Flow Battery Research Collective

@gus Awesome, feel free to post screenshots of your testing! You can also post some of your result files here if you need any help plotting the results.

@gus It doesn't matter what side you connect to WE/SE the first time you run the setup. The pumps aren't really associated with the electrodes at all on the program. In a new setup both sides are exactly identical.

The important thing is, whichever side (pump, reservoir, electrode) you use for WE/SE the first time, always use the same side for WE/SE, always same pump, same electrode side, same reservoir, etc.

@gus It doesn't matter on your first run, because the cell is perfectly symmetrical on start, but remember how you connect it as you would want to always connect it the same, this way elemental iodine will be limited to only one side of the cell. I always mark one of the endplates with an X to remember which one I connect as an anode and which one as a cathode. Where you connect WE/SE is where you will generate triiodide (your cathode) and where you connect CE/RE will be you will deposit Zn (your anode).

@gus no, that should be fine. Since all the tubing is at the front, leaks are usually contained to the front of the cell, I've never had a leak splash back and damage the Arduino, even when we didn't have it inside a box but just bare.

@sepi Glad to be of help! I'll ask @kirk who is currently in charge of the documentation to make the additions.

@gus Great! As a first test please run the following:

This should take around 1 one hour per charge/discharge cycle, experiment should take around 5 hours total. If the cycle ends because the potential reaches the upper safety limit too quickly, reduce the currents to 10000 and try again. At first the cells can require some time cycling at low SOC at lower current, to build all the Zn nucleation sites. Do not cycle to a potential higher than 1.7V because you will start having nasty side reactions at this point.

If this cycles successfully you can then increase the currents to 30000uA and repeat, see that it goes well.

After that you can then start going to high SOC values at 40000uA. I would recommend first cycling to 100mAh (set Upper bound to 100000uAh). Enclose the battery when cycling to higher SOC - you can put it inside a plastic tub - because leaks due to any problem will spray highly charged electrolyte, which, even if the volume is low, can be dangerous.

If you let me know how each test goes I can provide further feedback.

@gus Nice work! Please let me know what electrolyte, concentration and volume you want to run and I can give you some guidance on exact settings.

@sepi We are in Europe and we sourced everything as private individuals, so it can be done! Let us know if we can be of any additional help.

@sepi Thanks for your reply! We would love to, but sadly Aliexpress/Amazon sellers change their links every few weeks (my guess is to stay relevant with the search algorithm of the platforms), so if we added them they wouldn't work in a month.

Thanks for posting! About your material choices.

1. First one is ok for the silicone gasket material. Get 0.5mm silicone, which is the one we have tested.

2. This is not adequate. It needs to be conductive graphite felt. For example this one. We use 3mm so the best deal would be the 3x200x1230 one piece.

3. That is the correct link for the pumps. Make sure to get the KPK200 24B motors. Send them a message after you make the purchase to specify the pump. Make sure to be clear about needing the specific pump model because you have a project with dimensions for that specific pump. They have sent me more expensive pumps in the past because these were just not in stock. Tell them it HAS to be KPK200 24B. These pumps are expensive (probably close to 70EUR each), the cheaper versions just don't work for this purpose. Trust me, we tried every pump tier starting from 3EUR and this was the cheapest pump that would run reliably without getting killed by the chemistry or the stress of running continuously pumping the higher viscosity fluids. Cheaper pumps are not strong enough to compress the chemical tubing and the softer tubings leak active material that destroys the pumps at the contact points.

4. This is not adequate, it has to be a graphoil sheet. Like this one. We have routinely used 0.5mm sheets. Graphoil used to be readily accessibly through Aliexpress, but China has recently banned the export of several carbon based materials, including graphoil. Alternatively thin Titanium foil could also be used as an electrode material for the cell. Even 0.1mm Ti foil should be adequate (like you can find here).

5. This is not the proper tubing. The specific tubing from Kamoer we ordered (which is a PTFE lined BPT), is called "Tygon Chemical tubing" by them, they don't have an Aliepxress link for it but if you contact them directly they can sell it to you. The closest tubing they have a link for would be this BPT (here). The diameter would be 4ID 6OD.

Let me know if you have any other questions!

@sepi Unless you are very experienced soldering small components I would suggest not doing this. Some of the expensive components are tiny and easy to damage, either by overheating or by soldering multiple legs together and causing shorts. I tried soldering my own mystat 3 times (without success) before I got my first working one from pcbway. I am however, not good at all at soldering.

@Vorg I'm no expert in manufacturing, if you find anything that would work and you want to try it please feel free to share your results!

Since our design prevents direct contact with the current collector, almost any metal should work. you could even use aluminum foil if you really don't want to machine anything (although I wouldn't recommend it! lol). Bear in mind that the graphite foil is not impervious to the electrolyte, it diffuses slowly through it, so if you have highly charged electrolyte some reactivity with the current collector will happen through time (across weeks of cycling).

There is a paper on creating an electrode using graphoil and wax that might be much better (https://chemistry-europe.onlinelibrary.wiley.com/doi/abs/10.1002/slct.202103996). I will probably get a rosin press to test this and construct some better quality electrodes for our kits.

I think a potentiostat is a must if you want to do proper quantitative experiments. However, not all experiments require it, you can definitely use a power supply with the ability to set constant current to charge the cell and then discharge it across some load to make it work. Using the time it takes to charge and discharge using a setup like this you can get charge and discharge capacity values, but Coulomb efficiency, Energy efficiency, etc will be outside your reach.

Potentiostats can also make experiments safer as you can establish limits for upper voltages or currents, make sure you do not overload the cells. It is important to not do this as this can cause clogging, gas generation, etc.

A potentiostat is the heart of modern battery research, I would give it a high priority if possible.

The cost of a single mystat is usually in the 150-200USD range, depending on the availability of the components. If you get 5 or more the cost per unit can drop a little bit, but not a lot below that.