Therma 4910 stainless steel could prevent weld failures in thermal energy storage tanks

Concentrated solar power plants with thermal energy storage face the challenge of stress relaxation cracking in the current 347H austenitic stainless steel grade used commonly to manufacture molten salt tanks. Andy Backhouse, Outokumpu Lead Technical Manager, outlines the result of a consortium project that suggests a switch to Therma 4910 could eliminate these failures.

CSP and thermal storage

A recent report by the IEA (International Energy Agency) concluded that, with suitable governmental support, concentrated solar power (CSP) could generate over 11% of global electricity demand by 2050. However, to make CSP a cost-effective competitor to fossil-fueled base generation it needs the dispatchability made possible when it is used in combination with thermal energy storage (TES) systems.

For CSP, the thermal storage is often in the form of huge storage tanks containing thousands of tons of molten salt cycling between temperatures of 300°C and 600°C. Since the molten salt might lose only one degree of its heat a day, it is possible to store the thermal energy for months. But in practice, the storage is used for daily time-shifting, to make the energy produced during the day available for the evening periods of peak demand.

SRC issues with 347H in thermal storage tanks

The austenitic stainless steel grade 347H is the established material for the construction of molten salt storage tanks. This is because, when in contact with the storage medium, it offers enhanced stress corrosion cracking (SCC) resistance and elevated temperature mechanical strength in comparison to 304H and 316H grades. However, storage tanks used with CSP plants have known issues with intergranular stress relaxation cracking (SRC) emerging after months to years of service in 347H weldments.

SRC occurs due to the presence of weld-induced residual stresses and susceptible microstructures operating under elevated service temperatures of around 550°C and above. This causes some alloying elements to migrate to the grain boundaries, resulting in a concentration of brittle precipitates. At the same time, there is local strain accumulation near these brittle features, leading to fracture during stress relaxation.

It is possible that post weld heat treatment (PWHT) may alleviate this cracking by stress relief and stabilizing the microstructure. But implementation of PWHT in the field can be challenging and might even contribute to higher temperature SRC if the process is not carefully designed.

Various industries, including CSP, have looked at 316H stainless steel as a potential solution. Yet, the nuclear industry in the past decades has reported hundreds of SRC failures in advanced gas cooled reactors using 316H, and the material is accepted to have a high susceptibility to SRC based on this record and laboratory scale testing. That is why it has not been pursued for this application.

Could Therma 4910 make a difference?

To address these SRC concerns, there is an increasing demand to evaluate alternative alloys and weld fillers with improved SRC resistance and comparable thermomechanical properties (such as creep and fatigue) and corrosion resistance. A promising alternative to 316H is Outokumpu’s Therma 4910 (Alloy 1.4910) also known as 316LNB (low carbon and added nitrogen and boron content). It offers excellent creep resistance and comparable molten salt corrosion resistance compared to 347H In addition, the use of 16-8-2 filler has been demonstrated to improve toughness and thermomechanical properties as an alternative filler to matching fillers in 347H and 304H SS welds.

It should be noted that Therma 4910 is not a wholly new material.  EN 1.4910 was developed originally in the late 20th Century and found application in European coal-fired power plants during a time when there was a push to increase operating temperatures and pressures. This has left a legacy of useful experience and long-term high temperature strength and creep test data.

Outokumpu identified the potential usefulness of this material for TES applications and has re-introduced the grade as Therma 4910 with a specific focus on the SRC challenge, to investigate the suitability of Therma 4910 for CSP thermal storage, a collaborative project was established between the Colorado School of Mines, Vast – a world leader in CSP, Outokumpu - a world-leading supplier of stainless steel and the construction partner CYD.

Putting Therma 4910 to the test

The purpose of the program was to evaluate the SRC resistance of Therma 4910 and its weldments using 16-8-2 filler, under a Gleeble® thermomechanical testing procedure, in comparison to 347H.

The Gleeble 3500 is a fully integrated digital closed loop control thermal and mechanical testing system capable of heating specimens at rates of up to 10,000°C/second, while exerting as much as 10 tons of static force in tension or compression.

The thermomechanical testing revealed promising results for the Therma 4910 heat affected zone (HAZ) and 16-8-2 filler (ER16.8.2) fusion zone (FZ). No cracking was observed within the specified 22-hour test period at elevated temperatures ranging from 600 to 800°C and initial true stress conditions of 650 MPa (0.174 strain) for HAZ and 460 MPa (yield strength) for FZ.

In contrast, the 347H HAZ and matching filler FZ experienced cracking within a few hours at 800°C.

Future work will focus on completing the test matrix, such as testing for longer times at 600°C and investigating the microstructural conditions needed for fracture, such as secondary phases in the partially melted zone (PMZ).

Conclusion - Therma 4910 is a viable candidate for molten salt storage tanks

The clear conclusion from this initial test program is that, having simulated the in-service temperature and stress effects in laboratory tests, Therma 4910 austenitic stainless steel offers greater resilience to SRC in conditions comparable with molten salt storage tanks than the currently used 347H grade.

Because Therma 4910 is slightly more alloyed it is more costly to produce than 347H. However, the additional cost is a small consideration when weighed against the potential risk and reputational cost of a failure in a thermal energy storage tank.

Addressing material challenges is becoming even more pressing as CSP is used increasingly to provide heat for both power generation and industrial processes. While thermal energy storage is used for time-shifting of power generation at temperatures up to 600°C,  even higher operating temperatures are under investigation by several research groups to improve efficiency and reduce operating costs of CSP.  The combination of excellent high temperature creep strength and SRC resistance of Therma 4910 could prove to be a valuable asset in such circumstances.

Further reading

Full details of the research program can be found in a paper published at SolarPACES 2024: Pickle et al. “Evaluation of Alternative Base Materials for Mitigation of Stress Relaxation Cracking in Thermal Energy Storage Tanks”.