System-Supporting Operation of Solid-Oxide Electrolysis Stacks

Round 1

Reviewer 1 Report

The paper systematically showed a system supporting operation of solid oxide electrolysis reactor. By comparing the effects of the load profile on electrical stack performance, stack degradation and produced gas compositions, some interesting conclusions and suggestions were made, like "the guarantee of a continuous feed supply to the SOEC and its thermal management represents a major obstacle to flexibility","sophisticated heatmanagement is essential for HTCoEL's operation", etc. It should say the paper was well organized and written, and the figures and tables are full enough to help readers to understand the conclusions. It can be accepted after the minor revisions are made. For the timeplot of U/V in Figure 5, only L1 and L4 were showed, where were L2 and L3? What was H₂o/CO₂ on Line 351 in Page 13? was that H₂O/CO₂? In addition, the conclusion was too long.

Author Response

Dear reviewer,

thank you for your positive comments, they are much appreciated. I will start with answering your questions first and then cover your other comments:

1. For the timeplot of U/V in Figure 5, only L1 and L4 were showed, where were L2 and L3?
   The timeplot of U/V in Figure 5 does contain the cell voltages for layers 2 and 3, but they are so similar to the voltages of layer 4, that is overlaid be the curve of layer 4 (magenta). This can be seen at really high magnifications, but is very easily overlooked. We added a hint for this issue to the manuscript.
2. What was H₂o/CO₂ on Line 351 in Page 13? was that H₂O/CO₂?
   Line 351 on page 13 does correctly mention the H2O/CO2 ratio in the feed gas. This ratio (which also determines the ratio of oxygen/carbon in the feed) is important to prevent any soot formation in the pipes and components of the test rig and the stack and cells. Otherwise any soot formation will quickly clog the gas supply are destroy the cell substrates. Therefore our intention was, that during transitions of power levels the H2O/CO2 ratio may shortly increase but never decrease. We added an explanation to the manuscript for reduce the potential for confusion.
3. In addition, the conclusion was too long.
   We agree that the conclusion should be shorter. We rephrased it accordingly and moved some of the statements concerning limitations to the results&discussion section.
4. Extensive editing of English language and style required 
   We submitted the draft to our instutional editor who checked for spelling mistakes and language issues. We fixed some minor issues during revision, but he stated, that there are no major problems with the text. We want to point out, that he is an experienced, professional, native-speaking editor. Therefore we think, the manuscript should be acceptable with respect to language and spelling. If there are any serious issues remaining, we would be happy to hear about them.

Best regards and best wishes for 2021,
Dominik Schäfer & co-authors

Reviewer 2 Report

The work by Schader reports a load-following study of a 5-cell SOEC stack in response to a simulated load change to reflect that in grid power variations. To deploy SOEC technology into grid storage, this work is important and necessary from an engineering viewpoint, as it gains key information that is needed to design SOEC components and develop control algorithm. The finding of response limiting gas supply system to the load as well as conversion is very useful and important to the future system design. This reviewer recommends it for publication. Below is a list of comments for the authors to consider.

• The definition of relative power level (Pr) is not clear?
• Why would the conversion be higher at higher relative power level?
• What is the gas composition for load cycle experiment with static gas supply?
• What is the reason for the increase of relative power level with time in Figure 3 and 4? Should it be kept constant during operation when gas flow and current density are constant?
• It appears that steamer has some issues affecting the testing. What is the best solution?
• Figure 6 ahead of Figure 5??
• Could the stack be operated at thermoneutral potential instead of galvanic current?

The stack temperature seems to be minimally affected. Discussion is needed to why: is it because it is close to thermoneutral potential of the cell voltage?

Author Response

Dear reviewer,

thank you very much for your helpful comments and questions and your overall positive verdict. In the following we will answer your questions and mention the corresponding changes to the manuscript:

1. The definition of relative power level (Pr) is not clear?
   We agree and added additional explanations and extensively re-phrased the corresponding paragraph in the section "Stack operation".
   
