Learnings

Outcomes

The work packages undertaken resulted in the following outcomes:

Hydrogen Logistics

·       Grid electricity could be used to generate hydrogen but it would involve multiple energy conversions, with efficiency losses at each stage.

·       Renewable electricity sources have inherently variable outputs and would require secondary energy inputs such as a grid connection to manage intermittency. Electrolysis systems must be able to operate at variable energy inputs, with potentially fast changing demand.

·       With Steam Methane Reformers (SMRs), carbon capture and storage (CCS) is needed to achieve reductions in greenhouse gas emissions. For smaller scale supplies such as those required for the three potential demonstration sites, electrolytic generation is more feasible.

·       In terms of using bottled gas delivered to site in high-pressure cylinders - a typical hydrogen cylinder (50 litre water capacity at 200bar) contains about 0.83 kg so the capacity of a standard Multiple Cylinder Pack (MCP) (16 cylinders) is only about 13 kg of hydrogen. As winter use can be 30 kg/hr the use of such MCP’s is considered impracticable. 

·       Tankering was deemed unsuitable due to the higher costs in comparison to utilising the existing turbine and generating on site. Also a supply chain risk of using a single supplier for tankering the hydrogen to site

·       The element concluded the most optimum hydrogen production solution for Levenmouth would be electrolysis via wind power and above ground storage

Consequences

·       The Hydrogen consequences element undertook a series of tests to determine the differences in events of a hydrogen ignition event vs a natural gas ignition event. The results and review of previous incident data highlighted appropriate risk mitigation measures that would ensure risk of a hydrogen event was no higher than a natural gas ignition event.

·       Hydrogen excess flow valves on the service, using external meters where possible, smart meters with automated shut off valves that trip when a leak is present and crimp fittings in the property were all identified as effective risk mitigation measures for consideration in future trial.

Characteristics and QRA

·       Hydrogen can travel between 6% and 15% further than natural gas from a below ground gas release.

·       A small number of cases can result in an increased distance of up to 25% (highly dependent on ground conditions and other utilities such as ducts, cables and water pipes and or the presence of a sealed top surface.

·       Installing a full polyethylene network offers a significant reduction in the likelihood of an upstream release.

·       The risks associated with the production and storage facility are in the range which can be considered tolerable if they have been reduced as low as reasonably practicable (ALARP). 

·       The likelihood of an ignited release of hydrogen within a domestic property is estimated to be of the same order of magnitude as the likelihood of an ignited release of natural gas within a domestic property. This estimate assumed that all properties and relevant appliances will be fitted with appropriate risk reduction measures to prevent, control and mitigate releases of hydrogen should they occur. 

·       The likelihood of a release from the new all PE network with additional quality assurance checks on all fusion joints and connections will result in a significantly reduced likelihood of a release on the network compared with the current gas distribution network (of around 66%). This risk is being further mitigated by the use of Excess Flow Valve devices inserted in the top tee of the service.

Odorant and gas detection

·       The report concludes that the odorant used in the current UK natural gas network - odorant NB is suitable for use with hydrogen.

·       Sulphur free odorants did not induce corrosion on hydrogen fuelled boilers.

·       Hydrogen fuelled boiler corrosion was comparable to Natural gas when sulphur-based odorants where used, indicating there were no issues associated with boiler/materials compatibility.

·       Odorants containing sulphur caused noticeable degradation of the fuel cell.

·       A small reduction in fuel cell voltage was noted when testing with sulphur free odorant.

·       When testing for environmental stress induced cracking, the presence of any of the odorants in hydrogen had no effect on Polyethylene (PE80) and steel (X42 carbon steel).

·       The technoeconomic report concludes that all odorants cause fuel cell degradation to some degree and that the sulphur based odorants are by far the most damaging, it also concludes that Odorant NB is the best technical and the most cost effective option.

·       Sulphur free odorant (GASODOR-S-FREE) was suitable for fuel cells, and was similar in cost to Odorant NB, this is the recommended option if a sulphur free odorant is required.

·       Of the hydrogen Gas Detection Instruments (GDI) tested, instruments were found to be suitable for use in all the ranges required (0>100% gas in air, 0>100% LEL and ppm).

