Introduction: Create Large Refuelable Metal-air Battery.

Harness the power of rust! Metal-Air batteries are a way to capture the power of metal oxidation in a salt solution. In this instructable, I'll be building an iron-air battery, turning iron into rust using salt and atmospheric oxygen.

For a quick background, take a look at the energy that can be released from various metals:

While aluminum would release more energy, My goal here is to create a power source that is cheap to refuel (using rebar or scrap as the fuel source), and could be made from easliy obtainable materials in a third world environment. additionally, iron oxide is easily smelted back into raw iron using charcoal, and is non toxic - unlike aluminum, which is nasty stuff.

A metal-air cell is just: ANODE of metal (in this case iron), CATHODE of carbon/air, and a salt electrolyte. So far, my experiments have yielded Epsom Salts are a pretty darn good electrolyte.

What DOESN'T WORK? Seems like urea-based sidewalk salt is a huge failure, avoid everything containing urea.

COVER PICTURE: This is a carbon-block water filter. It was a little difficult to build the cell, as there is no inside liner. The newer filters (in the rest of the instructable, you will notice a white liner).

UPDATED: 3/10/14 - Added tables and additional data on most pages. Added diagram of how it works to main page.

UPDATED: 3/11/14 - added positive/negative to diagram and added Polarity step. Updated Filter page to indicate preferred filter (block) and style.

UPDATED: 3/12/14 - Polyethylene Glycol (PEG)- otherwise known as Miralax (the laxative).. immediate boost in power under load. 10Ohm resistor 0.5v fell under 0.01.. after PEG, jumped to 0.113 and was climbing when I left the basement... will let it run a few hours to see what happens.

UPDATED: 3/14/14 - Updated results with current output from a two cell in the final step, a few other edits in the mix as well.

UPDATED: 3/17/14 - 2 cells running with PEG since 3/12, 0.88v, 10Ohm resistor since 2/12 = 10.560 Ah * 0.88v = 9.2928 Watts expended. I think it's time to scale this up to 12v.

UPDATED: 3/17/14 - removed all iron from the outside of the cells, voltage went from ~0.4 to 0.55 immediately. Added two more cells to scale up to an initial voltage of 2.0v .. once it stabilizes I'll add the 10 Ohm resistor for a 200mA load .. and we'll see how it goes.

UPDATED: 3/18/14 - I attempted a recharged on the 4-cell at 3v for about an hour, which made the cells hold a nice 2.34 volts .. the two fresh cells seem to have a higher internal resistance, and don't show the surface crystals the older ones do. I think building up this lattice of epsom, salt, and PEG has something to do with efficiency... keeps the unit wet (from atmospheric water) without dripping.

UPDATED: 5/6/2014 - realized Maalox is incorrect, the product is Miralax, composed of only Polyethylene glycol.

Step 1: Water Filters

Here's my latest experimental filters.

They're cheap, but the carbon isn't a block like my current best cell. The carbon seems to be a roll of paper. I'll update on the progress with this filter, it *might* work - we'll see.

The idea is to create a damp, spongy CATHODE, so that oxygen can meet the electrolyte and iron ions. But you don't want it TOO loose, or you won't keep the elctrolyte damp. Additionally, having some kind of lining inside will prevent accidental short circuits, iwhen you fill the chamber with metal or activated carbon.

The second image shows how I cut off the white plastic netting, so I could wrap my CATHODE electrode around the body of the cell.

The bottom hole has to be plugged with a rubber stopper. You may have to trim the filter's hole to get the stopper to fit. This will prevent your electrolyte form just pouring through.

Finally, once prepared I filled the elctrodes with distilled water to ensure everything was moistened and ready to roll. Finally, I just let them sit in the water to soak up what they could.

Carbon-block water filters: :) - will make you happy.

You're more likely to be successful with these. The carbon is brittle, but not difficult to handle. Becuase the carbon is so thick, you're more likely to have a dry connection for your wire, and less likely to have the "reverse current" problem of the carbon paper. I was able to find $8/pair carbon block filters in my local hardware store's water filtration section.

"Carbon Paper" filters (like the one's in the picture): :( - will make you sad.

Seem to work well enough, but I think they might not be as conductive. Additionally, because they stay completely wet, you are at risk of having your metal electrodes form a secondary air-battery, sapping all your power. I'm experimenting with making carbon-paste electrodes (glue and activated charcoal) or some other similar method to extract power without the part you clip to being wet.