2. Why would the conversion be higher at higher relative power level?
   In general, a power increase wouldn't necessarily do that. A power increase at constant current and constant feed gas flow does not change the conversion ratio.
   But in this case the power levels were differing in current density. This directly increased the amount of converted gas along the cells and in case of stack A with a fixed feed gas flow, the conversion ratio. For stack B the conversion ratio was constant at all power levels by adjusting the feed gas flow rates to account for the different currents.
3. What is the gas composition for load cycle experiment with static gas supply?
   There was a slight mistake in the manuscript on line 251, stating the feed gas composition was constant for stack A. In fact, the feed gas composition was constant for both stacks, being 60% H2O, 30% CO2 and 10% H2. However, it was meant to state that the feed gas flow rate was constant for stack A. (While it was dynamic for stack B to enable the constant-conversion operation.) This has been fixed and also a sentences added for describing table 2 to further clarify the values given for j and CR in the table.
4. What is the reason for the increase of relative power level with time in Figure 3 and 4? Should it be kept constant during operation when gas flow and current density are constant?
   The increase of the relative power during in figures 3 and 4 while current density and feeds remain constant are due to the degradation of the cells in the stack which caused an increase in internal resistances. For maintaining the constant current the individual cell voltages have to increase and thus the power of the stack. It therefore also reflects the amount of degradation. It is feasible to keep the power constant, but this requires a trade-off, because the current density (and thus the conversion) becomes a variable. In this experiments we opted to keep the conversion (ratio) constant. The effect is also present for stack B in figures 6 and 7, but much less pronounced due to the lower degradation. We added a note about the increasing stack power.
5. It appears that steamer has some issues affecting the testing. What is the best solution?
   This is complex to answer comprehensively. We performed follow-up investigations on the carry-over of steam generators and gas preheaters in the meantime and plan to publish the results in detail soon. However there are two main issues that might be relevant in the experiment reported here: (a) Certain substances (e.g. boronic acid and silicic acid) have a high steam soluibility and can quite easily pass common ion-exchangers. For such substances, it is important that the water supplied to the steam generator is properly cleaned of them. (b) The once-through steam generators employed here have a high physical carry-over of water droplets (e.g. by entrainment) because they lack steam separators. This may carry even substances with low steam soluibility towards the stack. Also, our investigations showed that deposits of impurities in the steam generator could be mobilized in vast amounts by unstable evaporation, e.g. in case of overloaded steam generators, transients or quick load changes. For mitigating these issues, steam generators should be kept free of deposits and the use of steam generators with proper steam separators are advisable.
6. Figure 6 ahead of Figure 5??
   This seems to be a processing issue, because the source latex file does contain the graphs in the correct order and indeeed the sequence is right in the latest PDF. We suggest to solve this technical issue together with the latex support team at MDPI at the copy-editing step if the manuscript is accepted and it happens again.
   
7. Could the stack be operated at thermoneutral potential instead of galvanic current?
   Yes, it could be operated potentiostatically. However, the drawback is that the conversion (ratio) is not fixed in this mode (because the current changes) and therefore the product gas composition changes. Therefore, the stacks are usually operated with a galvanostatic control loop. For a real system, a "quasi-potentiostatic" operation is an option: the direct control loop is galvanostatic, but the current and feed gases are quickly adjusted by a secondary (slower) control loop to keep the layer voltages near around the thermoneutral voltages.
   In this case we also intended to keep all layer voltages during the experiments at or below the thermoneutral voltage to avoid a transition between the two operating regimes (endothermal vs. exothermal) mid-experiment. Therefore the current density of the highest power level was chosen in a way that the resulting voltages at the given feed gas composition, (maximum) conversion ratio and stack temperature did not exceed 1.35V (Uth being 1.344 V). Additionally we planned to have a large voltage reserve allowing for a pronounced degradation of the cells during the experiment.
   We also added a few more comments to the section "Stack operation" to clarify our design choices.
   
8. The stack temperature seems to be minimally affected. Discussion is needed to why: is it because it is close to thermoneutral potential of the cell voltage?
   We agree, this is important and added a paragraph discussing the heat demand of the stack for the different power levels to the manuscript:
   The stack temperature changes for the entire profile remained below 2 K. Given a thermoneutral voltage of 1.344 V for the given gas mixture, the heat consumption at the 125% power level is estimated to about 32 W for the complete stack. The lower power levels have lower layer voltages (i.e. larger difference to the thermoneutral voltage), but also much lower currents. This results in estimated heat demands of 36 W (100%), 35 W (75%) and 29 W (50%) for the other power levels. This explains why the stack temperature is lowest (and nearly the same) for the 100% and 75% levels and highest (and also nearly the same) for the 125% and 50% levels. The maximum difference in heat demand is 7 W. Taking into account the small stack size and very good thermal conductivity within the stack, this explains the very small difference in stack temperature.

Best regards and best wishes for 2021,
Dominik Schäfer & co-authors