Design of a miniature flow cell for in situ x-ray imaging of redox flow batteries

Abstract

Flow batteries represent a possible grid-scale energy storage solution, having many advantages such as scalability, separation of power and energy capabilities, and simple operation. However, they can suffer from degradation during operation and the characteristics of the felt electrodes are little understood in terms of wetting, compression and pressure drops. Presented here is the design of a miniature flow cell that allows the use of x-ray computed tomography (CT) to study carbon felt materials in situ and operando, in both lab-based and synchrotron CT. Through application of the bespoke cell it is possible to observe felt fibres, electrolyte and pore phases and therefore enables non-destructive characterisation of an array of microstructural parameters during the operation of flow batteries. Furthermore, we expect this design can be readily adapted to the study of other electrochemical systems.

Full text

1 © 2016 IOP Publishing Ltd Printed in the UK
Introduction
Redox ow batteries (RFBs) are seen as a promising tech-
nology for grid-scale storage, given their rapid reversibility
and separation of power and energy capacities [1–7]. Many dif-
ferent ow battery chemistries exist or have been proposed, but
one of the most promising is the all-vanadium ow battery, or
VRFB [8, 9], a schematic of which is shown in gure1. Despite
not employing complicated solid state electrodes as in lithium
ion batteries they still suffer from little-understood degrada-
tion issues stemming from the corrosion of the commonly used
carbon felts at high voltages in acidic environments. In addition,
the wetting and compression of the carbon felts and their effect
on performance of RFBs is poorly understood, with a range of
felt modications also presented in the literature [10–23].
In situ and operando x-ray imaging techniques have
been used to great effect to study microstructural changes in
energy devices such as lithium ion batteries, summarised in a
recent review by Weker and Toney [24], solid oxide fuel cells
[25–32], and polymer electrolyte membrane fuel cells and
electrolysers [33–36]. The unique non-destructive nature of
x-ray imaging allows characterisation of energy materials
under different operating conditions and stages of cycle life-
time, as well as extraction of useful parameters for modelling,
such as porosity, tortuosity, surface area, and shape and size of
electrode particles—changes in which can be followed in 3D
by using increasingly rapid acquisition of x-ray tomographic
data with synchrotron radiation sources [37–41]. Lab-based
x-ray imaging techniques can offer high spatial resolution,
comparable with synchrotron imaging; however, due to sig-
nicantly lower photon ux they have lower temporal resolu-
tion, meaning that it is often not possible to study dynamics
Journal of Physics D: Applied Physics
Design of a miniature ow cell for in situ
x-ray imaging of redox ow batteries
RhodriJervis1, LeonDBrown1, TobiasPNeville1,2, JasonMillichamp1,
DonalPFinegan1, ThomasMMHeenan1, DanJ L Brett1 and
PaulRShearing1,3
1 Department of Chemical Engineering, Electrochemical Innovation Laboratory,
University College London, Torrington Place, London WC1E 7JE, UK
2 Department of Chemical Engineering, Centre for Nature Inspired Engineering,
University College London, Torrington Place, London WC1E 7JE, UK
E-mail: p.shearing@ucl.ac.uk
Received 30 June 2016, revised 7 September 2016
Accepted for publication 14 September 2016
Published 4 October 2016
Abstract
Flow batteries represent a possible grid-scale energy storage solution, having many advantages
such as scalability, separation of power and energy capabilities, and simple operation.
However, they can suffer from degradation during operation and the characteristics of the
felt electrodes are little understood in terms of wetting, compression and pressure drops.
Presented here is the design of a miniature ow cell that allows the use of x-ray computed
tomography (CT) to study carbon felt materials in situ and operando, in both lab-based and
synchrotron CT. Through application of the bespoke cell it is possible to observe felt bres,
electrolyte and pore phases and therefore enables non-destructive characterisation of an array
of microstructural parameters during the operation of ow batteries. Furthermore, we expect
this design can be readily adapted to the study of other electrochemical systems.
