEP).

DNV-GL, 2017. Verkenning waterstofinfrastructuur. In opdracht van het Ministerie

Energystock, “Aanvraag Instemming Wijziging Opslagplan Zuidwending”, 2017.

Evans, D.J., 2008. An appraisal of underground gas storage technologies and

H2Tools, 2020. Link

HyINTEGER, 2017. Long-term underground storage of hydrogen.

Maliva R.G., 2020. Clogging. In: Anthropogenic Aquifer Recharge. Springer

National Research Council, 2008. Prospects for Managed Underground Storage

NOGEPA OPCOM, 2016. Industry Standard No. 41 – Well Engineering and

Panfilov, M. (2010). Underground storage of hydrogen: In situ self-organisation

E. Ruigrok, J. Spetzler, B. Dost, and L, Evers, 2019. Veendam event, 09-01-02019.

SODM, 2019. De integriteit van onshore putten in Nederland. https://www.sodm.nl

SPCS, SAFETY PLATFORM CARGO SURVEYORS (SPCS, 2020) – a Bureau

Tarkowski, R., 2019. Underground hydrogen storage: Characteristics and

Appendix 1 – Risk Inventory

TAB: Risk Inventory LSES