diy solar

diy solar

My adventures building a DIY Mn/Fe flow battery

How come you "stumbled upon" Ligand or the Mn- and Fe-chelate and decided to experiment with it and not Lignin (Lignosulfonate). Do you see advantages and disadvantages (for both) that eludes others ?
For a chemist, metallic chelates are very familiar, they have very reversible electrochemistry and - Fe and Mn ones - can be bought for affordable prices. They are also very well studied. Things like Lignosulfonates are much less studied, so I would have never considered something like this for a DIY approach. Families of molecules that are less studied will require substantially more basic research, which requires more time and equipment.
 
So I have some good news and some bad news.

On the good side, I confirmed the commercial Fe-EDDHA standard potential at pH 7 to be around -564mV Vs Ag/AgCl. I also confirmed the reaction to be reversible, as expected. Given the redox potential for Mn-EDTA +660mV, a battery using both reactions would have a top potential of 1224mV. Below you can see a CV experiment with Fe-EDDHA. My resolution isn't very good atm (my working glassy carbon electrode needs polishing) but you can see the redox peaks.

1673440342897.png

On the bad news, turns out that a cellulose+PVA membrane is not robust to Mn3+, the oxidized Mn is a decently powerful oxidizer and reacts with the membrane, slowly dissolving it through time. I know it is dissolving the membrane because I can see gas bubbles on the membrane only on the Mn side and I can see the red color of the Mn3+ fading as it is consumed by the membrane.

This fade is not due to discharging due to crossover, as there is absolutely no Fe-EDDHA crossover (because it would be obvious due to the color change, even at minute concentrations). Also crossover to cause so much discharging would need to be quite massive.

Most probably any organic membrane containing either carboxylic or alcohol groups will not be robust to Mn3+. This battery would therefore need to work with a sulfonated - sadly much more expensive - membrane.
 
Last edited:
1224mV is in the grand scheme of things very useful. It is not as high as some other common chemistry's, but very useful. So good news indeed. Are you content with that potential ?

I did a quick search for sulfonated membrane and there are several papers/study's. I'm not going to pretend I know the slightest, so I'll happily take the backseat on this one .. but lets mourn the cellulose+PVA membrane, it was a good contender.
 
1224mV is in the grand scheme of things very useful. It is not as high as some other common chemistry's, but very useful. So good news indeed. Are you content with that potential ?

I did a quick search for sulfonated membrane and there are several papers/study's. I'm not going to pretend I know the slightest, so I'll happily take the backseat on this one .. but lets mourn the cellulose+PVA membrane, it was a good contender.

I am very happy with the voltage, that is more than I thought I could get for an all-soluble water chemistry with chemicals already available in bulk at neutral pH.

Sulfonated membranes are very well established. You cannot DIY them easily though, because the synthesis involve sulfonating agents, which are all quite dangerous to handle (chlorosulfonic acid, concentrated sulfuric acid, etc). So this means buying commercial ones. SPEEK membranes are the cheapest - still quite expensive - and should work fine for this purpose. This is probably the cheapest available (https://ionexchangemembranes.com/cmi-7000-cation-exchange-membranes-technical-specifications/).
 
I am very happy with the voltage, that is more than I thought I could get for an all-soluble water chemistry with chemicals already available in bulk at neutral pH.

Sulfonated membranes are very well established. You cannot DIY them easily though, because the synthesis involve sulfonating agents, which are all quite dangerous to handle (chlorosulfonic acid, concentrated sulfuric acid, etc). So this means buying commercial ones. SPEEK membranes are the cheapest - still quite expensive - and should work fine for this purpose. This is probably the cheapest available (https://ionexchangemembranes.com/cmi-7000-cation-exchange-membranes-technical-specifications/).
excellent work.

Do you believe flow batteries can/will replace Lithium Ion in Automotive applications. If so, when?

Also have you seen the Iron Air Battery plant scheduled to be opened in WV? I haven't heard of Iron Air batteries only Zinc Air from years ago.
 
I am very happy with the voltage, that is more than I thought I could get for an all-soluble water chemistry with chemicals already available in bulk at neutral pH.

Sulfonated membranes are very well established. You cannot DIY them easily though, because the synthesis involve sulfonating agents, which are all quite dangerous to handle (chlorosulfonic acid, concentrated sulfuric acid, etc). So this means buying commercial ones. SPEEK membranes are the cheapest - still quite expensive - and should work fine for this purpose. This is probably the cheapest available (https://ionexchangemembranes.com/cmi-7000-cation-exchange-membranes-technical-specifications/).
Alright, then I am happy too :) and indeed, 1.2V via water based electrolyte and neutral pH is really outstanding.

