]]>
I'm not sure the batteries operation would actually involve metallic sodium. It depends on how the sodium is trapped in the anode. Is it kept in metallic form, or as positive ions in a negatively charged structure? If the latter, there shouldn't be a problem.
The impression I get from the KurzweilAI article is that the cellulose anodes store them as ions, so it should be okay.
It has been a couple of decades since I studied chemistry. But from what I remember, sodium has only two oxidation states, either zero or +1. Iron and nickel are transition metals and have variable oxidation states. Hence, sodium must be included as a metallic substance, as its oxidation state must change for it to produce energy. The same with lithium.
The obvious advantage that sodium and iron have over lithium are their abundance. They are both amongst the most common elements in the Earth's crust. Sea water contains sodium in effectively infinite quantities as far as humans are concerned.
]]>This novel electrolyte contained triethyl phosphate as the sole component of the solvent. This substance is known as a flame retardant. It has been tested in lithium-ion batteries, but only very high concentrations provided enough stability for long-term operation, too high for industrial applications.
The battery industry demands dilute electrolytes, which are cheaper and ensure better performances. They combined the phosphate solvent with a commonly available potassium salt and obtained an electrolyte that did not burn and allowed stable cycling of the assembled battery concentrations of 0.9 to 2 moles per liter, which are concentrations that are suitable for larger scales; for example, in smart-grid applications.
Based on some research it seems like you're right that Lithium does exist in the +1 state in LiC6. That would mean that the graphite layers between which the Li is intercalated have a charge of -1 per 6 carbon atoms.
However, I think it would be going too far to say that because the Li exists in the +1 state that this compound is unreactive. Looking at Wiki's table of standard electrode potentials, we see:
Li+ + e- -> Li (s)
Has a standard electrode potential of -3.04 V and:
Li+ + C6(s) + e- -> LiC6(s)
Has a standard electrode potential of -2.84 V. For comparison Na neutralization is -2.71 V.
I guess what I'm saying is that it might well react with water to form LiOH and H2. For example,
Li3N + 3 H2O -> 3 LiOH + NH3
Will happen spontaneously even though Lithium Nitride is otherwise stable.
I guess this does change my understand of lithium ion batteries though. It's actually powered by Carbon and Cobalt--Lithium is just an intermediary.
]]>To maintain the charge balance in the cathode, an equal number of some of the positively charged intercalated lithium ions are dissolved into the electrolyte solution. These travel over to the anode, where they are intercalated within the graphite. This intercalation reaction also deposits electrons into the graphite anode, to ‘tie’ up the lithium ion.
During discharge, the lithium ions are de-intercalated from the anode and travel back through the electrolyte to the cathode. This also releases the electrons that were tying them to the anode, and these flow through an external wire, providing the electric current that we used to do work
That doesn't sound like lithium being stored as metal. That sounds like a negatively charged lattice storing positively charged ions.
]]>Again, that doesn't seem to be how it works in the cellulose anode. It seems to be instead storing electrons as a negatively charged structure, with the charge balanced by the inclusion of sodium ions.
Electrochemistry has never been my strength, but I don't think that's what happens in a battery. I read the article n KurzweilAI and the very similar article on the UMD news site and I'm sorta unclear on various points.
The novel development here seems to be that they used wood coated with tin instead of a different substrate, and that this improved the lifetime of the battery by deforming rather than snapping as the coating changed shape.
I don't believe that batteries "store" electrons; If they did, that would make them capacitors. Batteries are fundamentally chemical things. Chemical reactions happen at the electrodes that absorb or release electrons. Fundamentally, in order to work, reduction or oxidation needs to be happening at each electrode. Specifically, reduction happens at the cathode and oxidation happens at the anode.
For either of those to happen, the oxidation state of something has to change. Otherwise, it's not a battery. Most batteries, it seems, use the difference in electron affinity of two different metals to create an emf between their two terminals.
