Trilogy on electricity storage

Part II: How new generation technology makes big storage work

Most of my professional time, I spend on thinking about batteries. Every day, I’m breaking my brain on how batteries work and how I can make them better. And I love talking about that too. That’s why I decided to write this three-part series on LinkedIn. To share what I am doing and why. In this second post I reveal the technology behind the battery we developed, the ReFlex™. I have done my best to describe it in an accessible way, but you might find it a bit harsh here and there. If you are more interested in how the batteries are made and deployed, see part I.

Go with the flow, from electrical energy to chemical energy in flowing liquids, or vice versa

The ReFlex™ battery is a so-called redox flow battery (RFB). As a battery, RFB is an electrochemical energy conversion device. However, unlike Li-ion and many other batteries (referred as “solid-state” batteries) that store energy in solid electrodes, RFB stores electrical energy in liquid electrolytes.

As shown in Figure 1, an RFB cell consists of two electrodes made from carbon felt (or other materials) and typically two circulating electrolyte solutions (a positive/ cathode-side electrolyte or catholyte, and a negative/ anode-side electrolyte or anolyte).

Figure 1  Schematic illustration of the structure of a redox flow battery

 

A membrane, or separator, stands in between to separate the catholyte and the anolyte. The energy conversion from electricity energy to chemical energy (charge or store energy) and vice versa (discharge or release energy) occurs instantly within the electrodes when the liquid electrolytes flow through the cell. It is essentially a regenerative fuel cell.

Not one size fits all

A flow battery is a flexible battery. To build up a useful voltage, a series of cells are connected in an “engine” called stack. An RFB storage unit can have one or multiple stacks that are flowed with the same catholyte and anolyte. This feature brings in a big advantage. It allows for separate design of energy (kWh) and power (kW). The energy/power ratio (or charge/discharge duration) can be tuned, and thus accommodate all sorts of user requirements. So, the stack decides the power, and the electrolytes decide the energy. With the same stack, additional volume of electrolytes increases energy and makes the duration longer. We can design a flow battery in several ways; it can give a lot of power during a shorter time, or give less power, but during a very long time. And pretty much every option in between that. In contrast, in a “solid-state” battery the energy and power are corelated, and cannot be fine-tuned to fully meet customer needs.

Keeping it cool, safe

A flow battery is a safe battery. During operation, the flowing electrolytes carry away the heat generated from the electrode reactions and ohmic resistance, such that the electrolyte tanks act as a large heat sink. That prevents overheating of the battery stacks and the individual cells within them. The physical separation of cells/stacks and electrolytes also helps avoid building up of a large heat in the cells/stacks. The generally safe profile of the RFB is further enhanced by the fact that operation can be stopped at any time (including any emergency) by turning off the pumps. There is only a limited volume of electrolytes left in the stack cells, which self-discharge. Furthermore, RFBs generally do not involve uncontrolled, violent reactions that can lead to a fire.

Unlimited cycles

There is more. Another enabling feature of the flow battery is its long life. The elimination of repeated ion insertion and de-insertion in electrodes as occurs in Li-ion and other batteries, preserves the structural and mechanical integrity of the cells/stacks, enabling a long-cycle life of the battery. In a true RFB that utilizes liquid electrolytes on both positive and negative sides, its cycle life is generally independent of the battery’s stage of charge (SOC) and depth of discharge (DOD). This is not the case with Li-ion and other non-RFB batteries that store energy in their solid electrodes.

From RFB to VFB, not all flow batteries are the same

Not all flow batteries use the same chemicals. The RFB may be traced back to zinc-halide batteries that use “consumable” Zn/Zn2+ anodes. True RFBs were first developed in the early 1970’s by Dr. Thaller’s research group at NASA. They used iron-chrome (Fe-Cr) chemistries and carbon electrodes, and had both reactants and products dissolved in liquid electrolytes. To avoid the cross-contamination and improve electrochemical activity over the early Fe-Cr RFB, the all-vanadium RFB, or VFB, was first demonstrated by Professor Skyllas-Kazacos’ research group in the 1980’s at the University of New South Wales.

For non-engineers this Vanadium might sound as an exotic name, but you probably have several items made with vanadium metals (as additives) in your house. Just check the screw drivers or spanners in your toolbox.

From here on it is getting a bit more technical. As shown in Figure 2, the VFB utilizes a V2+/V3+ aqueous sulphate solution on the anode side and a V4+/V5+ aqueous sulphate solution on the cathode side.

