Participation in the Flow4U Battery Conference

On April 9, the Flow Battery Research Collective attended the Flow4U flow battery conference at Eindhoven University at the invitation of Prof Sanli Faez of Utrecht University and the FAIR Battery project. We gave a presentation on our project which you can see below:

The presentation was uploaded to ECSarxiv and given a DOI here.

The presentation was immediately followed by a hands-on workshop using a beta version of our flow battery development kit, the manufacturing of which was sponsored by Sanli’s team at FAIR Battery. For the workshop we built 10 RFB cells and 5 complete kits in total, each including a MYSTAT potentiostat. The cost, including assembled MYSTAT PCBs from PCBway and machined plastic cell bodies from Xometry, was less than 500 EUR per kit. We demonstrated a real cell cycling with a zinc-iodide chemistry, safely contained within a sealed plastic box near the speaker podium.

Preparation

Prototyping before the workshop happened in an asynchronous, decentralized manner during the preceding months in the home labs of Kirk and Daniel and in the prototyping lab of Utrecht University, where Josh diligently fabricated many of the printed and laser-cut parts as well as many other items needed for a successful workshop. Kirk would make design changes in FreeCAD, Daniel would print them and refine his chemistry formulations, and then once we had working prototypes, Josh would make more of them and handle the logistics of a 10x scaleup.

In our prototyping process, though we had some limited success with FDM and resin-printed cells, for the workshop we decided to use a prototyping service to machine the cell bodies. We used Xometry EU and had a good experience. At a part quantity of 1, for CNC milling of polypropylene, the cost is 158 EUR, but at 20 cell bodies the price per part drops to 33 EUR (22 EUR if switching to polyethylene).

Plastic machined parts on a MDF work surface with tools and work gloves in the background. — 20 machined cell bodies, enough for 10 test cells, produced by milling of solid polypropylene, for the workshop.

There were also a lot of other parts needed for the workshop, including:

pumps
tubing
reservoirs
cell components: separator, flow frames, graphite felts, current collectors, nuts and bolts
and more…

The whole bill of materials is currently partially documented here, in the rendered documentation from our repository.

Series of toolboxes and storage containers filled with toolboxes on a laboratory work bench — Materials gathered the day before the workshop to assemble kits

Of course, we spent the whole previous day of the workshop making sure we indeed had everything and that it all worked and fit together. This was the first time the whole team had met in-person!

Two people at a laboratory workbench, one tightening a nut and bolt assembly with a torque wrench. — Daniel and Kirk testing the cell the day before the workshop

We of course forgot one type of USB adapter cable in the BOM and sorted this out, and prepared the cell to be cycled with real zinc-iodide electrolyte the following day during the workshop.

Flow battery test cell in jig on lab bench, with multimeter, power supply, and laptop nearby — Assembled cell in jig the day before the workshop, to be used for live demonstration of Zn-I chemistry

Workshop

Presentation

For our first-ever public presentation of the project, we felt well-received. At least two members of the audience were co-authors on papers we cited as examples of other academic cell designs.

A conference room full of seated attendees with one person at the front addressing the room — Workshop participants gathered for the initial presentation

A man with a microphone in front of a projector screen addressing an audience while gesturing — Prof Sanli Faez of Utrecht University introducing the workshop

A man addressing an audience from a podium while holding a microphone — Kirk addressing the workshop

Hands-on assembly

Right after the presentation we got into assembling the flow cells. People then brought the assembled cell to a station where we would pump pure water through the cell and check for leaks.

The components of the flow battery test cell individually laid out on a table, arranged in order of assembly — The components of each kit laid out

We got a lot of useful feedback during this workshop, both directly and indirectly. A major one being that most people don’t know how to use a torque wrench! I will add that the only team that built a leak-proof cell on the first try was the team of mechanical engineers… who how to use a torque wrench. We will add this to the documentation and workshop instructions in the future now.

Three people seated at a table assembling a test cell — Workshop participants assembling the cell

Three people seated at a table assembling a flow cell kit — Workshop participants placing a flow frame with graphite felt into the cell assembly

A person tightening a nut on the flow cell with a wrench — Tightening the flow cell with a wrench (torque wrench used for final tightening)

A flow cell with tubing connected sitting in a jig with pumps and reservoirs — The flow cells were put in to a jig to test for leaks by pumping water through them. Eventually we switched to just pumping manually with a syringe, since this was easier than disconnecting and reconnecting all the tubing each time!

Three people watching a leak test in the flow cell jig on a table in a conference room, all looking closely at the setup — Daniel and workshop participants watching a cell being tested for leaks in the jig

A group of people watching an individual pump water with a syringe through an assembled flow cell — The much simpler method of injecting water with a syringe

Feedback

We got feedback simply by watching the participants but also asked for their direct written feedback on the day. Here is a summary of what we received:

Good things

“easy to assemble”
“easy to align”
“sturdy”
“CHEAP ! ! !”
“fun activity”
“nice/understandable demonstration” (x2)

Things to improve

“fittings hard to fit” (presumably barbed fittings)
graphite foil is fragile/flimsy (x2)
need to include o-rings on drawing
indicate in instructions that some parts are not symmetrical
square copper plate with can be placed two ways, one which results in a working cell, the other way which results in a leaky cell
could add alignment options for quicker assembly
numbering of gaskets
flow field design can be improved / is very simple currently
overtightening of barbs and/or cell bolts
add washers for better pressure distribution
a symmetrical flow frame could avoid assembly in the wrong orientation
threaded hole for barb should be further from bolt hole
barbs are vulnerable to overtightening (at least one group overtightened and sheared them off completely)
closed and tightened cell appears slighly bigger in middle
(silicone) rubber gasket can loosen over time, more rigid plastic gasket might avoid this
possible to close cell with a clamp instead of bolts?

Roadmap for the test cell

We will complete the entire documentation for the version of the cell we produced for the workshop, and mark this version as the “first release” of the cell. We’ll then incorporate any breaking design changes from the feedback just mentioned which will go into the second release.

One design change that could address multiple feedback points is adopting some poka-yoke design strategies, such as making both the copper current collector and corresponding recess in the cell body rectangular, so that it can only be inserted one way.

Another suggestion was to address the ordering of gaskets and flow frames by placing a notch or other visual indicator along the external perimeter of each part, so that when assembled there should be a large visual indicator showing that the notches of all parts are aligned and therefore assembled in the correct order.

Reflection

Group of people posing for photo in a conference room — The team behind the conference - thanks to all, and in our case especially to Josh Hauser on the right for his efforts to set up a successful workshop!

Overall we are quite happy with how are first workshop went. We are very grateful for the support of Sanli Faez and his FAIR Battery project, and by extension Utrecht and Eindhoven Universities and the student team at https://team-energy.nl/ who all performed an excellent job organizing the entire conference.

Two workshop participants were chosen at random to take home complete kits to their own labs to experiment with, with the stipulation that they share their findings and feedback.

Now, we have our hands full with completing the documentation for the first release, before making design changes with the valuable feedback received for version two.