Learnings

Outcomes

Summary of Results

Fracture toughness tests have been carried out on a range of different NTS pipe grade materials in 70 bar hydrogen environments. Measured fracture toughness values (KJ0) are in the range 72.7 to 149.8 MPa√m. Measured fracture toughness values are higher than the suggested lower bound value derived from Charpy Impact Energy as per the IGEM/TD/1 hydrogen supplement. Fatigue crack growth rate tests have also been carried out on a range of different NTS pipe grade materials in 70 bar hydrogen environments. The fatigue crack growth rate was observed to be dependent on the cyclic loading frequency, with growth rate increasing as the frequency was reduced from 1 Hz to 0.001 Hz. Fatigue crack growth rates were similar for the range of material grades tested. The majority of fatigue crack growth rate data were lower than the upper bound curve recommended in the IGEM/TD/1 hydrogen supplement, however, at low ∆K some data were observed to be higher.

Pipe stress analysis and fracture mechanics assessment of critical components within the FutureGrid demonstrator high-pressure transmission system has shown that a maximum allowable operating pressure of 70 bar is acceptable. Fatigue lives are predicted to be less than 20 years in some cases, so a programme of periodic inspection and monitoring is recommended to manage this risk.

Fracture mechanics analyses of example NTS pipelines were found to be highly sensitive to assumed input parameters. In particular, the assumed initial flaw size is a dominant factor, along with other inputs such as welding residual stress and assumed stress concentration factor associated with weld misalignment. A number of analysis options were presented. For existing pipelines operating at design factors in excess of 0.6, the fracture mechanics assessments suggest that meeting current maximum operating pressures (MOPs) will be challenging and could require additional data from inspections or testing to validate critical input assumptions. These additional inspections or testing would need to focus on quantifying:

• The flaw sizes present in welds, specifically focussing on the flaw depth (through wall extent) which is available from ultrasonic inspections, but not radiographic inspections. The defect position, i.e. whether surface breaking or embedded, could also provide useful information.

• The extent of weld misalignment which contributes towards the derivation of a stress concentration factor. Worst case assumptions can be made from limits given in material specifications, but these are onerous in many cases and significant benefit could be derived from statistical analysis of real data. 

• The magnitude of welding residual stresses, particularly for assessment of girth welds where this has a strong influence. Worst case assumptions can again be made, but this may limit the maximum allowable operating pressure of many lines.

When considering carrying out this type of assessment for an entire network of pipelines, it is also considered that a staged approach may be of use, i.e.

• An initial screening activity based on design factor to identify those pipelines that will be most challenging to operate under hydrogen without a pressure reduction. This information could then be used to focus any inspection activities on pipeline populations where they will provide the most benefit.

• The assumed fracture toughness could be refined based on age and grade to group pipelines into different categories, rather than taking a worst case for all assessments. Materials data will need to be collated in order to determine the extent to which pipelines can be grouped with regards to assumption regarding fracture toughness. This information includes: year of fabrication, material grade, welding process / specifications, mill certificates giving chemistry and mechanical property data, weld NDT records (if available). The test data described in this report are considered sufficient to allow the screening and prioritisation exercises described above but might not be sufficient for carrying out fitness for service assessments (which typically require triplicate tests of relevant materials) without further justification.

• An inspection campaign focused on determining the flaw sizes present in seam and girth welds will assist in clarifying the extent of any further information that may be required. If the worst-case welding flaws are moderate or large, then further investigations will be necessary to quantify the extent of weld misalignment and/or welding residual stress. Use of in-line inspection crack detection tools would provide the best means of collecting large quantities of data. For seam welds, EMAT or UT tools have the potential to collect suitable data. For girth welds, there are greater technical challenges and tool capabilities are less well defined. Data from girth weld inspections at excavations should also be collated to provide additional information. These inspections should incorporate phased array UT and reporting requirements should be improved so that all inspection findings are quantified (i.e. defect depth and length) as opposed to just reporting compliance with a relevant specification.

• Initial fracture mechanics assessments should be “deterministic” in nature, i.e. by considering worst case values for all inputs. However, where such assessments “fail” (which is likely for pipelines operating at high design factor) it is recommended to use a “probabilistic” approach, which avoids making the assumption that worst case values of inputs such as flaw size, weld misalignment and welding residual stress will coincide at the same location. This method would consider the statistical distributions associated with critical input parameters rather than worst case values. The data obtained from inspection and testing activities can be used to develop the required input parameter distributions.

The project value tracking is listed below: 

·       Maturity 

o   TRL 4-5. Laboratory testing of pipeline steels.

·       Opportunity 

o   100% of single asset class. Project assessed materials representative of the pipeline steel grades on National Transmission System.

·       Deployment Costs  

o   £0.00. Project is research and there will be no technology developed to be deployed.

·       Innovation Cost 

o   £ 1,099,733. Cost of entire project following project scope expansion (following change control).

·       Financial Saving 

o   £0.00. Data generated in the project could enable operation of gas network assets at higher pressures than would otherwise be permissible thereby avoiding the need for new pipeline assets to be installed.

·       Safety 

o   0% Improvement. No direct safety improvement expected. Data generated in the project can be used to define operational regime to ensure safety (e.g. mitigation against fatigue failure).

·       Environment 

o   0 tonnes CO2e saving. No direct environmental benefits expected. Data generated in the project could enable operation of gas network assets at higher pressures than would otherwise be permissible thereby avoiding the need for new pipeline assets to be installed.

·       Compliance 

o   Support compliance. Data will be used in safety case evidence submission.

·       Skills & Competencies 

o   Group. Work will augment knowledge of hydrogen team by feeding into decisions on Project Unite and Project Union. 

·       Future Proof 

o   Supports business strategy. Results will support operation of future hydrogen national transmission system. 

Lessons Learnt

Throughout and after the project the following lessons learnt have been captured:              

·       Testing Timescale

o   Testing of materials in high-pressure hydrogen is a technically complex activity with low maturity in the supply base. Whilst DNV are leaders in the testing, delays were still experienced, compounded by the size of the test programme. The time-dependent nature of hydrogen embrittlement means that testing is ideally undertaken at long durations to ensure conditions representative of pipeline operation, this can further extend testing timescales.

o   Future projects should ensure the scope of the testing is defined as well as possible and close liaison with the testing supplier is needed to forecast realistic and achievable timescales.

o   If possible, the status of material (e.g. pipes) for testing and supporting data (e.g. wall thicknesses and length of available welds) should be available before or at the beginning of projects to ensure efficient use of the project timescale.

·       Reporting

o   Delays were experienced with the reviewing and updating reports, predominantly due to resource availability.

o   Sufficient time should be incorporated into the project plan for reviewing and feedback especially on projects which contain significant technical information.

·       Collaboration

o   Regular project reviews were held between DNV and NGT throughout the project, these were critical in ensuring an appropriate test programme was defined and followed.

o   These discussions lead to the scope modification to investigate sub-critical crack growth the information from which has been used to inform the selection of the material fracture toughness value (Kmat) for fracture assessments.

o   It is recommended that close collaboration and discussion is continued in future projects.