Skip to main content

Low Cost Interconnects with highly improved Contact Strength for SOC Applications

Periodic Reporting for period 2 - LOWCOST-IC (Low Cost Interconnects with highly improved Contact Strength for SOC Applications)

Reporting period: 2020-07-01 to 2021-12-31

The solid oxide Cell technologies (SOC), i.e. the solid oxide fuel cell technology (SOFC) and solid oxide electrolysis technology (SOEC), are highly energetically efficient technologies. The high efficiency makes them strong alternatives to other fuel cell and electrolysis cell technologies. Besides no noble metals or other scarce resources are used in the SOCs. In comparison to alternative electrolysis technologies (alkaline (AEC) or proton exchange membrane (PEMEC) based technologies), the SOEC technology is far more efficient and thus operating cost saving, especially because it can utilize waste heat from downstream processes to increase the overall plant efficiency.

The SOFC technology can be used to convert fuel, preferably renewably made, into electricity. Again the technology is characterised by very high efficiencies, especially for energy dense fuels such as ammonia, which make them ideal for e.g. shipping, where liquid fuels should be converted to electricity with high efficiency. Also land based energy supply based on e.g. biogas could be envisaged.

A very important factor for the competitiveness of the technologies is their capital cost. The main factors playing in on the cost is the lifetime of the cell stacks and the raw material cost.

The LOWCOST-IC project has the objectives of increasing the robustness of solid oxide fuel cell stacks while decreasing the cost of the steel components, i.e. the so-called interconnects. It brings together a consortium covering the entire value chain from the interconnect production to their final use in solid oxide cell stacks. To achieve the goals, a novel production route is tested together with a number of proposed material and stack manufacturing and design improvements.
One of the new materials used in the project is a pre-coated chromium rich ferritic steel from Sandvik (SAN), Sanergy® HT 441. The thin coating is deposited by physical vapour deposition (PVD). It was investigated whether the coating sustained the chosen hydroforming method conducted by Borit (BOR). No change of corrosion protection could be observed by Chalmers University (CU) after the shaping. It could thus be concluded that the proposed cost-effective roll-to-roll production of interconnects is feasible.

With respect to processing, the hydroforming by BOR was shown to be able to easily reproduce Sunfire’s (SUN) interconnect shape using a 0.3 mm steel plate, whereas some deviations on the curvatures were present when using a stiffer 0.5 mm steel plate. To mitigate this limitation to the degree of freedom in the design process, a novel shape of interconnect ribs has been proposed by Forschung Zentrum Jülich (FZJ).

Besides the standard materials from Sandvik also another steel substrate, K44M, from Aperam (APE) and different coating solutions are investigated. Comparing the different steels, it was concluded by CU that coated K44M and AISI 441 offered similar performance in the lab, as the far more costly state-of-the-art steel, Crofer 22 APU. A replacement of this would induce a significant steel cost reduction (> 80 % reduction). Furthermore, various steel compositions with different Cr content - all coated with Sandviks CoCe coating were investigated to detect the lifetime of the steels. It was shown that with the good coating the Cr evaporation was the same for all steels, and failure of the steel from breakaway corrosion was thus only determined by the Cr content.

To join the interconnect with the ceramic cells, a so-called contact layer is needed, and the interface here is critical as loss of contact will lead to failure of the stack. Here a novel contact layer material from Technical University of Denmark (DTU) show a very good performance in terms of high robustness (>200 % tougher, i.e. 4 J/m2) and low area specific resistance (~15mOhm.cm2). This was achieved with contact layers made by CoMn spinels, but contact layers with CuMn spinels showed even better mechanical performance best mechanical performance were from the spinels of CuMn (~20 J/m2). However, the CuMn spinels attracted the Cr of steels and accelerated the Cr evaporation. A second generation of contact layers based on different compositions Co, Cu Mn were thus proposed and investigated. With these it was possible to mitigate the increased Cr diffusion and evaporation from the spinels, while an intermediate fracture energy was obtained (~10 J/m2). A third generation of contact layers were proposed and tested. The initial performance was better than the second generation but after ageing the two solutions ended up with similar results.

It was shown that the very fast drop on demand printing technology (40 m/min) could be used to print ceramic inks on the interconnect tops by Techno Italia (TI). This will save both material and make fast large-scale production feasible. Later in the project different deposition methods were investigated; inject printing, micro extrusion, drop on demand printing and rotary screen-printing. From all the methods investigated the rotary screen-printing technique appeared to be the most suitable technique for the industrial deposition of the conductive inks on the interconnector ribs. The reason was that the printing method permits thicker contact layers, which is needed for some stack designs.

For the interconnect design process, a new very fast multiphysics stack model, which includes the local failure at the interconnect / air electrode interface, was developed by DTU. This reduces the time for a single computation of all physics relevant for a full stack from at least 40 hours to 15 minutes, making iterative design studies feasible. The flow distribution in this was verified using a detailed flow simulation by FZJ. Eventually it was realised that with the design of the Sunfire stacks only moderate stresses would be generated at the cell-interconnect interface, which is consistent an observed low failure rate at this interface in the actual stacks. Thus, a different design target was set, which targeted minimising the stress in thinner electrolytes. A design for an interconnect was proposed based on this. This was manufactured by hydroforming and a stack is currently being tested.

The cost analysis was initiated by AVL, with preliminary results showing that it is possible to reduce the cost below 5 € per interconnect with the different proposed changes to the processing and materials.

Stacks with Sanergy® HT 441 were produced with the existing interconnect shaping methods and tested. For high temperature operation (850C) at SUN, it was shown that Sanergy 441 HT corroded faster than the rather expensive Crofer 22 APU steel. The reason for this is believed to be that a dense Si scale is forming below the corrosion protective coating. In lab experiments the minor gaps in the Si layer is however sufficient for the resistance of the interconnect not to be apparent. The reason for the discrepancy will be investigated in the final part of the project.
The project has thus at this stage already has shown a larger number of improvements over state of the start in terms of possible material as described, and it is approaching the end as well. Therefore, only few activities for the final test and conclusions on those are in progress.
Close up of details of interconnect shaped with hydroforming
Novel concept developed to model the stresses local in stacks using a multi scale modelling approach
Concept of the project illustrating the value chain and the supporting research institutions.
Sunfire stack manufactured with new interconnect materials, ready for testing
Developed reactive bonding concept for the novel air electrode contact layers.
Flow of fuels and the current density distribution in a Sunfire stack at low current density.