A standard voltage of 1.25 V is generated by the VFB through the following reactions:

 

Cathode: VO2+ + 2H+ – e–  « VO2+ + H2O

Anode: V3+ + e–  « V2+

Cell: VO2+ + V3+ + 2H+ « VO2+ + H2O + V2+

 

The VFB demonstrates an excellent electrochemical reversibility and virtually an unlimited cycle life. Over 10,000 full cycles has been demonstrated in the field, without limits to SOC and DOD. The use of aqueous-base catholyte and anolyte further strengthens the safety record of VFBs. The aqueous electrolyte also offers a large heat sink due to the high heat capacity of water. And the electrolytes are non-flammable. There are also no violent reactions that can lead to thermal runaway. All of the aforementioned make the VFB inherent safe. The favourable general features of RFBs and in particular, the advantages with the VFB have attracted wide interests in using the VFB for utility and other large-scale stationary battery applications.

Figure 2  Scheme of all-vanadium redox flow battery, in a charge process.

Still need for a better, new generation

Unfortunately, the traditional VFB is limited in the stability of its basic electrolyte chemistries. Vanadium oxides, e.g. V2O5, tend to precipitate out at elevated temperatures from the electrolytes via irreversible reactions, resulting in capacity loss and degradation in the battery’s durability and reliability. As such, the operation of the traditional VFB is limited in a temperature range to 10~40°C (practically, 35oC), imposing the burden of heat management and energy loss due to heat management. Additionally, the limited stability of the traditional VFB electrolytes limits vanadium solubility in the electrolytes and as thus energy capacity or density of VFBs, a disadvantage compared with many other batteries. The limited solubility also burdens mass balancing that is critical to the health of VFB flow batteries. Furthermore, the carbon felt electrodes tend to degrade due to building up oxygen activity and subsequent oxidation or burning of the carbon felts.

Discovery of a new electrolyte chemistry and a new VFB

To improve performance, reliability and economics of VFBs, my team at the US Department of Energy’s (DOE) Pacific Northwest National Laboratory (PNNL) developed a new electrolyte chemistry, with the support of the US DOE Office of Electricity Delivery and Energy Reliability’s Grid Storage Program. The new electrolytes still use vanadium as its active element, but has a chloride base solution, instead of the traditional sulphate one.

Figure 3 Electrolyte chemistry: sulfate base vs. chloride base.

 

With new vanadium complexes formed in the solution, the new vanadium electrolytes have a much-improved stability over the traditional sulphate based chemistry, allowing a higher vanadium solubility. (See Figure 4.)  The VFB that uses the new electrolytes demonstrated stable operation under broader conditions with a much higher energy density, among other improvements. The good thing is it remains at least as reversible as the traditional sulphate VFB, via the following reactions:

  

Cathode: VO2Cl + 2H+ + e–  « VO2 + Cl + H2O

Anode: V2+ – e–  « V3+

Cell: VO2Cl + V2+ + 2H+ « VO2+ + V3+ + Cl + H2O

 

Optimization and start of commercialization

With all the promises of the new generation VFB at the lab, I left PNNL in 2012 and started UniEnergy Technologies (UET) in Seattle and Vanadis Power in Europe to commercialize the technology. After licensing the new generation VFB from the lab, UET started the journey to engineer the new generation VFB into products at scales. The years of endeavour have now proved the new VFB at scales and in the fields with the following advantages:

  • A practically doubled energy density due to a higher solubility and a higher utilization of vanadium, enabling the compact ReFlexTM design that is a few times smaller than the traditional VFB;
  • Along with the optimized stack and system design, the new generation VFB uses up to 30% less vanadium, thus saving cost and preserving vanadium resources;
  • Operation in temperatures up to >55oC, easing heat management, mitigating solid vanadium oxide precipitation and capacity fading, and improving energy efficiency;
  • No oxygen evolution (chlorine would come out first, if any) on the carbon felt cathode electrodes during any overcharging (even locally), avoiding oxidation of the electrode and degradation in the electrode reactivity, and thus raising the voltage up-boundary and electrolyte utilization;
  • Enabling automatic balancing management between cathode- and anode-side electrolytes due to higher vanadium solubilities;
  • Improved tolerance to impurities in the electrolytes due to the chemistry changes over the traditional sulphate VFBs;
  • Improved battery system reliability and durability due to the all aforementioned.

With the new generation electrolyte chemistry, the last version of the ReFlex™ battery demonstrates many improvements over the traditional sulphate VFB. Here is an overview:

What comes to your mind?

It is my great ambition to help everyone further in realizing a clean and sustainable global energy system. In my view, sustainable energy storage is crucial to achieve that. It is a long haul from Alessandro Volta’s first tiny electric battery in the year 1799 to today’s advanced vanadium flow battery that easily powers an entire island. Of course, my team and I did not get it all 100 percent right at the first try. I will devote Part 3 of my series to the tests that we ran on the ReFlex™. We literally asked independent testers to try and break it. I will soon reveal how that ended.

The greatest things have originated from interaction and exchange of ideas and opinions. What options do you see for the use of vanadium flow batteries? Please place a comment to start a conversation, or take it further.