What would be the discharge voltage according to this test ? .. (I am yet to learn how to interpret)

The CXM-200 is way cheaper than the dialysis membrane I hinted at in an earlier post, so I am not complaining there and it does make the assembly of a cell simpler. Membranes aren't always the easiest to deal with or produce and often contain many expensive steps and like you say, hazardous chemicals which really require one to have proper training and a lab. So this is a very good option and if something else pops up, then one can consider the options.

Do you know the life expectansy of the CXM-200 in solution ? if you don't know, I can send Membranes International Inc. and check.

BTW: The states of the metal species, are they Iron(II) and Iron(III) as well as Mn(II) and Mn(III) ?
 
Last edited:
Do you believe flow batteries can/will replace Lithium Ion in Automotive applications. If so, when?
Unless you are willing to drive around with a trailer containing big tanks of solution weighing a few tons, then no :) - Look at it this way. Tesla's Li-ion sits at roughly 272-296 Wh/kg while lead acid as an example is 35–40 Wh/kg in comparison and the reason I mention lead acid is that this is a good energy target for flow battery's and non Lithium. Lead acid is 3.5 times less energy dense and not a viable solution if you want decent driving range. Doable sure, but practical.. no. Mobile applications are different compared to stationary and for stationary energy storage, this MnFe flow battery is showing some remarkable performance.
 
Last edited:
To add to this debate in regards to mobile solutions lets compare:

Non renewable:
Gasoline: 12,2 kWh/kg ..

Renewable:
Ethanol: 7,85 kWh/kg
Birch (wood): 4,25 kWh/kg
Pine (wood): 3,74 kWh/kg

Renewable to an extent.
Li-Ion battery: 0,15-0,296 kWh/kg
NiMH battery: 0,10 kWh/kg
Lead Acid battery: 0,04 kWh
Flow Battery; 0,01-0,04 kWh ... (various types)

If we compare the top 4 energy sources: Gasoline, Ethanol and wood, they are by far many times more energy dense than any battery chemistry currently available and while we want to move away from fossil non renewable fuel, battery's aren't and cannot be the solution to everything. While wood gasification (85% efficient) can't serve the entire planted, it can serve a smaller population and is carbon neutral, plus, it would give many ICE's a second life. Brewing some Ethanol is yet another solution over battery's.

Looking into the future also means looking at our past and often the future doesn't contain just one solution, but must rely upon many.
 
Back to the MnFe flow battery. I played with some numbers for fun but these will change as new and more relevant data starts flowing in at a later stage.
 
Last edited:
Back to the MnFe flow battery. I played with some numbers for fun but I am sure these will change as new and more relevant data starts flowing in at a later stage, but for the time being and if we use 20mA/cm^2 and A4 area for size (21,0 x 29,7 cm):

21 * 29.7 cm = 623.7 cm^2
623.7 cm * 20mA = 12,474 mA > 12,474 mA / 1000 = 12.47 A

1.224V @ 12.47A and 15.26W

If we connect 40 pcs in series, we end up with

1.224V*40 = 48.96 V
12.474*40 = 498.9 A
15.26*40 = 610 W or 0.61kW

At around 300 USD per m2 of membrane that would cost around 750 USD, not counting import taxes. Probably a little bit above 1000 EUR when all those taxes are paid. This membrane is almost always the most expensive part of a flow battery.
 
I did check their price list and the price will drop a bit when buying in bulk, but yes, this will be the most expensive part of the system.

CXM-200 (CMI-7000S)
A 1.2m*0,5m sheet cost $200 not including shipping an taxes. If I add taxes and VAT which is around 30% and convert it to €, then 241€ as of the time of writing this.

A 1.2m*0,5m area is equal to 0,6m^2 (sq-m) and 6000cm^2 (sq-cm). The cost per cm^2 is therefore 241€/6000 = 0,040€ or 40 euro cents per sq-cm.
 
Last edited:
SO what is the plan now that we know the cellulose+PVA isn't working, are you looking to obtain some CMI-7000S samples for the demo flow cell ?
 
SO what is the plan now that we know the cellulose+PVA isn't working, are you looking to obtain some CMI-7000S samples for the demo flow cell ?

