Zinc Bromine Batteries: Proposing the next scaling level, a petri dish battery

My experiments with Zn-Br batteries in Swagelok cells have now matured to the point where a relatively stable battery has been achieved with no large dendrite problems and acceptable Coulombic and Energy efficiency values (~85% and ~65% respectively). I believe these results would already allow some of you – who want to explore larger scales – to construct some prototypes with my exact configuration, but at a larger scale. In this post I talk about a larger scale setup constructed using a petri dish.

The building material is a difficult choice, since Zn-Br batteries form pretty corrosive chemical species, in particular elemental Bromine and perbromide salts. In order to ensure stability, the cheapest and most easily available material that resists Bromine is glass. The cheapest casing to build a relatively small scale, yet modular and practical battery that fits the geometry of my design would be a petri dish, you can get a 100mm diameter petri dish here, which you could use to put together the battery I’ll be describing.

Suggested geometry for the petri dish cell.

Knowing the area and height of the cell (100mm x 18mm) we can calculate the energy density and charge capacity you should get. Given my previous research, this should be around 2Ah of stored charge with an energy stored of around 2.85Wh. To obtain this ratios you would charge the cell at a constant current of 1A to 2Ah, which matches a current density of around 11.8mA/cm2 in my Swagelok cells, charging to 15mAh at 15mA.

For the anode, you will want to get 0.2mm Zn metal, which you can obtain here. For the cathode you will want to get GFE-1 graphitic felt, which you can buy here. Note that this is NOT a regular carbon felt, it is a highly activated and conductive felt. For the separator a non-woven Fiber glass tissue will do the trick, like this one. The GFE-1 cathode will need to be soaked in a 10% solution of Trimethylphenylammonium bromide (TMPhABr, CAS 16056-11-4) and then air dried. I bought my TMPhABr from alibaba. Using Tetrabutylammonium bromide (TBABr) gives significantly worse results because of the much higher molar mass and lower solubility of the salt in Zinc Bromide solutions. As a current collector for the cathode and to connect it to the outside I suggest you use graphite foil, like this one. For the Zn anode make sure you leave a strip of anode when you cut the piece so that you can make the connection to the outside, cover all the metal that will go through the cell in some non-reactive insultation, electrical tape should work well enough for your first tests so that it doesn’t short with the cathode material.

Geometry of suggested components for petri dish battery

For the electrolyte you can either use Zinc Bromide at 2.7M or a solution of this concentration prepared using Sodium Bromide and Zinc Sulfate Monohydrate. I suggest you browse some of my previous posts on solution preparation if you’re interested in using these sometimes cheaper and more widely available chemicals. However, in both cases you will need to prepare the solution with 3% hydrogen peroxide instead of distilled water, to ensure all Fe is precipitated and removed from the solution before putting it into the battery (Fe impurities cause huge problems in the cells). After preparing the electrolyte, leaving it 24 hours to react and then filtering, drop some strips of Zinc foil in it so that all the peroxide reacts, then use peroxide measuring strips before loading it into the cell to ensure all hydrogen peroxide has reacted. After a couple of days all peroxide should be gone. Also, make sure you also add 1% Tween 20 to the solution, this is a key component to improve wetting and prevent dendrite formation.

To put the cell together, place the Zn foil at the bottom of the petri dish, then cut as many layer of fiberglass separator so that once you put the GFE-1 cathode on top and close the lid you’re able to lightly compress the cell. Then the easiest thing is to take out the separators and cathode, soak them in electrolyte and put them back in the cell. Now you should be able to close the lid, lightly compress it, clean any spillover and seal the petri dish using a couple of rounds of electrical tape.

I am also very aware that Bromine is denser and will want to sink into the cell, most of it won’t though as it will be sequestered as perbromide within the carbon electrode. There are very good reasons why this cell has the cathode at the top and why a normal geometry just does not work well for long term cycling at these energy densities. Please read this post to better understand why a practical static Zn-Br battery actually requires an inverted geometry to work well.

Zinc Bromine Batteries: Iron impurities in Zinc Bromide solutions derived from Zinc Sulfate

An important issue with Zinc-Bromine batteries is the need for a high purity Zinc Bromide solution in order for the battery to work properly. The use of low-cost ZnBr2 sources, in particular the use of Zinc Sulfate and Sodium Bromide to generate Zinc Bromide solutions, can be problematic due to the presence of large amounts of Iron impurities. In this post I will discuss what happens when you have these impurities in a battery and how you can purify a Zinc Bromide solution to remove iron and achieve better results.

