Separator material
-
I have started the first tests with the RFB. I have encountered the problem that the electrolyte migrates from one reservoir to the other. I assume that a better separator material would help. I am currently using three pieces of photo paper. Could more layers of photo paper reduce the effect?
The pumps I use are probably a little too big, which certainly doesn't help.
I assume that with larger cells, the problem will intensify and that a better separator material than photo paper will be necessary.It seems to me that there is no way to obtain small quantities of a proper separator. At least, I couldn't find any online.
Now to my actual questions:
What other materials could be used as separators?
Which one would be the most suitable to buy at the moment?
Has anyone of you been able to obtain small quantities of a separator material?@SamAuc Congratulations on your first tests with the kit! Please share some pictures of your setup with us if you want/can, we're always excited to see new people build our kit from scratch.
If the migration is happening quickly - within a few minutes you see volume increase on one side vs the other - chances are the photopaper is getting pierced and electrolyte is just flowing straight through the separator or you have internal leaks. In the case of the internal leaks the electrolyte is able to bypass the membrane by flowing between the membrane and the gasket. Try cutting the photopaper membrane squares larger, as large as you can with the corners almost touching the screws. Also make sure you start flow on both pumps at the same time and ramp it up slowly (take it to a max of 25-40% of total pump power, more isn't really necessary), this will prevent flow ripping holes through the photopaper because of pure pressure.
If the migration is happening slowly (takes a few days) then this is due to hydrostatic pressure, it will happen regardless of which microporous separator you use as it happens because of water migration between both sides due to differences in their ionic strength. In this case the solution is to adjust the speed of the pumps (have one run faster than the other) so that migration doesn't happen. In Zn-I devices solution tends to migrate to the anolyte, so the catholyte normally has to run faster to prevent this from happening, some research papers have found 7x faster to work best.
About other separators, you can make the photopaper more resistant by treating it with PVA. Prepare a 5% w/w solution of PVA and then cover the photopaper and allow it to dry, usually 2-3 passes are enough to heavily reduce the porosity of the photopaper.
Other separators do exist, commercially we would use a polyethylene based separator, like the ones Daramic offers (sadly only b2b). We have been able to obtain a significant amount of separator from them for research (for free, to help us develop the kits) and now that we have started a company to sell the kits we will likely buy a significant amount from them so that we can sell at small amounts in retail so that users of the kit can buy and experiment with these separators.
-
First of all, congrats on getting it to work!
Disclaimer: I have almost no clue but am trying to learn. About the migration, is this about the volume of electrolyte (the water basically) migrating, or rather the charged redox species migrating? In any case both are probably normal and might be alleviated by using a thicker separator. Afaik the trade off might be that internal resistence might go up. If I understand correctly, both the parasitic reactions (redox species crossing) and added cell resistence would lead to a less efficient cell. The first reduces coulombic efficiency while the second reduces voltage efficiency. @danielfp248 did I get that right?
If I understand correctly, there are fundamentally two types of separators. Selective membranes and microporous separators. The first are usually pretty expensive and probably hard to source but can prevent migration much better.
@sepi Thanks for your reply. Actually a thicker separator reduces cross-over so it tends to increase the CE of the cell at the expense of VE. This is a known tradeoff of cells using microporous separators. Thicker membranes will always perform better in CE and worse in VE and vice versa. In the end we tend to choose the thickness that maximizes the overall energy efficiency of the device.
You are also right about types of separators. Ion exchange separators are also not hard to source - you can easily buy Nafion from aliexpress or amazon - but they are just very expensive. Another problem is that ion selective separators will get damaged by dendrites (once they are pierced they are gone), so they are quite incompatible with Zn chemistries due to this reason (and expensive to test as each test likely ruins a separator).
-
@danielfp248 @sepi Thank you for your replies.
The migration is not noticeable at first, but becomes apparent during discharging. The longer the photo paper is in use, the more it leaks.
I accidentally purchased the incorrect pumps.
They are larger than intended, and the flow rate is too high even at the lowest speed. I have already tried different speeds for both pumps, but this has only helped to a limited extent.I find the topic of separators very interesting, and it seems to me to be a very complex subject with much research still to be done.