Keywords: x-ray imaging, computed tomography, redox ow battery, energy storage, in situ cell
(Some guresmay appear in colour only in the online journal)
R Jervis etal
Printed in the UK
434002
JPAPBE
© 2016 IOP Publishing Ltd
2016
49
J. Phys. D: Appl. Phys.
JPD
0022-3727
10.1088/0022-3727/49/43/434002
Paper
43
Journal of Physics D: Applied Physics
IOP
Original content from this work may be used under the terms
of the Creative Commons Attribution 3.0 licence. Any further
distribution of this work must maintain attribution to the author(s) and the title
of the work, journal citation and DOI.
3 Author to whom any correspondence should be addressed.
0022-3727/16/434002+9$33.00
doi:10.1088/0022-3727/49/43/434002
J. Phys. D: Appl. Phys. 49 (2016) 434002 (9pp)

R Jervis etal
2
or transient phenomena associated with normal operation.
Nevertheless, the advantages of lab-based x-ray CT are often
underestimated—namely, ease of use, accessibility, lower cost
and potentially large scanning areas—and it can be a highly
effective complementary tool to synchrotron imaging [42].
Recently, the potential of x-ray imaging as a diagnostic
tool for corrosion of carbon felts has been demonstrated [43]:
the results showed visible damage and agglomeration of the
carbon bres relatively early on in the charge–discharge
cycle of the RFBs. Combination of the tomography data
with image based modelling revealed the microscopic
transport properties of the felt. However, the micro-CT was
conducted ex situ on felts that had been removed from the
operating environment. Particularly for wetting and com-
pression effects, in situ imaging at large working distances
of operating RFBs would provide valuable insight into the
processes that change the porosity of the felts as charging or
discharging proceeds. Given the hydrophobic nature of the
felts and that the performance of RFBs is intrinsically linked
to the surface area available for reaction, the proportion of
the pores lled with electrolyte is a vital metric in assessing
the quality of different carbon felts, and at different com-
pressions [19, 20]. In addition, calculation of porosity and
tortuosity of the felt materials allows an understanding of the
transport properties of the electrolyte through the electrodes,
a vital factor in the amount of parasitic pumping power
required in the whole system and affecting balance of plant
component design [44, 45].
This paper presents the design of a miniature ow cell with
rotational symmetry, suitable for operando x-ray imaging
in both synchrotron and lab CT setups, allowing real-time
imaging in three dimensions of electrolyte ingression at
varying compressions, calculation of wetting and saturation
parameters, and assessment of the changes in the material
properties upon degradation. Such a device enables the cou-
pling of electrochemical performance with microstructural
evolution and could provide vital insight into the causes and
effects of electrode degradation in ow cells. Details of the
design of such a cell are presented, as well as initial imaging
using both synchrotron and lab sources demonstrating the
potential of x-ray CT as a tool for evaluation of performance
of ow batteries and electrode materials.
For the rst time, we show the capability for imaging of
ow battery carbon felt materials with electrolyte intrusion in
a bespoke cell designed for in situ and operando x-ray tomo-
graphic imaging. The spatial and temporal resolution achiev-
able in lab-based and synchrotron tomography allows vital
properties of a ow cell to be studied in situ and operando
under various operating conditions using such a cell.
Methods
Cell design
Most ow cell designs incorporate a square planar geometry,
with two carbon felt or carbon paper electrodes separated by
an ion conductive membrane, and metallic or graphite cur-
rent collection plates, all held under compression by metallic
end plates with bolts or tie rods. Peripheral pumps deliver
electrolyte from tanks to the electrodes, through the end-
plates, via tubing. Various typical designs are discussed in
the review by Alotto et al [46] and a schematic of a typical
stack is shown in gure2. The materials used in these cells
are highly attenuating to x-rays and therefore a new design
of cell is required to allow imaging of a ow battery. During
collection of tomography data, a sample is rotated through
180–360° relative to an x-ray source and detector, with pro-
jection images collected at discrete angular steps. These pro-
jection images can be reconstructed using back-projection
algorithms to form a 3D volume [47]. To maximise data
quality, the primary concern in cell design is to incorporate
rotational symmetry in the region of interest for imaging.