Layers of a Zn-Br battery using a 4.2M ZnBr2 electrolyte derived from the reaction of commercial grade Zinc Sulfate monohydrate and Sodium Bromide. The battery was taken apart after 15 charge/discharge cycles at an energy density of around 20 Wh/L.

The image above shows you the layers of a Zn-Br battery that was taken apart after 15 cycles, charging the cell to 15mAh at 15mA and discharging to 0.5V, the cell has a diameter of half an inch. As you can see, the fiberglass separator layers look very yellow and, although it would be tempting to assign this color to the presence of elemental Bromine, which is true in part, cells constructed with high purity ZnBr2 solutions do not show such a strong yellow coloring. The yellow color is in large part caused by the oxidation of Fe+2 to Fe+3.

Moreover, the Zinc anode is usually a grey or black color in batteries created with high purity ZnBr2 solutions, while in this case, it looks a lot like a rusted piece of Iron. This is no coincidence, as Fe is actually reduced in the anode and then oxidized to insoluble Fe oxide/hydroxide species. The first two layers of the battery show the strong presence of these red Fe+3 oxides, which are non-conductive and significantly hamper the performance of the battery.

Battery performance characteristics for the battery that was dismantled and showed above.

The characteristics of this inverted battery were actually not bad to begin with, achieving max CE and EE values of 81% and 61% respectively. However the discharge curves start showing significant deterioration around the 10th cycle, with strong drops in the discharge voltage around 3-4 mAh into the discharge cycles. This was also shown as a strong decay in the average discharge potential, which during the 15th cycle was around 6% lower than the starting value. This speed of decay is dramatically faster than for Zn-Br batteries created using high purity ZnBr2 solutions.

The above implies that you need to get rid of these Fe impurities if you expect to be able to run a Zn-Br battery that lasts for a long amount of time. Thankfully there is a pretty easy way to do this, which we know from the literature surrounding the purification of Zinc brines (see here). The trick is to use a 3% hydrogen peroxide solution for the preparation of your ZnBr2 solution, instead of distilled water, when you dissolve the Zinc Sulfate and Sodium Bromide salts. This causes all the Fe to be oxidized to Fe+3 and to precipitate out of solution.

A reaction containing 62g of NaBr, 52g of ZnSO4.H2O and around 50mL of a 3% hydrogen peroxide solution. You might want to add around 50-100mL of distilled water before filtering to make your processing easier.

The image above shows you how this reaction looks. The Fe compound generated is very red, while the precipitated sodium sulfate is white, giving you this sort of look. The reaction takes around 6 hours to fully complete – for all the Fe to be precipitated out – time after which you can filter the solution and measure the density of your resulting brine to figure out what approximate concentration you have (per my previous post). It’s crazy how much Iron the “high purity” agricultural grade Zinc Sulfate Monohydrate contains!

It is also worth noting that this ZnBr2 solution will contain a significant amount of peroxide so you either need to heat the solution up to fully decompose the peroxide or wait till all the peroxide is decomposed at room temperature before actually using the solution in a battery. To be absolutely sure, you can also use H2O2 testing scripts (like these), to figure out whether you have any peroxide left before considering a solution to be battery-ready.

With that said, this process is absolutely necessary to build Zn-Br batteries if you’re deriving your Zinc Bromide from an impure source, as a Zn-Br battery containing large amounts of Fe will degrade and stop working pretty fast as a function of time. If you want to test your ZnBr2 solution you can add a couple of drops of H2O2 to a sample of it and see if any Fe precipitate forms.

Zinc Bromine Batteries: Is there anything that can work just like propionitrile?

Trying to improve on Zinc-Bromine batteries, it seems that the use of non-aqueous solvents in order to better sequester the bromine would be ideal since successfully sequestering the bromine in a static battery configuration is hard, even when ammonium salts are used as sequestering agents. This is because the salts are largely insoluble in ZnBr2 solutions, so the sequestering agent needs to be included as a solid, which can only be done in very limited amounts without strongly degrading battery performance. This is what I have tried to do with TMPhABr saturated GFE-1 cathodes, but the amount of Bromine I can sequester before it leaks out into the rest of the cell is quite limited. This is evident when I take apart fully charged cells and the entire cell is tinted with a strong orange color.

Ideally we would want to have a phase that is separate from the aqueous ZnBr2 phase but that has a higher affinity for bromine, is able to dissolve a significant amount of ZnBr2 (>1.5M) and is non-toxic. Propionitrile has been shown to be a decent candidate to achieve this, see here, but it is not an ideal candidate for a DIY solution because it is hard to get in most cases and can be a significant hazard for human health. It is catalogued as a dangerous chemical for transport in the US and EU and its use therefore poses a lot of additional restrictions and problems.