The battery being proposed is exactly that: Iron has a higher electron affinity than Sodium. According to Wikipedia's table of Standard Electrode Potentials, the reaction:
Na+ + e- -> Na(s)
Has a standard electrode potential of -2.71 V, and:
Fe3+ + 3e- ->Fe(s)
Has a standard electrode potential of -0.04 V.
Putting the two together you'd expect a cell voltage of 2.67 V (minus inefficiencies). But the important thing is that these reactions are actually occurring at the electrodes, with the movement of negatively charged ions (such as Cl-) completing the circuit. (If you were to do Fe3+ + e- -> Fe2+ instead of going all the way to the metal, you could actually get a cell voltage of 3.48 V instead! In a real cell you would expect voltage to be 3.48V until the Fe3+ is used up, and then 2.27 V thereafter).
By the way, this is basically what happens in Lithium Ion batteries. At the anode LiC6 -> Li+ + C6 + e-, and at the cathode CoO2 + Li+ -> LiCoO2. The fact that Co doesn't go anywhere hides it, but the Cobalt is actually changing oxidation states from +4 to +3 and absorbing an electron.
There's a few ways to try to make this kind of cell work. The traditional way would be to find some nasty organic chemical in which Iron and Sodium compounds are both soluble and run with it. Another way (for stationary applications) could be a kind of gravity cell. You might have water and another liquid, totally insoluble in each other. The liquids both dissolve the relevant compounds of Fe and Na. Cathode (where Fe metal is produced) is in water and the anode (where Sodium oxidizes) is in some organic compound that dissolves salts but doesn't react with sodium.
It's a tough problem because Sodium is so reactive, but it should be possible. Lithium batteries use slightly-polar organics.
]]>As far as I can tell, the proposal succeeds or fails on the question of isolating metallic sodium from water. If it's stored as ions, no problem. If it's stored away from the water, also no problem.
]]>As I understand it, in lithium-ion batteries, the lithium starts in a neutral state when it's discharged. and is stored as lithium ions when the battery is charged. But either way, it gets stored as ions at some point.
How I'm thinking the battery would work is, electrons are put into the cellulose anode, which exchanges hydroxyl ions for sodium ions to balance the charge, resulting in a negatively charge cellulose mesh incorporating positive sodium ions to keep the charge neutral. The hydroxyl ions are then stored by reacting with the iron cathode to form iron hydroxide, releasing electrons which completes the circuit. This is reversed when the battery is discharged.
]]>The battery reaction (assuming we use chlorides) is as follows:
Fe+3 NaCl+energy <-> FeCl3 + 3 Na
The flow of electrons from Fe to Na is the current flowing through the wires attached to the battery.
]]>The impression I get from the KurzweilAI article is that the cellulose anodes store them as ions, so it should be okay.
]]>I suspect you'll end up with something more like an electrolysis cell, where reduction of the Iron anode reduces the required voltage for electrolysis somewhat.
In that case NaCl might actually be an even better electrolyte as FeCl2 and FeCl3 are soluble in water while Fe(OH)2 and Fe(OH)3 are less so. But I digress.
This would work as a molten salt battery, presumably, but that's a bit harder than Iron, paper, and lye. It looks like Lithium Ion Batteries use LiXFy compounds (For example LiPF4) in various organics (Wikipedia suggests ethylene carbonate, dimethyl carbonate, and diethyl carbonate).
Also, I learned today that in a lithium ion battery Lithium is being reduced, and Cobalt is being oxidized from the +4 to +3 state. Interesting stuff!
It looks like you might be able to get away with Methanol, actually. NaCl is somewhat soluble and FeCl3 is very soluble. Wiki doesn't say anything about FeCl2, though, so you may want to check. I don't remember offhand how to check that it won't produce chlorine gas, but you'll definitely want to look into that as well.
I take it back--Sodium reacts violently with alcohols. My bad!
The trick I suppose is finding a solvent that won't react with Sodium but will dissolve a compound of Sodium and Iron. Formamide might do the trick but information is hard to find and that's not a nice chemical to deal with because it has the unfortunate habit of sometimes decomposing to HCN gas.
]]>I really ought to dig out my chemistry kit in that case. It shouldn't be difficult to test.
]]>