Flow Battery Research Collective

Thx Kirk for sharing, I almost missed this important release. Awesome documentation as always with great methodology and insights!

Oh, pardon me. I probably used the wrong term. I noticed that the catholyte tank electrolyte level had risen higher than the anolyte and there was electrolyte migration. I guess I "re-leveled" the electrolyte. There is a 1/4 inch (6.35 mm) shunt tube between both tanks but at the time, the shunt was not able to keep up with the migration. Since I do not have a pump or a siphon or extra valves and tubing that are acid resistant, I blocked the gas pressure equalization tube between both tanks, reduced the pressure in the anolyte tank, and overpressured the catholyte tank until the atmospheric pressure on the catholyte allowed more flow through the shunt. Then I restored the pressures to as expected. In recent times, the shunt has been able to keep up with any electrolyte migration pressures.

Time-stamping the electrolyte formulation for Zn-I, I am making the documentation to build the dev kit chemistry agnostic (so I am taking these weights out of the docs for now) zinc chloride - 1.36 grams potassium iodide - 3.32 grams ammonium chloride - 1.07 grams deionized water - 8.50 grams triethylene glycol - 0.55 grams This usually ends up preparing around 11 mL of solution which we split evenly between the two reservoirs.

I did a google search on using fiberglass to stop dendrites and some stuff did come up.

from this paper, seems like nanomaterial synthesis needed: https://www.nature.com/articles/s41560-026-02091-w#Sec25 also they use triflate salts --- that's likely more expensive than the Zn (and, fluorine-containing, with all that entails)

Did a few cycles at low SOC and then attempted to cycle at 10Ah/L. [image: 1782403366618-c6cbcd66-b4d5-4594-97c7-8266ae815951-image.jpeg] Catholyte (left) anolyte (right) below: [image: 1782406513535-d429c06d-318c-4865-b5a4-8eab115d9f66-image.jpeg] pH of catholyte was 0.24 and anolyte was 4.7. As you can see on second cycle to 10Ah/L the CE dropped from 93% to 83% and then capacity continued to decrease. On disassembly a lot of unreacted passivated iron remained on the anode, likely due to the pH increase. As Fe deposits water activity likely increases, which increases HER, which increases pH and leads to Fe passivation. I am going to run a test adding 1M ZnCl2 to the electrolyte, to see how the additional salinity changes water activity and metal passivation.

@bipolaron @kirk This project in Switzerland is currently building a 2.1 GWh capacity / 1.2 GW output redox-flow battery. Scheduled to go online in 2029, and by their own PR material "the largest in the world". https://www.swissinfo.ch/eng/climate-solutions/switzerland-builds-worlds-most-powerful-redox-flow-battery/91181119

I caught the version on YT, but having the same vid on PT is great.

I'm no chemist, so this may not apply here, but I've been looking at info on different batteries. I looked at Nickel/Iron because of their known long life. One change mention from original Edison design is adding a little Lithium oxide or dioxide ( forget exactly) to the electrolyte To extend the life of the Iron electrode. There was also something about making plates (instead of just using sheets of iron) using iron powder with carbon (for conductivity) and something else which I forget that reduced the production of hydrogen.

Edit: after original post, realized that one connection lead was loose, and the negative pump tubing was blocked/degraded---PTFE liner had detached from the wall and formed some sort of blob inside the tubing, it seems.

The battery looks interesting but has some problems. From what I've read, it tends to use up the water in the electrolyte fairly quick by splitting into hydrogen and oxygen. Also, It's working voltage is ~1.2v but it needs 1.6v (1.65v if added lithium) to take a change? The inverter I'm leaning towards now, GTIL2000, for low cost and simple connection has an input range of 45v-90v. When using a 48v LFoP battery, it only can produce about 1300w. With a 72v, it's around 1700w according post on a solar forum. One of these batteries with enough sells to put the charge voltage just below 90v with give me a battery that only has about 60v when no solar to charge it. That would really cripple the inverter. Here is a few videos of someone messing with this battery making his own electrodes and a gell electrolyte in an effort to improve the battery. https://www.youtube.com/watch?v=NaOzDt83XWY https://www.youtube.com/watch?v=0mYaei0O1sU https://www.youtube.com/watch?v=pjoxC4kwA9I

https://www.youtube.com/watch?v=t09aBKUejSs&list=PLYDTEjzfnaSVHeUiCOX5Z9G7i_AiSifpR&index=10

@idlestate also, some studies of these systems I think point to complexes being formed in solution, that contain both chloride and iodide/iodine, like ICl-, for example. They might show up with some spectroscopic techniques like Raman? So chloride can play more of a role than just supporting electrolyte.

@Methylzero "I heard you like batteries, so I got you a battery for your battery..." Definitely it's backing up a headless Raspberry Pi, a potentiostat, Arduino, and two peristaltic pumps.