·       No single gas detector was found which could cover all three measurement ranges required by the GDN’s.

·       All of the hydrogen GDIs tested were cross sensitive to natural gas and therefore not suitable for use by GDN’s as they could not differentiate between hydrogen and natural gas.

·       The current fleet of GDIs used by SGN - the GMI Gassurveyor is not be suitable for use with hydrogen. 

·       The outcomes of the gas detection phase resulted in a new NIA being borne to manufacture, test and certify a hydrogen gas detection instrument (NIA_SGN0156).

Safety Case and procedures

·       The structure and arrangements within the SMF have been defined and set out in such a way that it demonstrates that risk will be reduced to as low as is reasonably practicable (ALARP).

·       A roadmap has been developed that has 14 key milestones that cover the whole process from production through to burner tip.  This structured approach to identifying hazards and risk and putting controls in place will ensure that the assessment and development of the SMF is robust.

·       A review and assessment of all standards and procedures has been undertaken to identify and evaluate where the differences lie when operating a hydrogen network and highlighting the impact of the changes required.   

Transportation of debris in hydrogen pipelines

·       New pipe systems constructed to current standards should be clean and contain negligible amounts of debris.  The H100 network is a stand-alone system that is fed with filtered clean dry hydrogen gas which means that it will not be exposed to the effects of debris mobilisation, erosion and deposition from older networks upstream. The only debris present will be the very small amount related to construction activities usually tiny PE shavings associated with joint preparation.  For these reasons the project concluded that new H100 network will not be subject the effects of velocity induced saltation.

·       The theoretical work conducted indicates that gas flow velocity for hydrogen will increase and could be up to 3.5 the current value for natural gas.  This means that there is an increased likelihood of saltation in existing older mixed material natural gas networks where the existence of debris such as black dust could be an issue.  However, reinforcement and increased operating pressures could alleviate some of these issues and keep gas velocity within acceptable limits.

·       It was recommended that further work be undertaken to investigate the velocity issues and potential for erosional velocity in a conversion scenario.

Metering solution

·       Hydrogen flow can be metered using existing diaphragm gas meters, providing the meter flow rate is sufficient to meet the demand requirements of the connected appliances.

·       Current domestic meter installation designs are suitable for hydrogen.  The only addition being the fitment of a downstream Excess Flow Valve to protect pipework and appliances. 

·       Conventional diaphragm meters do not give rise to ignition potential in the case of a leak, providing they are housed in an external meter box with no risk of ignition from external sources

·       Excess flow valve installation in the service pipe, close to the top tee, and as part of the meter installation enables effective mitigation of high-volume gas escapes

·       External gas meter installation in an approved meter box is an effective downstream risk mitigation measure and the preferred option.

·       Investigating if mechanical meters can be calibrated for use for hydrogen flow rates was a secondary option for the hydrogen demonstration. The project’s primary choice of a metering solution is to use a hydrogen smart meter with an automated slam shut valve, currently under development in the BEIS led programme - Hy4Heat 

PE materials

·       Mathematical modelling conducted during the test programme has shown that hydrogen has a reduced ability to maintain pressure at the crack tip when compared to natural gas, hence the risk of rapid crack propagation is much reduced.

·       Polyethylene pipes and components will not experience hydrogen embrittlement at the temperatures and pressures experienced on a low-pressure gas distribution system.

·       Pipes and fittings fully permeated with hydrogen were successfully benchmarked against the performance criteria in the relevant industry standard for electrofusion, butt fusion and flow stopping techniques. In addition, no premature/early life failure mode where evidenced as a consequence of the activity.

·       The gas industry standard (GIS/EFV1) and natural gas EFV’s are not suitable for hydrogen as flow rates exceed the specification for both.  A new hydrogen EFV and standard is required for the H100 demonstration (currently being developed under NIA_SGN0154)

·       Time dependant failure modes - The overall conclusion from testing was that polyethylene pipe systems used to distribute hydrogen will have the same strength classification as those used to distribute natural gas, important for the calculation of maximum operating pressures (MOP) for the network.