-
@SamAuc Could you maybe bend the pipes a bit to reduce the flow? That might not be the best idea with a peristaltic pump though. Do you know what kind of motor your pumps use? You might be able to somehow hack it to make them slower.
About the separators: I'm sure there is a lot of optimization to be done in this domain. Especially if you want to design a separator for a specific chemistry. I also imagine that you can achieve quite a lot using day to day materials. I'm interested in trying out ceramics at some point. Unfortunately I'm still waiting for my damn pumps to get started with any experiments.
-
@danielfp248 @sepi Thank you for your replies.
The migration is not noticeable at first, but becomes apparent during discharging. The longer the photo paper is in use, the more it leaks.
I accidentally purchased the incorrect pumps.
They are larger than intended, and the flow rate is too high even at the lowest speed. I have already tried different speeds for both pumps, but this has only helped to a limited extent.I find the topic of separators very interesting, and it seems to me to be a very complex subject with much research still to be done.
@SamAuc Excessive flow rate causes both internal leaks and separator failure. It is critical to use the correct pumps for this reason. I would recommend changing the pumps, if this isn't possible, the reinforcing the membrane with PVA can help but it definitely won't protect you from internal leaks caused by overpressure from the pump.
-
Just a thought when using oversized pumps, try adding a bypass valve. With the valve fully open, the output of the pump just goes right back to the input. Close the valve a little at a time to get the pressure you want.
@Vorg Great idea.
-
Fuel injection works this way on cars. They use a bypass regulator to give you a smooth continuous flow. Pump is always pumping and extra goes back to the tank.
Another thought, someone might be able to come up with a regulator that balances each side. Pressure goes up on one side, it reduces that side and raises the other.
-
The idea of a bypass valve has also crossed my mind.
Hoses with a smaller diameter are already installed.
I have conducted a few tests on the flow rate. Currently, at low speed, I have a flow rate of 70 ml/min.@SamAuc We would want to be at 20-40mL/min for a reservoir this size.
-
Fuel injection works this way on cars. They use a bypass regulator to give you a smooth continuous flow. Pump is always pumping and extra goes back to the tank.
Another thought, someone might be able to come up with a regulator that balances each side. Pressure goes up on one side, it reduces that side and raises the other.
@Vorg said in Separator material:
Another thought, someone might be able to come up with a regulator that balances each side. Pressure goes up on one side, it reduces that side and raises the other.
You mean a governor that keeps the volumes on both sides equal by adapting the relative speeds? The only issue is measuring the volume. I guess that's not worth it for an experimental cell but it would probably be vital for a real world cell.
-
@Vorg said in Separator material:
Another thought, someone might be able to come up with a regulator that balances each side. Pressure goes up on one side, it reduces that side and raises the other.
You mean a governor that keeps the volumes on both sides equal by adapting the relative speeds? The only issue is measuring the volume. I guess that's not worth it for an experimental cell but it would probably be vital for a real world cell.
@sepi We have a model for this approach. It is basically a mixed reservoir that is communicated at the top, so if it overflows one side you basically will just sacrifice capacity to rebalance the cell. You will need a separate stand for it though as it isn't compatible with out base jig (https://codeberg.org/FBRC/RFB-dev-kit/src/branch/main/CAD/exports/Double Reservoir.stl). A common approach in research is to communicate the reservoirs with a needle valve to have them automatically balance on overflows.
-
@Vorg said in Separator material:
Another thought, someone might be able to come up with a regulator that balances each side. Pressure goes up on one side, it reduces that side and raises the other.
You mean a governor that keeps the volumes on both sides equal by adapting the relative speeds? The only issue is measuring the volume. I guess that's not worth it for an experimental cell but it would probably be vital for a real world cell.
@sepi It's the pressure difference that rupture the separator correct? So what I was talking about is some kind of double bypass valve (one for each side) that is linked together to keep the pressure the same on each side. Might be too complicated and likely yuo can already buy something like that though $$$. Anyway, just a thought.
-
@danielfp248 ah, that makes sense as a last resort measure but would it not be nice to not lose charge? Well it all depends on how easiy it is to counteract by setting a different flow for the different electrolytes.