This reduces the artefacts produced on reconstruction of the
3D image from the 2D projections and allows the attenuation
of the beam to be constant at all angles. Thus, the electrodes,
membrane and current collectors all take the form of disks,
contrary to the square designs more commonly seen in ow
cells. Additionally, exible tubing that connects to the cell is
allowed to wrap around the body of the cell, ensuring that the
electrolyte can be delivered without being in the eld of view
of the imaging.
The cell is constructed from polypropylene to maximise
transmission of x-rays whilst providing chemical resistance
to the 3 M sulphuric acid used in the electrolyte of vanadium
RFBs. The wall thickness of the cell is minimised in order
to reduce x-ray attenuation while maintaining mechanical
strength. The diameter of the electrode volume was selected
to be 11 mm, a compromise between a size that is small
enough to be included in the full region of interest of a CT
scan and large enough that the volume is representative of
Figure 1. A schematic of a VRFB showing two tanks containing
electrolyte with electroactive vanadium species, the positive and
negative electrodes (usually carbon felts) and the proton conducting
membrane. The power of the battery is dictated by the size of the
electrodes, while the energy storage is decoupled and dependant on
the size of the electrolyte tanks. In charging mode V(IV) is oxidised
to V(V) at the positive electrode and V(III) is reduced to V(II) at the
negative electrode. The reactions are reversed for discharge.
J. Phys. D: Appl. Phys. 49 (2016) 434002

R Jervis etal
3
the behaviour of a larger volume used in the majority of ow
batteries. Adjustable pistons containing channels for delivery
and removal of electrolyte, as well as a channel for a cur-
rent collector wire, allow the compression of the electrodes
to be controlled (gures 3(a) and (b)). The range of move-
ment of the pistons allows felts of up to 1.2 cm thickness to
be used in the cell. Currently, the design does not incorpo-
rate ow eld channels but can easily be adapted for future
studies into the effect of ow eld design on electrolyte intru-
sion and distribution. In the design and conguration outlined
in this work, electrolyte is delivered to the surface of the felt
through one of the channels in the piston, ows through the
felt to a second channel in the piston where it exits the elec-
trode volume. Aluminium spacer disks allow the compression
to be controlled accurately and consistently between experi-
ments, and the assembly of the cell is swift and trivial with
aluminium tie rods providing overall compression. The com-
pression provided by the pitons is even and aids in the preven-
tion of ow of the electrolyte around the edges of the felt, and
is easily adjustable allowing for the study of electrolyte ow
behaviour under various compressions. Separation of the elec-
trolyte between the two electrodes is achieved by clamping
the membrane (with two PTFE ring gaskets) onto a shelf in
the main central body of the cell (gure 3(d), red) using a
second internal piece (gure 3(d), blue), and two Viton o-rings
on the adjustable pistons prevent electrolyte leakage via the
outside of the piston. Manufacturing of the cell was carried
out by GGM Engineering (Middlesex, UK). The felts used in
this study are SIGRACELL GFD graphite felts (SGL Group,
Germany) that have been heat treated in air at 400 °C for 30 h
in air in order to thermally oxidise the surface of the bres, as
is often done in RFBs to increase the hydrophilicity [48].
Figure 2. Exploded view of a typical ow battery stack with a square geometry and many metallic components with high x-ray attenuation,
making it unsuitable for x-ray CT investigations.
Figure 3. Design of the mini ow cell (a), with transparent view showing internal workings of the cell (b). The path of electrolyte and
working part of the cell (membrane, felt electrodes and current collection) is shown in (c), and (d) shows a detailed view of the imaging
region, including the sealing that separates the two electrodes achieved by clamping the internal body (blue) onto a shelf created in the
external body of the cell (red). O-rings prevent electrolyte from owing through the gap between the internal body and the adjustable
piston.