Step 2: Rebar

I bought some 24-inch rebar at Lowe's as my fuel source. Avoid zinc or other coatings that might prevent oxidation. Oxidation is exactly what we want... The iron ions will be matched up to oxygen ion from the air to form rust, and give us a few free electrons. 1 foot of rebar should give is a few hundred watts of energy.

Why not Aluminum? Why Iron?

Aluminum would give you a FAR more powerful battery (on the order of 11kw/kg), but I'm not a huge fan of the Aluminum Oxide (AlO) that would result. One reason, is that Aluminum is kind of nasty stuff, biologically speaking, and another is the AlO is almost impossible to recycle at home.

Magnesium, Nickel, and other metals have similar problems.

Iron, and it's by product - Rust, is pretty biologically benign. Also, given enough rust (and the carbon filters) you could re-smelt it into a bar. So, lower power, yes. Better for your health since you're handling it? yes.

Step 3: Electrolyte

This picture shows me using a urea "pet safe" electrolyte. I'm not really sure it worked well, since I didn't let it sit very long, but I definitely can say Epsom salts (Magnesium sulfate) do an excellent job. I suspect milk of magnesia might also help (Magnesium hydroxide).. but haven't done any longer term studies.

I will try to make this more structured (with actual voltages and times), once I have enough cells to analyze.

Best Combo, so far:
Sea Salt + Epsom Salts + PEG (Polyethylene Glycol)

Good Electrolytes:
Sea Salt and water .. NaCL (sodium chloride) and some trace minerals
Sea Salt and vinegar .. becomes Sodium Acetate - seems to work well.
Epsom Salt ............... Magnesium Sulfate

Possibly good, requires some more experimentation:
Cream of Tartar ........ Potassium BiTartrate

"Adjuncts": (Other junk I poured in to see if it helped)
Milk of Magnesia....... Magnesium Hydroxide ... not sure if it helped.
Nickel Hydroxide ...... left over from an edison cell project, again - not sure it made a difference
Nickel Chloride ........ seems to have helped somewhat.
Titanium Dioxide ...... didn't seem to affect it either way.
Polyethylene Glycol (PEG) ..... (Over the counter as Miralax, the laxative) - WOW!!!! Immediate boost in power under load. 10Ohm resistor 0.5v fell under 0.01.. after PEG, jumped to 0.113 and was climbing when I left the basement... will let it run a few hours to see what happens.

Calcium Chloride (Damp Rid) - effective, seems a good trade-off for epsom salts. Both draw in moisture the keep the battery operating without needing to be topped up with water.

Definitely Does not work:
Urea-based sidewalk salt - seems to just "clog" the system, and require a LOT of flushing to remove.
If you see Urea in the ingredients, it will kill the cell.

Need to try, or couldn't find "just by itself" yet:
Potassium Chloride
Lactic Acid - Iron reacts quite a bit in a lactic acid solution. This is also easy to make (with simple veggie fermentation), need to try.
Anthocyanidins - purple color from various berries, used to boost power in dye sensitized solar cells, might help here as well.

Step 4: Keeping It Wet.

The electrolyte needs to be wet... because the filter is naturally "leaky" - water moves through it - you'd have to install a pump to keep circulating electrolyte through the system. Short of putting cloth or something similar in there, I decided to put activated carbon, which also helps conduct electricity to the metal anode.

Activated carbon is available at your local pet store in the aquarium section. I filled the tubes as best I could, having to wiggle the iron to get the carbon to fall into place.

The carbon will hold the water and salts nicely, and act as a medium for the ions to move back and forth. Additionally, you can recycle it if you need to "refuel" in the future.

Polyethylene Glycol (PEG, also known as Miralax laxative) is a great way to keep the electrolyte from moving, and also seems to contribute a lot to the process. Added PEG not only made the battery work better, it quit leaking.

To prepare the PEG+Epsom+Salt Filter Cartridge: Make a strong solution (1 Cup epsom, 1 cup salt, 1/2 c or so of PEG) and soak the filters.. they seem to work best once crystals start forming on the outside.

Step 5: Wiring.

Just an old spool of bailing wire I had in the basement. giving it a little scrub with sandpaper seems to make it a pretty good conductor.

Unfortunately this CANNOT touch the wet surface of the carbon electrode. It forms a miniature air-metal cell, and seriously reduces the efficiency of your cell... running around 0.44v instead of 0.5-0.8v - half your power wasted!