That would be the plan. It will take me a while to save the money for all the things I need (3D printer, membranes, electrodes, pumps, etc), so it will probably be 6-12 months before I can even do some tests.
 
That would be the plan. It will take me a while to save the money for all the things I need (3D printer, membranes, electrodes, pumps, etc), so it will probably be 6-12 months before I can even do some tests.
I read the 3D cell paper and was surprised that they didn't try PETG filament. It is very similar to PET, which is what most drink bottles are made from. It has low expansion, making it much easier to print than ABS. I have not tried to print PP, but it sounds like it didn't work well for them anyway.
 
I read the 3D cell paper and was surprised that they didn't try PETG filament. It is very similar to PET, which is what most drink bottles are made from. It has low expansion, making it much easier to print than ABS. I have not tried to print PP, but it sounds like it didn't work well for them anyway.

They didn't try PETG because it is well known to be unstable against sulfuric acid and strong oxidants, so it would definitely be unstable in a vanadium flow battery environment.
 
They didn't try PETG because it is well known to be unstable against sulfuric acid and strong oxidants, so it would definitely be unstable in a vanadium flow battery environment.
Gotcha. I wasn't sure about that.

Perhaps it would be a filament variety to try for your neutral pH electrolyte. It is just about all I print any more since it prints so well and is cheap. Do you have the .stl files for the cells you want to print?
 
Got it. It takes time to develop projects under strict finances, but something is always better than nothing :)

********
I contacted Membranes International with this question and they replied with in quote

Q: What is the expected life span of your CXM-200 in 3M NaCl solution ?

Good morning.

The expected shelf life of the membrane in the dry state and packaged as shipped is one year. The shelf life in solution is short term. In addition to the salt solution it is recommended that a biocide be added as well.

Regards, Customer Service, Membranes International Inc.

I sent a follow up: "Can you clarify the short term in solution please. I am considering using this membrane in a water based solvent with sodium chloride support (NaCl) and Ligand active ingredients, pH neutral, in a long term situation for a redox flow battery." .. waiting for reply.

If their answer is as expected, then we press forward.
 
I mentioned on Tuesday that my entry's would slow down a bit while we wait for updates from Daniel - well, that changed yesterday. Since I am not properly trained nor have a degree in chemistry (besides high school lvl), for me to find solutions that might be beneficial are slim, but I did mention that I want to contribute as much as possible and I might just have something. I say might because this could be a dud for reasons beyond me.

I sat down and did a search for "diy Nafion" on YouTube and indeed, a video popped up [How to make alkaline membrane for fuel cell] and that seemed relatively straight forward and I was going to mention this video and now I have... LOL. I figured I would check the channel for other interesting stuff and indeed, there was one. [Battery using ion conducting ceramic separator and dual electrolyte]. If you scroll down and read in the comments section, he does mentions - translated from Portuguese - in quote since it is only supporting info:


I did this experiment a while ago and ended up posting the video. What is not very evident in the video is the fact that I used a traditional ceramic water filter (the common ones that go in pots) and then I modified it so that the pores were in the nanometric order. This experiment was also made from a paper that talks a little about the modification of some materials. Then the filter ceramic starts to act as a nanometric separation membrane and I demonstrated it using 2 electrolytes with different pH. When this happens I can raise the battery voltage using a water based electrolyte. Ceramics in this case is useful because it resists well to more aggressive electrolytes. It is more difficult to use polymeric membranes for different pHs, they break down and also do not have good selectivity. But here it was just a test, you can explore this subject much more, I found it interesting to post here because the power generated in the engine was quite considerable.

and...

In fact, I hadn't made this video for YouTube. I only recorded it because I wanted to document it, so I didn't worry about the audio. I decided to put it later to illustrate how it works in a rudimentary way with "dual electrolyte" batteries. This type of battery is very recent, and is currently being studied a lot. I used a ceramic filled with ionic material to contain the 2 different electrolytes and the result turned out to be much better than I expected. The power of this assembly was very considerable, see that I am using an engine that needs a higher amperage to function. There is an interesting theory about this type of assembly that I want to explain in future videos, that's why it reaches 2.4v.