J. Phys. D: Appl. Phys. 49 (2016) 434002

R Jervis etal
4
Lab-based x-ray CT
Prior to synchrotron experiments ex situ imaging of felts was
conducted using a lab-based micro-CT machine (Zeiss Xradia
Versa 520, Carl Zeiss XRM, Pleasanton, CA), operating with
a source voltage of 40 kV. Imaging of the ow cell was con-
ducted in the same manner, except with a source voltage of
80 kV to take into account the higher attenuation of the cell
housing and longer absorption path of the x-rays. The number
of projections and exposure time in each scan differed in order
to obtain sufcient counts for each image and the details are
given in the relevant image captions. Electrolyte consisting of
3 M sulphuric acid and 1.6 M vanadium (III/IV) species was
pumped through the cell before the tubes were clamped and
the cell was imaged containing a static volume of electrolyte.
The experimental setup is shown in gure4.
Synchrotron x-ray CT
Following characterisation of the cell using lab based CT,
imaging of the ow cell was also carried out at the TOMCAT
beamline of the Swiss Light Source synchrotron [49], which
owing to the vastly improved x-ray ux allowed for much
quicker acquisition times. These tomograms were collected
using 1801 projections over a 180° rotation with 333 ms expo-
sure time each, with a parallel monochromatic beam at 25 keV.
A 10× optical magnication was employed, giving a pixel size
of 0.65 µm, and radiographs were captured using a lutetium
aluminium garnet:cerium (LAG:Ce) Scintillator and PCO.Edge
5.5 camera. Tomograms were reconstructed using the Gridrec
algorithm [50], and in some cases Paganin reconstruction was
used to enhance the contrast between phases [51].
Results and discussion
Figure 5 shows the x-ray CT of the ow cell containing elec-
trolyte. Virtual slices can be taken through the reconstructed
data at any point and angle, and show a good contrast between
electrolyte, metal, air and the polypropylene cell housing
(gure 5(a)), allowing separation and visualisation of comp-
onent materials, this has been achieved using Avizo Fire soft-
ware (FEI VSG, Mérignac Cedex France) (gure 5(b)) which
allows the cell to be viewed at any angle and with any part of
the cell removed/displayed.
The metal tie rods used to compress the cell cause signi-
cant artefacts due to the high x-ray attenuation of the material
causing areas of low transmission, and hence a lack of informa-
tion that is required in the mathematics of the reconstruction.
This can be seen at the vertical extremes of the xz and yz ortho-
slices in gure5(a), and so the volume was cropped to allow
easier segmentation of the more relevant central region of the
cell. As well as viewing specic parts of the cell at any angle, it
is also possible make virtual cuts into the reconstructed volume
and view the internal parts of the image (gure 5(c)).
In order to ascertain the quality of scans possible in situ in
this cell design, ex situ scans of fresh felt were collected on
the lab-based CT system (gure 6). By using a high number of
projections and minimising the amount of sample outside of
the region of interest it is possible to obtain very good quality,
high resolution images of the felts. The segmented image
could then be used in structure-based modelling to determine
parameters such as pressure drop through the felt, transport
phenomena and local reactant concentrations [45, 52]. The
high quality of the imaging is due in part to there being no
extra material surrounding the felt, and therefore minimal
artefacts and good transmission of x-rays through the sample.
Conversely, gure 7 shows an in situ scan of the ow cell
at a higher magnication, using a 4× objective lens to achieve
an isotropic voxel size of 3.8 µm—high enough resolution to
resolve the bres of the felt volume. The image highlights an
area of the electrode that contains incomplete wetting of the felt
by electrolyte. From the orthoslice (gure 7(a)) it is possible to
differentiate the bres of the felt from the areas of vanadium
electrolyte, and therefore segment the image into the current
collector, felt and electrolyte materials (gures7(b) and (c)).
Comparison with the ex situ felt scans in gure6 shows the
complications of including extra material outside the region
Figure 4. Miniature ow cell containing static electrolyte in the lab-based x-ray CT system. The source (on the left of the image) and
detector (on the right of the image) are brought as close as possible to the cell during imaging to improve the quality of the scan. Inset
shows the cell in more detail.
