Lab Notebook Entry #14

Made an argon sparging/blanketing setup with some equipment I had. Cleaned the setup (took all morning) and built a new cell, let it sparge for an hour, but after I stopped sparging and tried to keep an argon overpressure/blanket, I got leaks from the barbs. System isn’t designed/can’t take much over ambient pressure.

I really don’t like the push-fit barbs. I redesigned them with Luer-Lock style, have a hunch that may seal better.

Here’s how it looks like:

Double reservoirs with internal overflow, but with Luer Lock connector ports instead of print-in-place barbs

I did the same think for the flow frames. This makes building kits way easier (if it works). But they seem at least as good as the print-in-place barbs so far. I did have to change the slicer a bit to print the Luer Lock part with more precision than the rest of the reservoir, to make it thread properly.

Barbed and Luer Lock flow frames in FreeCAD

Daniel recommended just sparging the system (through the septa) then removing the argon tubing completely, and not trying to keep an active argon overpressure, to both mitigate leaks and conserve the little inert gas I’m able to afford/store in my lab. Sort of like bubbling argon through a Schlenk glassware then closing it off from the line. I will do this with a ZnI cell to see if the voltaic bulge appears (later) than without argon sparging.

Ideally, I could limit the argon pressure over the reservoirs and avoid a constant flow over them (which can evaporate liquid from the system), but that would require an expensive forward or backpressure regulator, which I don’t have.

Zinc-iodide test with argon

Note to self for next time—remove the septa argon inlet/outlet tubing more carefully, lost a couple drops of electrolyte.

I started a test with filename 16-04-2026-KPS-22.txt with the same conditions as last time but sparged with argon first for 15 minutes with the pumps on. Something I noticed right away was the charging curve went straight to the 1.3 V plateau, no funny side reactions at the start.

# electrolyte component masses, in g
MASS_ZnCl2  = 1.38
MASS_NH4Cl  = 1.06
MASS_KI     = 3.32
MASS_H2O    = 8.50
MASS_TriEG  = 0.55

total_mass_kg = (MASS_ZnCl2 + MASS_KI + MASS_H2O + MASS_TriEG)/1000.

TOTAL_VOLUME = 11  # electrolyte volume in mL, approx, measured by taking as much electrolyte as possible up into a 12 mL syringe

MASS_TO_RESERVOIRS = 14.55 # g of electrolyte actually loaded into system, based on weighing syringe before/after loading reservoirs
# molecular weights in g/mol

density = MASS_TO_RESERVOIRS/TOTAL_VOLUME

MW_ZnCl2  = 136.315
MW_NH4Cl  = 53.49
MW_KI     = 166.0028 
MW_H2O    = 18.01528
MW_TriEG  = 150.174

molality_ZnCl2 = MASS_ZnCl2/MW_ZnCl2/total_mass_kg
molality_NH4Cl = MASS_NH4Cl/MW_NH4Cl/total_mass_kg
molality_KI = MASS_ZnCl2/MW_KI/total_mass_kg
molality_TriEG = MASS_TriEG/MW_TriEG/total_mass_kg
molality_H2O = MASS_H2O/MW_H2O/total_mass_kg

molarity_ZnCl2 = MASS_ZnCl2/MW_ZnCl2/TOTAL_VOLUME*1000.
molarity_NH4Cl = MASS_NH4Cl/MW_NH4Cl/TOTAL_VOLUME*1000.
molarity_KI = MASS_ZnCl2/MW_KI/TOTAL_VOLUME*1000.
molarity_TriEG = MASS_TriEG/MW_TriEG/TOTAL_VOLUME*1000.
molarity_H2O = MASS_H2O/MW_H2O/TOTAL_VOLUME*1000.

if False:
    print("Electrolyte Composition:")
    print("Molarities (moles/L solution): {:.2f} M ZnCl~2~, {:.2f} M NH~4~Cl, {:.2f} M KI, {:.2f} M triethylene glycol, {:.2f} M H~2~O\n".format(molarity_ZnCl2, molarity_NH4Cl, molarity_KI, molarity_TriEG, molarity_H2O))
    print("Molalities (moles/kg solution): {:.2f} m ZnCl~2~, {:.2f} m NH~4~Cl, {:.2f} m KI, {:.2f} m triethylene glycol, {:.2f} m H~2~O\n".format(molality_ZnCl2, molality_NH4Cl, molality_KI, molality_TriEG, molality_H2O))
    print("Density approx. {:.1f} g/mL\n".format(density))

Electrolyte Composition (molarity, per L solution)	0.92 M ZnCl2, 1.80 M NH4Cl, 0.76 M KI, 0.33 M triethylene glycol, 42.89 M H2O
Electrolyte Composition (molality, per kg solution)	0.74 m ZnCl2, 1.44 m NH4Cl, 0.60 m KI, 0.27 m triethylene glycol, 34.31 m H2O
Electrolyte Volume	5.5 mL (anolyte) + 5.5 mL (catholyte)
Electrolyte Density	1.3 g/mL
RFB Dev Kit commit	e655c88f06
Cell Geometric Area	2 cm2
Separator	Daramic AA-900
Anode Configuration	Grafoil + 3.2 mm graphite felt
Cathode Configuration	Grafoil + 3.2 mm graphite felt
Gaskets	Outer and inner, 0.40 mm silicone (measured with micrometer)
Current Density	20 mA/cm2
Charging Conditions	To 100 mAh (about 9 Ah/L) or 1.7 V
Discharging Conditions	To 0 V
Flow Conditions	Kamoer KPK200 24 V brushless peristaltic pumps at 40% duty cycle (about 1500 rpm) with 3x5 mm Tygon Chemical (PTFE-lined BPT) tubing, double reservoirs with Luer Lock barbed fittings
Inert Gas	Sparged electrolyte through septa with argon for 15 minutes prior to cycling with pumps on

It’s running now, hopefully it can go for a while 🍀.

Large-cell testing

Starting to prep for this. Finally got the MightyWatt software talking to hardware after mucking around with Windows VM on my Linux PC (don’t have enough disk space to dual boot, lol).

I don’t miss Windows AT ALL. Why are there flowers in the start menu? Why are there stock prices in the notification area?

The MightyWatt is an OSHW electronic load that I will be adapting as a battery cycler, similar to this application note, likely with a relay to change automatically between charging and discharging. The MYSTAT doesn’t have the current range for this, and we don’t need a galvanostat/potentiostat three-electrode setup for this anyway. Gonna hook this up to a lead-acid battery I have here to test and see if it works. Also want to use OwnTech stuff eventually for more custom stuff but trying to keep it simple for now.

I bought the MightyWatt a while ago now off the creator on Tindie, when I mentioned I wanted to set it up as a cycler he mentioned this:

For the tester, there are some options. You can use MW to manage charging and discharging but you will need a way to switch power wires. A single DPDT relay should be sufficient. You can control the relay from the MW control program if you use one of the unused Arduino pins.

I have seen some 3V power supplies but the current is relatively low (2.25 A).
https://www.tme.eu/cz/details/zsi-53998/zasuvkove-napajeci-zdroje/goobay/53998/

You can also use a higher voltage power supply and then reduce the voltage using a step-down module.
https://www.tme.eu/cz/details/oky3504/menice/okystar/

All the voltage you drop on MightyWatt goes into heat so you will obviously be limited by the 75W capability. Having a step-down pre-regulator and leaving 0.5 to 1 V to drop on MW would probably be my first try.

You can also bypass MW for charging entirely and attach (positive terminal via a SPST relay if you want an automatic system) a power supply with a current-limited step-down module directly. Then switch between charging and discharging.