Ceramic ion filter, interesting. So I did a search for ceramic water filter and many hits appeared. I expanded the search to "ceramic ion exchange membrane" and after scrolling a bit .. ding ding ding. Something of huge interest popped up and my gut feeling told me this could be it. This could be the membrane this cell needs in order to be even more cost effecting than using CMI-7000 and the overall cost of that membrane was to be honest, one of the reasons I wanted to explore some options. I have incorporated the interesting bits from the article:

Catalyzing Commercialization: Flexible Ceramic Membranes Enable Low-Cost Energy Storage - Feb 2019

Molecular separations are at the heart of chemical engineering, and nanostructured membranes are emerging as a low-cost alternative to traditional separation processes. However, most membranes struggle to maintain performance in harsh environments. For example, redox flow batteries (RFBs) rely on a cation exchange membrane to store energy. To create an RFB capable of storing enough energy to power a city (>1 MWh) for 20+ years without capacity loss, an extremely robust membrane chemistry would be required, such as a perfluoro sulfonic acid (PFSA) membrane (NAFION etc). However, PFSA is expensive to manufacture, has mechanical and performance issues, and generates substantial quantities of bio accumulative and toxic substances during production. In many cases, PFSA membranes account for more than 20% of the total battery system costs. Researchers have been searching for more than three decades to find a material capable of delivering better performance at a lower cost.
With funding from the National Science Foundation (NSF), researchers at Membrion, Inc., have patented a potential solution that uses the desiccant commonly found at the bottom of a beef jerky package. Membrion has developed a self-assembly process for silica gel that enables its use as a low-cost and high-performance ceramic cation-exchange membrane.

“We were trying to lower the costs of RFBs by finding a cheaper alternative to PFSA that could handle the extreme conditions. Ceramics seemed like a logical choice, but they tend to be brittle and not well suited for the high compression in stack applications like RFBs. Our real innovation was in making a flexible and compressible ceramic membrane,” says Membrion’s Founder and Chief Technology Officer, Greg Newbloom.

This novel membrane processing method starts with a nonwoven glass mesh commonly used as roofing insulation in the construction industry. Membrion scientists add a polymeric edging to the outer rim of the mesh to act as a gasket for the compressive loads in stack environments. They dip this polymer-edged glass mesh in a silica solution that wicks into the mesh, and then dip the mesh in a commodity acid. The end product is a percolated nanoporous ceramic membrane with a 2-in. bending radius.

“The acid dip is where the real science happens,” Newbloom says. “The acid causes the silica to electrostatically destabilize and form a network of nanopores. By controlling the kinetics of the process, we can make pores of different sizes and network structure. We use size and charge screening to dictate which molecules pass through the membrane and which are rejected.”

20191115_144338-scaled.jpg
1-s2.0-S0015188220302354-gr1.jpg

1-s2.0-S0015188220302354-gr5.jpg



Membrion, Inc have two patents which is linked to under sources.
1. WO2020247467A1 CERAMIC CATION EXCHANGE MATERIALS
2. WO2020247472A1 CERAMIC ANION EXCHANGE MATERIALS

Closing words.
If this membrane turns out to be a viable option, then it would be advantageous to incorporate a diy process just like Daniel was working on regarding the cellulosa+PVA. Now, I don't know how much our chemists know, so I dug up an article / book called: Current Trends and Future Developments on (Bio-) Membranes - Silica Membranes: Preparation, Modelling, Application, and Commercialization - Chapter 1 is literally called: Chapter 1 - Preparation of Silica Membranes by Sol-Gel Method - the book is linked down bellow. I will take a look what it has to offer.

I sincerely hope this membrane is the solution because if so, then this battery will be made using 3 of the most abundant elements: Silicon, Iron and Manganese - wouldn't that be awesome

... waiting for feedback

Sources:
 
I mentioned on Tuesday that my entry's would slow down a bit while we wait for updates from Daniel - well, that changed yesterday. Since I am not properly trained nor have a degree in chemistry (besides high school lvl), for me to find solutions that might be beneficial are slim, but I did mention that I want to contribute as much as possible and I might just have something. I say might because this could be a dud for reasons beyond me.