J. Phys. D: Appl. Phys. 49 (2016) 434002

R Jervis etal
5
of interest and the subsequent inuence it has on the resulting
image quality. Nevertheless, the miniature ow cell shows the
possibility of assessing the degree of wetting of the felts, under
differing conditions of compression, owrate and with different
felt types. Incomplete wetting of the felts not only reduces the
active volume available for the redox reactions to take place in
(and hence introducing mass transport overpotentials and sub-
sequent voltage inefciencies), but may cause areas of local
uctuation in electrolyte concentration and high potentials as
the electrode is starved of reactants [52]. This could lead to
accelerated degradation of felt materials and unwanted side
reactions, such as hydrogen evolution, that could reduce the
lifetime and efciency of the ow battery. The miniature ow
cell also has the potential for operando imaging, which would
allow investigation of bubble formation and possibly degrada-
tion of the electrode materials in real time, particularly when
Figure 5. X-ray CT image of the ow cell containing static electrolyte: (a) three orthoslices of the reconstructed volume along the xy,
xz and yz planes, (b) segmented image created using Avizo software allowing separation of the separate pieces and materials of the cell
(labelled) and showing the position of the xy and yz orthoslices taken in (a) (red lines, xz is in the plane of the image), (c) a virtual cut into
the cell allowing visualisation of the inner sectionsof the ow cell. The scan was conducted on a lab based CT system with 0.4× detector
(effective pixel size of 31 µm) and comprised 3201 projections with an exposure of 10 s each. Comparison with gure3 shows the
components that make up the cell.
Figure 6. Ex situ felt scans obtained using a lab-based CT system with 3201 projections of 15 s exposure using a 20× optical
magnication, resulting in a pixel size of 0.80 µm: (a) a 3D reconstruction of the scan showing felt bres (blue) and air pores (transparent),
(b) an xy orthoslice of the same volume and (c) three orthogonal orthoslices of the felt in the xy, yz and xz planes.
J. Phys. D: Appl. Phys. 49 (2016) 434002

R Jervis etal
6
using a synchrotron x-ray source where the higher photon ux
would allow rapid acquisition of tomograms.
Figure 8 shows images of the electrode area of the ow cell
containing electrolyte fully wetting the felt volume obtained
using a synchrotron radiation source. In this case, the tomo-
grams show improved signal-to-noise ratio than those obtained
in the lab CT scan (gure 7) and allow for easier segmentation
of the bres from the electrolyte phase (gure 8(c)). The high
ux of the synchrotron source also has the benet of much
reduced scan times; in this case 10 min compared to 3 h for
each tomogram which will allow much more wide-ranging
studies of felt wetting and electrolyte ingression at varying
ow rates, compressions and felt types to be conducted in
reasonable time scales. Conversely, despite the long acqui-
sition times for lab-based CT scans, the accessibility of the
facility allows for much longer-range experiments where the
Figure 7. X-ray CT imaging of the ow cell containing static electrolyte obtained using a lab-based CT system with a 4× objective lens
(effective pixel size of 3.8 µm) comprising 1601 projections of 7 s exposure. Areas of incomplete wetting of the felt can be seen in the
orthoslice (a) and segmentation of the Ti current collector (grey), felt (red) and felt containing electrolyte (purple) is possible (b). The
electrolyte volume has been removed in image (c) and a Ti mesh current collector was employed in this instance.
Figure 8. In situ felt scans obtained using a synchrotron light source, containing static electrolyte. (a) Orthoslice showing good contrast
between bre and electrolyte phases, (b) orthogonal orthoslices showing, in this case, a complete wetting of the felt with no areas of gas
lled pores, and (c) a 3D reconstruction of the scan volume showing separation of bres (blue) and electrolyte volume (transparent).
Figure 9. In situ felt scans obtained using a synchrotron light source, showing areas of gas lled pores in the static electrolyte (a). Paganin
reconstruction [51] has been used in (b) and (c) to enhance the contrast between the pore volume and the wetted areas of the felt.