·       Charpy impact testing was used to measure fracture toughness. The results have shown that thick sections of PE tested with both hydrogen and natural gas have comparable fracture toughness results and when examined under the microscope no differences where noted. Thin sections of PE showed mixed results and further testing is being undertaken under the Hydeploy project. Results expected summer 2021

·       Evaluation of time dependent failure modes for polyethylene pipe, fittings and jointing techniques shows that there is no degradation due to hydrogen exposure of any of the components and that they have an expected lifetime in excess of 50 years.

·       Elastomeric seals were tested against the current GIS standard which recommends hand tightening the cap that contains the seal.  This recommendation was found to be unsuitable for hydrogen as hand tight can vary between 4 and 7 newton meters.  Testing has concluded that the correct torque setting for top tee caps needs to be between and 6 and 7 Newton Meters this setting will ensure the cap is hydrogen tight. 

Commercial and regulatory

·       4 potential value chain models have been developed that are not site dependant 

·       Two options were further refined to be specific to the awarded site - Levenmouth

·       The preferred option closely aligns with standard network operating model and hence would require minimal process change. This model has been assessed by industry and academic experts to ensure it is the best solution for the customers partaking the demonstration. 

·       Subsidy support is required as the capital and operational costs of hydrogen production and the required storage are currently higher than natural gas

·       Setting up a special purpose vehicle allows the project to carry out the unregulated activities relating to the demonstration i.e. gas production and sale of gas. 

·       The preferred model was reviewed and challenged by legal experts, academics and industry to ensure it was the model most suited from a customer’s perspective – least disruptive, does not cost the customer more than current gas costs.

The preferred model was put forward and progressed further in support of the demonstration phase.

Flame visibility

·       From the evidence gathered and reviewed for both high and low volume gas escapes, it was concluded that hydrogen flames are likely to be clearly visible for releases above 2bar, particularly for larger release rates. 

·       At lower pressures, hydrogen flame visibility will be affected by ambient lighting, background colour and release orientation.

·       A review of current risk reduction measures and additional measures identified through the H100 Hydrogen Consequences and Characteristics projects, has concluded that fires with the potential to be less visible are adequately controlled by current procedures and practices.  The possibility of injury is also extremely low due to the rarity of such events occurring and the consequence of contact being unlikely to be severe.

·       These conclusions are supported by cost-benefit analysis that shows that no additional risk mitigation measures are justified for the H100 project (ALARP).

Feasibility and FEED studies

·       The Aberdeen feasibility study highlighted the new housing estate would not be ready in time to meet the project timeline. The site did not progress into FEED based on this finding.

·       The outcome of the two FEEDs showed Machrihanish and Levenmouth are both credible for a 100% hydrogen network and each offer their own unique characteristics, socio-economic positions and technical solutions. 

·       Levenmouth enables the demonstration of changeover from existing natural gas customers to hydrogen gas which is representative of 85% of GB customers and explores the energy transition for gas infrastructure. 

·       Machrihanish presents an opportunity for testing hybrid head solutions, customers supplied by oil and electricity to be given an option to convert to hydrogen gas or exploring industrial heat demand with onsite industrial users. 

The outcomes have been carried over to progress in NIC phase of H100 Fife.

Lessons Learnt

·       This project identified 3 suitable locations for 100% hydrogen demonstrations, each with unique features to validate a number of decarbonisation issues. Preliminary design work undertaken allows for these to be picked ‘off-the-shelf’ for future projects, saving time and cost. 

·       The extensive research and development evidence base that has fed into the quantification of risk for operating a 100% purpose-built hydrogen system from production through to utilisation in the home, will inform other hydrogen distribution projects.

·       Commercial and regulatory gap analysis and modelling has identified the optimum model that allows the project to operate without derogation from licensed activities. The model has been assessed by industry experts and academically reviewed to ensure it is customer centric. The lessons learned enabled the project to bid for and win funding in the NIC to construct and operate the 100% hydrogen demonstration at Levenmouth, Fife. Project now known as H100 Fife. 

The evidence base and feasibility and FEED studies from this project are replicable across the UK gas distribution networks.