I sat down and did a search for "diy Nafion" on YouTube and indeed, a video popped up [How to make alkaline membrane for fuel cell] and that seemed relatively straight forward and I was going to mention this video and now I have... LOL. I figured I would check the channel for other interesting stuff and indeed, there was one. [Battery using ion conducting ceramic separator and dual electrolyte]. If you scroll down and read in the comments section, he does mentions - translated from Portuguese - in quote since it is only supporting info:




Ceramic ion filter, interesting. So I did a search for ceramic water filter and many hits appeared. I expanded the search to "ceramic ion exchange membrane" and after scrolling a bit .. ding ding ding. Something of huge interest popped up and my gut feeling told me this could be it. This could be the membrane this cell needs in order to be even more cost effecting than using CMI-7000 and the overall cost of that membrane was to be honest, one of the reasons I wanted to explore some options. I have incorporated the interesting bits from the article:

Catalyzing Commercialization: Flexible Ceramic Membranes Enable Low-Cost Energy Storage - Feb 2019

Molecular separations are at the heart of chemical engineering, and nanostructured membranes are emerging as a low-cost alternative to traditional separation processes. However, most membranes struggle to maintain performance in harsh environments. For example, redox flow batteries (RFBs) rely on a cation exchange membrane to store energy. To create an RFB capable of storing enough energy to power a city (>1 MWh) for 20+ years without capacity loss, an extremely robust membrane chemistry would be required, such as a perfluoro sulfonic acid (PFSA) membrane (NAFION etc). However, PFSA is expensive to manufacture, has mechanical and performance issues, and generates substantial quantities of bio accumulative and toxic substances during production. In many cases, PFSA membranes account for more than 20% of the total battery system costs. Researchers have been searching for more than three decades to find a material capable of delivering better performance at a lower cost.
With funding from the National Science Foundation (NSF), researchers at Membrion, Inc., have patented a potential solution that uses the desiccant commonly found at the bottom of a beef jerky package. Membrion has developed a self-assembly process for silica gel that enables its use as a low-cost and high-performance ceramic cation-exchange membrane.

“We were trying to lower the costs of RFBs by finding a cheaper alternative to PFSA that could handle the extreme conditions. Ceramics seemed like a logical choice, but they tend to be brittle and not well suited for the high compression in stack applications like RFBs. Our real innovation was in making a flexible and compressible ceramic membrane,” says Membrion’s Founder and Chief Technology Officer, Greg Newbloom.

This novel membrane processing method starts with a nonwoven glass mesh commonly used as roofing insulation in the construction industry. Membrion scientists add a polymeric edging to the outer rim of the mesh to act as a gasket for the compressive loads in stack environments. They dip this polymer-edged glass mesh in a silica solution that wicks into the mesh, and then dip the mesh in a commodity acid. The end product is a percolated nanoporous ceramic membrane with a 2-in. bending radius.

“The acid dip is where the real science happens,” Newbloom says. “The acid causes the silica to electrostatically destabilize and form a network of nanopores. By controlling the kinetics of the process, we can make pores of different sizes and network structure. We use size and charge screening to dictate which molecules pass through the membrane and which are rejected.”

20191115_144338-scaled.jpg
1-s2.0-S0015188220302354-gr1.jpg

1-s2.0-S0015188220302354-gr5.jpg



Membrion, Inc have two patents which is linked to under sources.
1. WO2020247467A1 CERAMIC CATION EXCHANGE MATERIALS
2. WO2020247472A1 CERAMIC ANION EXCHANGE MATERIALS

Closing words.
If this membrane turns out to be a viable option, then it would be advantageous to incorporate a diy process just like Daniel was working on regarding the cellulosa+PVA. Now, I don't know how much our chemists know, so I dug up an article / book called: Current Trends and Future Developments on (Bio-) Membranes - Silica Membranes: Preparation, Modelling, Application, and Commercialization - Chapter 1 is literally called: Chapter 1 - Preparation of Silica Membranes by Sol-Gel Method - the book is linked down bellow. I will take a look what it has to offer.

I sincerely hope this membrane is the solution because if so, then this battery will be made using 3 of the most abundant elements: Silicon, Iron and Manganese - wouldn't that be awesome

... waiting for feedback

Sources:

Thanks for sharing!

First impressions after reviewing this literature. This would be a nice to buy if they made it available, but sadly I don't think it would lend itself easily to DIY, as-is, at least not for me. This requires the use of TEOS and MPTES, both not easy to get or "nice to handle" chemicals. Definitely hard to get in the EU. While the chemical that is generated in the end is very inert, nanostructured silica, the precursors are quite reactive silicon containing materials.

The press releases definitely down-play the complexity of the process and make the chemicals used seem much more common and innocuous than they really are imo.
 
Glad you like it :) I already emailed them regarding sample, so lets see how that turns out and ofc I will make sure you get some. I am going to look over the book I linked to just to see what I can pick up in terms of info.

Maybe there is a way to make the ceramic silicon with nicer chemicals...
 
Back
Top