Vanitec – 2nd Energy Storage Meeting (Part 2)


4 July 2017

Electrolyte Standards – Peter Fischer of the Fraunhofer ICT

The VRFB group of the Fraunhofer Institute for Chemical Technology (ICT) based near Karlsruhe have built themselves their own 20MWh VRFB to store energy from their 2MW wind turbine which you can see in a beautiful animation here.

Peter is the group leader in charge of VRFB development and they have established significant programs in testing and characterising VRFB electrolytes and components from all over the world that are intended to be used in VRFBs. Fraunhofer ICT are also participating in the BMN and IDC backed Market study for Vanadium Flow Batteries announced on the 5th of October last year.

Dr Peter Fischer in the Fraunhofer VRFB building alongside the electrolyte tanks

Peter started by summarising the three basic categories of all-Vanadium electrolyte that are currently used for VRFBs.

Classic‘ Sulphuric Acid only Electrolyte  PNNL Mixed Acids High Energy Density Electrolytes
 Total Vanadium Concentration 1.5 – 1.8 mol-1 1.9 – 2.5 mol-1  2.6 – 3.5 mol-1
 c [V2(SO4)3] 0.375 – 0.45 mol-1  0.475 – 0.625 mol-1  0.65 – 0.875 mol-1
 c [VO(SO4)] 0.75 – 0.9 mol-1  0.95 – 1.5 mol-1  1.3 – 1.75 mol-1
 c [H2SO4] 2 – 2.5 mol-1  2 – 2.5 mol-1  2.5 – 4 mol-1
 c [HCl] NA  4 – 5 mol-1  NA
 Comments  Originally developed by UNSW – and used by Fraunhofer ICT battery  Patented by PNNL and licensed to specific manufacturers.  Being worked upon by UNSW – research only at present

Peter went on to describe a number of the issues with the ‘classic‘ electrolyte mixture that they are using for the Fraunhofer ICT battery – the effects of impurities on electrochemical kinetics – the stability of the V(II) electrolyte (especially sensitive to copper impurities) and issues with the slow coagulation/precipitation in the electrolyte.

Precipitation in ‘Classic’ electrolyte at high temperatures, courtesy of Dr Peter Fischer, Fraunhofer, ICT

The coagulation/precipitation at high temperatures is caused by polymerisation of some of the Vanadium species in the electrolyte, and is what puts an upper operating temperature of 35-40 degrees on the classic electrolyte. Fortunately the PNNL mixed acids electrolyte user by UniEnergy Technologies manages to avoid this problem – allowing higher operating temperatures to be tolerated without precipitation, and the Hi-Energy Density electrolyte formulations are now including precipitation inhibitors for the same reason.

Peter concluded by confirming that the Fraunhofer group are now providing multiple characterisation techniques for electrolytes and provide many services for testing of VRFB components such as membranes. He was keen to accept sample from VRFB researchers worldwide and has already undertaken assessment of dozens of samples.

Can VRFB compete with the Lithium Juggernaut ? – Mikhail Nikomarov of Bushveld Energy and Thomas Hohne-Sparborth from Roskill Information Systems

Mikhail pointed out that flow batteries have a well defined technical position in the energy storage phase space – 1-10 hours storage period and 20KW-100’s MW power rating, that positions them somewhat differently to Li-ion storage technologies  – those are typically sub hour storage and less than 10MW power rating.

He then pointed out that in the last few quarters reported energy storage projects based in the US are now averaging multihour storage times – so all seems good until you look in detail at the project and realise that this does not reflect installed VRFB projects but Li-ion projects which are encroaching upon the natural phase space that VRFBs should be occupying.

He then moved on to look at the analysis that is published on the Levelized Cost of energy Storage (LCOS) that is being published by industry analysts such as Lazards  – and which is used to inform executives on appropriate technology selection. He pointed out that in many of the largest-size categories, that VRFBs should be expecting to dominate, at present Li-ion currently appears to be able to claim lower LCOS costs.

