This is interesting so I thought I would share.
Been looking at a few RFB's available on or to the market in order to learn more about them and one type of general interest are the all organic variant, as in an organic molecule is used in both the anolyte & catholyte. One such organic molecule is "fairly" complex Anthraquinone or just Quinone in some cases, which is the basis for what Kemiwatt us using. They have a few patents describing their process of "readying" the molecule if you are interested in reading about it - see under sources.
Using organic molecules can be very attractive since it imply that its natural and therefore harmless, this is a bit misleading since harmless fully depend on not just what kind of molecule we are talking about, but also the amount and form of exposure.
What I wanted to get at with this post is that in Kemiwatt's patent, an energy density of 3,7 Wh/L is mentioned. In the hybrid energy article, 7,8 Wh/l is mentioned, so I guess we should expect something along those lines for the organic flow battery. That is not very impressive... even if the chemistry is by far more friendly than Vanadium which is considered toxic, but does yield greater energy density in the range of 15-25 Wh/L . Vanadium redox is prepared using sulfuric acid and the cell chemistry is highly acidic. The Kemwatt cell is on the other hand, highly alkaline.
Iron-Iron will achieve roughly 20 Wh/L and a potential of 1.2V
Zinc-Bromine, upwards of 65 Wh/L with a potential of 1.8V... but elemental Bromine is toxic.
Zinc-Manganese .. upwards of 60 Wh/L and a potential of 1.2V ... but zinc forms dendrites.
Most of the above is usually acidic or highly alkaline while Daniels Mn/Fe cell is pH neutral while maintaining 40-80 Wh/L at a potential of 1.2V. The higher Wh value ofc must be validated and cell health investigated over time, but the premises for a relatively high energy density while keep the cell pH neutral with a long cycle life is very good and this is an important selling point.
Anyway, just excited and wanted to share some perspectives.
Sources:
- https://cordis.europa.eu/project/id/101027793
- https://kemiwatt.com
- https://hybris-project.eu/high-level-concept-of-the-hybrid-energy-storage/
- https://worldwide.espacenet.com/patent/search/family/076807580/publication/EP4106060A1?q=KEMIWATT
- https://en.wikipedia.org/wiki/Anthraquinone
- https://en.wikipedia.org/wiki/Quinone
- https://en.wikipedia.org/wiki/Vanadium_redox_battery
- https://en.wikipedia.org/wiki/Vanadium
- https://en.wikipedia.org/wiki/Flow_battery
- https://en.wikipedia.org/wiki/Zinc–bromine_battery
Thanks for sharing what you have been reading!
Organics in flow batteries are definitely very interesting from a research perspective and likely the future in 20+ years, as the molecule manufacturing gets scaled and more stable molecules are found.
I wanted to find a chemistry using things that you could buy off the shelve right now and that, in bulk, could be purchased for less than 10 USD/kg. Both of the molecules I intend to use have been validated in different papers on RFB, so we know they have reversible redox behavior (this far I have verified personally using cyclic voltammetry as well).
RFB using Fe-EDDHA|Fe(CN)6 -
https://www.sciencedirect.com/science/article/pii/S0013468622011999
RFB using Zn|Mn-EDTA -
https://www.sciencedirect.com/science/article/abs/pii/S0378775321008119
By probably good luck, their highest solubility values are located near neutral pH (although their solubility at pH 7 is something I have to confirm since there are no publications, to the best of my knowledge, addressing this directly). They are also readily available as bulk fertilizers (you can buy them in bulk in the EU, no problem). Another important thing is that I wanted to avoid any plating, to make sure dendrites are never a potential problem. This also makes the capacity directly proportional to tank volume and fully decoupled it from the electrode area (as in the case of a Vanadium RFB).
The reactions also fit very snuggly in the potential window of water at pH 7, there should be no hydrogen evolution issues on the anode or oxygen evolution issues on the cathode. Something that acidic RFB often have problems with. The 1.2V potential of a cell with Fe-EDDHA and Mn-EDTA at neutral pH is great. I have also confirmed this to be the potential achieved with this chemistry. Having neutral pH was really important for me, because I don't want to be working with circulating solutions of concentrated acids in my home.
Another great benefit, is that ALL of the electroactive species are bulky anions.
(Fe-EDDHA)-2
(Fe-EDDHA)-1
(Mn-EDTA)-2
(Mn-EDTA)-1
This means that they are blocked very effectively by a cation exchange membrane (like Nafion or the DIY PVA membrane I discuss on my blog). Vanadium RFB have problems with energy efficiency attributed to the crossover of Vanadium cationic species, no such issue will happen with this battery.
To anyone interested in RFB I would recommend this talk:
It goes through a lot of the current technologies, the issues present and the future of this technology.
My proposal of Fe-EDDHA|Mn-EDTA suffers from several foreseeable issues:
- Molecular mass of the compounds is large, so the mass per electron moved is large. This increases the cost for scaling up as you need more kg/L per Wh/L. This limits energy density to, worst case, 20Wh/L.
- The large molecular mass also implies potentially slower kinetics because of likely slower diffusion coefficients. This makes it likely that efficiency losses at high current densities will be higher.
- Organic molecules are present, which means you will have some degradation as a function of time. Both of the above mentioned papers cycled Mn-EDTA and Fe-EDDHA hundreds of times with no problems, but this does limit the electrolyte life as these organic molecules are going to slowly get destroyed.
With the above said, my objective is not to make this the end-all-be-all energy storage solution, but at least provide something that could be put together at lower technological complexity that at least does a great job and that does it safely.
For the constructions of the cells, I am planning to use this paper - although I will scale down the cell around 30% - and a 3D printer:
Peristaltic pumps, Arduinos and my current DIY USB potentiostat will be doing the rest of the job.