Most polar solvents that can dissolve ZnBr2 to some extent and are electrochemically stable – meaning they won’t get oxidized at the cathode – are sadly also miscible with water. This means that they don’t form a separate phase, which is an absolute requirement. I tried using propylene carbonate mixed with tetrabutylammonium bromide (TBABr) – which does not mix with ZnBr2 solutions – but this was a complete failure due to the fact that the bromine and perbromides formed actually prefer the ZnBr2 phase to this organic phase, even at lower concentrations of TBABr.

2-Methyltetrahydrofuran BioRenewable, anhydrous, >= 99 %, Inhibitor-free |  96-47-9 | Sigma-Aldrich
Me-THF, a new organic solvent candidate I will be trying in Zn-Br batteries to serve as a cathode side electrolyte and sequestering media.

I then continued my search for the “perfect solvent” for this, reviewing a lot of the literature on non-aqueous solvents in batteries and on “green chemistry” to find the solvent with the perfect polarity, ability to dissolve Zn-Br, affinity for Bromine, immiscibility with water and low toxicity. I think I might have found a solution in the form of 2-methyltetrahydrofuran or Me-THF. This solvent is a highly polar ether, it can dissolve high amounts of ZnBr2, it has a high affinity for elemental Br, it’s significantly less toxic than propionitrile and, alike propionitrile, it does not mix with water or ZnBr2 solutions. It is also electrochemically stable and is also safe enough that it’s available for purchase on ebay/amazon, so I decided to buy some of it to do some basic experiments and see if this solvent finally provides the answer we have been looking for,

Interestingly the density of Me-THF is quite low – around 80-90% that of water – so it will naturally rest on top of it. This means that it is naturally suited for an inverted cell configuration (cathode on top, anode on the bottom), which is what we want (see here). I couldn’t find what the partition coefficient with ZnBr2 solutions in water is, but the hope is that we should be able to get enough ZnBr2 in the ether to have a successful battery starting with a 3M solution of ZnBr2 in water. Batteries with Me-THF likely won’t require the use of any ammonium salts – as in the propionitrile paper mentioned above – since the affinity of the Br for Me-THF is high enough to keep it away from the aqueous phase.

Alike with PC, it is hard to know what to expect, so I would advice waiting for some of my results before going out to buy this solvent. Who knows how it will work in practice!

Zinc Bromine Batteries: Why propylene carbonate will not work as a cathode electrolyte

I have written several blog posts in the past about the potential use of propylene carbonate (PC) as a potential non-aqueous solvent in Zn-Br batteries. However, through my research I have now discovered that this solvent will not work in these devices as a cathode electrolyte, due to the way it interacts with the chemicals that are generated within the cell. In this post I will explain to you the experiments I did and why I reached this conclusion.

The idea with using PC was initially to completely replace the electrolyte within the cell. This was discarded right away due to ZnBr2 solubility issues and low conductivity issues of the constructed cells. The idea then evolved to using a tetrabutylammonium bromide (TBABr) saturated PC solution (PC-TBABr) as an organic layer in an inverted device, since this layer can rest on top of a Zn-Br solution and can even remain on top after mixing if the Zn-Br solution is concentrated enough (>4M).

Small cell I used to visually study what happened with a PC-TBABr containing device. The cell was not characterized as the geometry is not reproducible and the surface area is too big for my testing equipment.

The organic layer is completely immiscible with the highly concentrated ZnBr2 layer and it was my hope that the TBABr3 produced in the cathode would be substantially more soluble in the PC-TBABr compared to the aqueous phase. Given that the PC phase is more than 50% TBABr, it seemed very likely that the produced perbromide would have a significantly higher affinity for the PC-TBABr.

To confirm whether this was happening, I constructed an inverted cell using a glass vial in order to be able to see what was going on (which you can see above). I placed a zinc anode at the bottom, used fiberglass as a separator and placed a GFE-1 cathode saturated with PC-TBABr on top using some C4 carbon cloth as a current collector. Before placing the GFE-1 cathode I filled the cell with a 4.2M ZnBr2 electrolyte, which makes the PC-TBABr remain less dense even after prolonged mixing. Since my objective was not to measure efficiencies with this device, but just to observe the chemical processes, I did not strive for a reproducible geometry.

After charging for a while with 3 charged AA batteries at more than 3V – just to make the process go fast – I noticed a lot of TBABr3 forming and precipitating within the cell. Sadly, the perbromide seems to form on the cathode and then immediately migrate out and into the aqueous phase. To my absolute surprise, the TBABr3 – which is more of a liquid rather than a solid – has higher affinity for the aqueous phase, although it doesn’t solubilize within it but rather forms a suspension with it.