Step 6: Additional Electrode Materials

Dawn sells a great steel pot scrubber that I've used for several projects. Be careful, it can cut you pretty good if you pull it... but with a pair of pliers you can make a "steel sheet" of wool

In this electrode, I'm going to try using it for additional surface are and see if it matters - or if my "simple spiral" of iron wire is all I need.


Yes, you can wrap the rebar in steel wool, and it will work. This might give you a big boost of power, but corrode and fail more quickly.

DO NOT USE AS AN EXTERNAL ELECTRODE - it will vastly lower the efficiency of your cell. use graphite or carbon rod.

Step 7: Assembled Cells.

Here I've joined two cells in series to see how well it works.

My inital voltage was only 0.3v, which was much less that what I expected - so I'm going to let it sit over nigth and see what happens.

Theoretical power:


From the first, Rust-type interactions seem to run around -0.89v
From the second, our rust runs around 0.6-0.7v per cell.

My original cells seem to run about 0.75-0.80v per cell. I'm trying to get this up,

Galvanic "half cell" calculations:
Fe2+ + 2 e− is in equilibrium with Fe(s) −0.44
O2(g) + 2 H2O + 4 e− is in equilibrium with 4 OH−(aq) +0.40

Carbon itself as an electrode is a +.25 - but I don't know if that contributes to the cell or not.

Step 8: Polarity

The fuel rod should be negative (-), and the outer carbon should be positive (+)

If the Iron rod in the center is positive (if you have to hook your red lead to the rod to get a + reading on your voltmeter), then your outer electrode is actually forming a small air battery, and you aren't getting any of the power from the central rod. This happens when your wire and the rod are of different qualities of steel, or different metals... you basically just made an air battery on the surface of the main battery.

How I got around this? I had some carbon rod laying around, and used that as my electrode - suddenly my voltage jumped into a useful range (>0.5v per cell). I will experiment with some other carbon electrode ideas and post them here.

Some experimental ideas I have: Making carbon electrodes my epoxy/gluing/mixing activated carbon and a binding glue (I've seen this work with epoxy).. finding carbon cloth online.

Step 9: Scaling Up.

Ok.. scaling up:

The image you see is:

1. Carbon Cloth
2. all electrodes in contact are either graphite ("carbon") rod or carbon filters (Brita makes them for sports bottles)
3. No iron in contact with wet surfaces <--- important. only graphite or activated carbon electrodes touching the filters.
4. PEG (Poly ethylene Glycol) + Sea Salt (no iodine) + Epsom (Magnesium Sulfate) + distilled water.

2.0v open circuit! Seems like it takes a day or so to build up enough strength to handle an actual load (the internal resistance drops as the bars degrade)... so, I'll add another picture if it improves.

It might even improve over that, as I was occasionally able to get the voltage up a bit higher. I think my realistic max is still around 0.8 per cell (3.2v for these 4 cells)

Step 10: Results and Next Steps.

Lessons learned:

1. use carbon block
2. if you use carbon paper filters, you need some kind of carbon/graphite electrode to collect current, not spool wire.
3. add PEG (Polyethylene Glycol / Miralax) and salt - the PEG doesn't seem to leak through - so you don't have to keep adding electrolyte or pump it around or whatever.
4. an inner liner and activated carbon + salt + PEG make a good matrix for the internal electrolyte. This might make the unlined carbon block filters a little tricky if I have to get a liner in there myself.

I haven' tried PEG in the carbon block filters (will try this weekend)

Carbon paper + carbon electrodes (still need to remove the old wire): 2 filters 0.8v (0.4v/cell) 10Ohm resistor, internal resistance is negligable (only dropped 0.03v or so).. running 24 hours now .. thats ((0.8v / 10)*24) = 1920mAh so far... so I'm well over a crappy standard battery in terms of output.

I believe the same fuel+matrix in the carbon blocks will be closer to what it should be well over 0.8v per cell (1.6 for two) instead of 0.4v per cell... as it started out in the 0.750v range without PEG.

Where I plan to go from here:

1. cut the rebar shorter, and have a plug on top as well. use bigger rebar for longer battery life.
2. try different electrolyte adjuncts (lactic acid, possibly anthocyanidins) and update this instructable with a table of results.
3. try my hand at making graphite electrodes.