Study of novel redox flow batteries based on double-membrane, single-membrane, and membrane-less cell configurations
Date
2016
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
Wide deployment of intermittent energy generation (e.g., wind and solar) calls
for low-cost energy storage system for smooth and reliable power output. Redox flow
batteries (RFBs) have been identified as one of the most suitable systems for largescale
energy storage. Different from conventional batteries that store energy in solid
electrode, RFBs take advantage of flowable electrolytes as energy-storage media and
therefore bring unprecedented freedom in independent tuning of energy and power of
RFB. The method to separate two chemically reactive electrolytes plays a key role in
RFB. Current RFBs adopt a single ion-exchange membrane (IEM) as separator, which
can physically separate two electrolytes but ionically conduct them with commuting
ions. Ever since the invention in 1974, the single-membrane configuration has enabled
a tremendous amount of new combinations of elements from periodic table for battery
application. However, single IEM configuration remains imperfect: 1) IEM is
designed to either conduct cation while excluding anion (cation-exchange membrane,
CEM), or conduct anion while excluding cation (anion-exchange membrane, AEM).
This property only allows the combination of redox pairs in the same type of charge,
leaving a lot of promising redox pair combinations useless; 2) IEM cannot reach 100%
selectivity of commuting ion, which results in an inevitable crossover of redox pairs,
causing electrolyte imbalance, coulombic efficiency and capacity loss; 3) IEM
contributes the biggest voltage loss due to its large internal resistance in many RFBs,
and is usually one of the most expensive components in the stack, both indirectly or
directly increasing the cost of RFBs.
Aiming at solving the problems in single-membrane RFBs, this work explored
three possible routes that provide alternative configurations to current RFBs: 1) a
double-membrane RFB that could combine redox pairs with different types of charge,
and of different supporting pHs; 2) a single-membrane all-iron (all-Fe) flow battery
that adopts the same elements on both sides, which is immune to the crossover of
metal ions; 3) a membrane-less RFB that utilizes immiscible organic and inorganic
electrolytes, which thermodynamically separate two redox species and eliminate the
usage of membrane in RFB.
In the double-membrane RFB design, both AEM and CEM are incorporated in
cell to isolate cation and anion redox pairs respectively. A middle electrolyte is used to
ionically conduct two membranes. Three examples have been successfully
demonstrated: Zn-Ce (Zn(OH)4
2−/Zn vs. Ce4+/Ce3+), S-Fe (S4
2−/S2
2− vs. Fe3+/Fe2+) and
Zn-Fe (Zn(OH)4
2−/Zn vs. Fe3+/Fe2+) RFBs. Zn-Ce RFB provides the highest cell
voltage among all aqueous RFBs as 3.08 V. S-Fe RFB combines very inexpensive
anion redox pair (S4
2−/S2
2−) and cation redox pair (Fe3+/Fe2+) together (1.22 V) and
brings low electrolyte cost. Zn-Fe RFB has the best balance between high voltage (2.0
V) and low electrolyte cost, thus bringing high performance and low capital cost.
Middle electrolyte was found to be an important role in controlling total cell
resistance. With optimally engineered middle electrolyte, Zn-Fe RFB shows high
power density (676 mW/cm2) and the lowest system cost so far among several notable
RFBs, under $100/kWh, which is below the cost target for energy storage system set
by Department of Energy of U.S. in the 2023 term. Such a low cost puts Zn-Fe RFB in
a very promising position for future development and commercialization.
In the single-membrane all-Fe RFB, the same element, iron, is used in redox
pairs in both positive and negative electrolytes with different coordination chemistries.
The adoption of the same element fundamentally eliminates the cross-contamination
in RFBs that uses two different elements. All-Fe RFB shows good durability and
stability over cycle test. The slow diffusion of coordinate agent, however, was
identified as a prominent concern in capacity retention in long-term. Nonetheless, all-
Fe RFB remains as a good attempt in combining redox pairs of the same element with
different coordination chemistries to extend the spectrum of redox pairs for RFB
application.
In the membrane-less RFB design, a new separation method of redox pairs is
introduced by employing immiscible organic and inorganic electrolytes. Redox pairs
are thus thermodynamically separated and require no membrane. A zinc-ferrocene
RFB was demonstrated as an example for this membrane-less design and good
durability and stability were proved in cycle test. This concept broadens the method to
construct flow battery and brings more possible combinations between organic and
inorganic redox pairs in RFB application.
The new designs and concepts studied in this work successfully demonstrated
that invention of new cell structure could greatly enrich and diversify the category of
RFBs, expanding new redox chemistries and enabling new redox pair combinations
for RFB. Setting those three cell designs as frame work, we are expecting and looking
forward to more exciting redox chemistries being explored.