I then proceeded to take the electrolyte out of the cell and perform an extraction using some additional PC-TBABr and surprisingly, all the perbromide stays in the aqueous phase after mixing and just refuses to get into the PC-TBABr. For this reason, PC is not going to work as a cathode electrolyte within the device, as the perbromide just exits it and never returns. This is probably why my devices trying to do this never seemed to work for too long.

Zinc Bromine Batteries: Preparing a solution of known Zinc Bromide concentration using Sodium Bromide and Zinc Sulfate

I wrote a post a while ago about how to prepare solid Zinc Bromide from Zinc Sulfate and Sodium Bromide through the use of water and isopropyl alcohol. However, this method has substantial issues when it comes to its practical implementation, as taking ZnBr2 out of solution is an arduous process that can also be dangerous due to the often aggressive splashing of hot concentrated ZnBr2 solutions. Taking the Zinc Bromide to a solid also seems unnecessary given that in the end we want to end up with ZnBr2 solution for batteries. What if we could just mix Zinc Sulfate and Sodium Bromide and somehow end up with a solution known Zinc Bromide concentration after precipitating the Sodium Sulfate? I will tell you how to do just that in this post.

Result of the process after cooling in the freezer for a couple of hours. An abundant sodium sulfate precipitate is forming.

The solubility of Zinc Bromide is exponentially greater than that of Zinc Sulfate, Sodium Bromide and Sodium Sulfate so mixing a solution of Sodium Bromide and Zinc Sulfate generates a concentrated solution of Zinc Bromide and precipitates almost everything else. At 0C the solubility of ZnBr2 is still 311g/100mL while that of Na2SO4 drops to 4.76g/100mL and that of ZnSO4 drops to nearly 0. It is important to keep an excess of Zinc Sulfate in the reaction though as the solubility of NaBr is still quite high at 0C, reaching more than 79g/100mL.

In order to carry out this synthesis I have followed this process:

  1. In a 250mL clean beaker, weight 45g of Zinc Sulfate Monohydrate
  2. Add 51g of Sodium Bromide
  3. Add 100mL of distilled water (it’s important to use distilled water)
  4. Heat the mixture with stirring till boiling starts or everything dissolves.
  5. Let the mixture cool until it reaches room temperature
  6. Optionally you can add 10mL of isopropanol here, which greatly reduces the Sodium Sulfate contamination. (rubbing alcohol works just fine)
  7. Place in a freezer for 24 hours.
  8. Filter the solution to remove all Sodium Sulfate and unreacted Zinc Sulfate.
  9. If isopropanol was added, boil the solution until all the alcohol is removed.

What you are left with is a concentrated solution of ZnBr2 of unknown concentration. Since we do not know how much the volume of the solution changed due to the reaction and some volume of solution is always left wetting the remaining solid, we cannot accurately determine the molarity of the solution from the things we added and what we obtained. We need to perform some measurements to get an idea about how much ZnBr2 we have in solution.

Density of a Zinc Bromide solution as a function of its molar concentration.

However, since this is a mostly ZnBr2 containing solution – with likely less than 5% of the salt weight being from sodium sulfate contamination – we can estimate the amount of Zinc Bromide by measuring the density of the solution and looking at experimental results showing the density variations of pure Zinc Bromide solutions. Using the experimental data from this paper from 1994, I was able to create the above graph, which allows you estimate how concentrated your Zinc Bromide solution is. Note that you should input the density in the equation expressed in kg/m^3.

In order to measure the density of the solution, I used a 10mL pycnometer, which you can buy here for a low price. A pycnometer allows you to very accurately determine the density of a solution since its volume is exact. By weighting the empty pycnometer and the filled pycnometer and then dividing the difference of this weights by 10 (volume of the pycnometer), you can obtain the measurement in g/mL which you can multiply by 1000 to get the value in kg/m^3.

Once you know the approximate concentration of your solution you will know how much you would need to dilute the solution to arrive at your desired concentration. Solutions produced with the above method are bound to be in the 4-7M region, so you will probably need to dilute them to arrive at a concentration that is better suited for the ZnBr2 batteries. With this information you can now prepare ZnBr2 solutions for your batteries without the need to prepare pure solid Zinc Bromide, have to deal with aggressively splashing solutions or have to go through any further purification processes.

Since Zinc Sulfate Monohydrate and Sodium Bromide are both widely available almost anywhere for really low prices, this should allow a lot of people to experiment with these batteries with low costs, yet retain the ability to understand what the concentration of their electrolyte is.