One might think that this indicates that the assessment methodology is somewhat flawed – for example the definition of peaker replacement assumes that batteries can only be charged once a day – this clearly under-appreciates VRFBs which can be charged and discharged many times a day without losing long term performance unlike Li-ion batteries. Longer term the assessment methodology might be made more reflective of reality – however this approach would miss the other reality, which is that Li-ion large scale energy storage projects have a significant marketing headstart on VRFB technologies.

Furthermore Lazard is of the opinion that Li-ion technologies will somehow benefit from a greater ‘Learning effect’ as their production ramps up and economies of scale plus experience come into play to reduce the LCOS costs for Li-ion. This idea is based upon experience with technologies such as integrated circuit and solar panel production and a few years of deployment of Li-ion batteries in large scale energy projects.

As an aside I am less convinced that such “learning effects” can be relied upon – often these come as the result of technical advances which cannot be predicted – Li-ion has already had many more years of research, development and optimisation compared with VRFBs so the low hanging fruit there will have already been taken. Also comparing the reductions in prices achieved with one technology (PV panels) with what might be achievable with a completely different technology (Li-ion) seems naive regardless of how scientific you appear to make such ‘analysis’ seem. Of course the reductions in unit price of any technology cannot continue forever – eventually the ‘Learning effect’ must stop, and no analyst can predict at what stage that will happen – for example the famous Moore’s law in semiconductor devices only works because transistors can be scaled to smaller and smaller dimensions with improving speed and performance – no such scaling to the small side can exist in Energy storage – in fact the converse is more likely to be true.

Mikhail also pointed out that Lithium constitutes only a small fraction (2-4%) of the cost of Li-ion batteries and so the issue of the Li-ion battery market being restricted by potential rising Lithium cost would not be an issue for that technology. On the flip side of course – is the fact that Vanadium currently contributes a large amount (30%+) to the upfront cost for a VRFB battery – thus if an electrolyte leasing model can be made to work then the initial upfront cost for VRFB customers could be drastically reduced.

In the Q&A session I asked Mikhail whether such an electrolyte leasing arrangement would also affect the LCOS costs that are reported by analysts such as Lazards. Mikhail answered quite positively YES – if the electrolyte leasing company can lease the electrolyte to the VRFB user at an effective interest rate that is lower than the rate that the VRFB user would otherwise have to pay to a bank to borrow the capital needed to buy the electrolyte outright – (following up on this – Lazards LCOS calculations certainly say that they take into account the cost of capital, but I cannot see in the report exactly which value or band of values are assumed.)

If you think about it then you will see that an electrolyte leasing company certainly should be able to offer a better rate than a bank :

  1. The leasing company understands the asset that is the security (the electrolyte) – Mr Banker is not going to have any idea how to value the electrolyte and will perhaps only lend against a small fraction of its real value.
  2. An electrolyte leasing company is not exposed to the risk of a single VRFB user – if one goes belly up you can always take back the electrolyte and lease it to someone else – an isolated bank-to-VRFB-user loan would attract a much higher cost on the capital loaned because of the potential trouble in dealing with a customer who defaults.
  3. The electrolyte company can be very big and can negotiate the very best rates on the capital that it would need to borrow to get the electrolyte in the first place – it might even have a strategic stake in the V or VRFB production.

The conclusion – as well as the electrolyte leasing model lowering the initial Capex barriers to entry for VRFB it may also reduce the LCOS figures reported by analysts for VRFB technologies. Far from the 30%+ Vanadium being a weakness, it may prove to be a strength that can be brought to bear rapidly in sales and marketing of VRFBs.

continue to part 3 …

This article only conveys the personal opinion of the author. Whilst every effort is made to ensure the content is accurate, we cannot guarantee the accuracy of the data shown. This article does not constitute professional, financial or investment advice and must not be used as a basis for making investment decisions.

Site content is not authorised by the FCA and you are not safeguarded by the Investor Protection measures of the Financial Services and Markets Act 2000. See our full disclaimer