J. Phys. D: Appl. Phys. 49 (2016) 434002

R Jervis etal
7
state of the battery can be periodically examined throughout
the course of many charge–discharge cycles. This can be done
by pausing operation, scanning the cell in situ (but not oper-
ando) and then returning the cell to operation—all of which
can be done without disassembling the cell and, therefore,
changing the operating conditions. Usually, access to a syn-
chrotron radiation source is limited to a short, nite period of
time, during which it might not be possible to study the cell
over many charge–discharge cycles, despite the rapid acquisi-
tion time, as the rate determining step becomes the length of
time the battery takes to charge. Figure9 shows an example
of a felt with incomplete wetting, leading to pockets of empty
pores, as well as areas of electrolyte wetted bres. The con-
trast between these pores and the electrolyte is enhanced in
gure9(b) and (c) by employing a Paganin reconstruction [51]
to improve phase contrast and ease of segmentation enabling
calculation of wetting amounts, pore size distribution and
contact angles, amongst other metrics. However in gure9
the electrolyte and felt phases are difcult to distinguish due
to their similar attenuation coefcients, though this could be
improved in the future with changes to the image acquisition
or by the adoption of a composite absorption and phase con-
trast imaging methodology [53]. Characterisation of wetting
amount and gas pore size and distribution could provide vital
insight into the operation of ow batteries.
It is therefore proposed that a miniature ow cell, such as
the one described in this work, could have the potential for
investigation into the operating properties of ow batteries
and degradation of electrode materials in a non-destructive
manner, not possible with other characterisation methods. We
expect that in situ and operando tomography of ow batteries
will provide vital insights into the behaviour, causes of perfor-
mance loss and degradation of these devices, with the aim of
increasing performance and improving the material properties
of their components.
Conclusions
A miniature ow cell incorporating rotational symmetry and
allowing variable compression of electrodes was designed and
constructed from polypropylene. X-ray tomography has shown
that it is possible to image the electrodes in three dimensions
and segregate the carbon felt material, vanadium electrolyte
and air phases. The cell design shows promise for operando
tomography of RFBs and could incorporate many different
chemistries or even different electrochemical systems, such
as microbial fuel cells. X-ray CT characterisation of ow bat-
teries and the materials used in the devices could provide vital
information about their operation and degradation, as well as
enabling improvement in performance and durability.
Acknowledgments
The authors thank the UK EPSRC for funding (under the grants
EP/L014289/1, EP/N032888/1), and from an EPSRC ‘Fron-
tier Engineering’ Award (EP/K038656/1). PRS acknowledges
the Royal Academy of Engineering for funding. Synchrotron
imaging experiments were performed on the TOMCAT beam-
line at the Swiss Light Source, Paul Scherrer Institut, Villigen,
Switzerland and we acknowledge the beam line staff, in par-
ticular Dr David Haberthür for their contribution. The authors
also thank GGM Engineering for valuable discussion on the
design and manufacturing of the cell.
References
[1] WeberAZ, MenchMM, MeyersJP, RossPN, GostickJT
and LiuQ 2011 Redox ow batteries: a review J. Appl.
Electrochem. 411137–64
[2] KearG, ShahAA and WalshFC 2012 Development of the
all-vanadium redox ow battery for energy storage: a
review of technological, nancial and policy aspects Int. J.
Energy Res. 361105–20
[3] JoerissenL, GarcheJ, FabjanC and TomazicG 2004 Possible
use of vanadium redox-ow batteries for energy storage in
small grids and stand-alone photovoltaic systems J. Power
Sources 12798–104
[4] Skyllas-KazacosM and GrossmithF 1987 Efcient vanadium
redox ow cell J. Electrochem. Soc. 1342950–3
[5] LeungP, LiX, De LeónCP, BerlouisL, LowCJ and
WalshFC 2012 Progress in redox ow batteries, remaining
challenges and their applications in energy storage RSC
Adv. 210125–56
[6] NoackJ, RoznyatovskayaN, HerrT and FischerP 2015 The
chemistry of redox-ow batteries Ang. Chem., Int. Ed.
549776–809
[7] Ponce de LeónC, Frías-FerrerA, González-GarcíaJ,
SzántoDA and WalshFC 2006 Redox ow cells for
energy conversion J. Power Sources 160716–32
[8] BartolozziM 1989 Development of redox ow batteries. A
historical bibliography J. Power Sources 27219–34
[9] Skyllas-KazacosM, ChakrabartiM, HajimolanaS, MjalliF
and SaleemM 2011 Progress in ow battery research and
development J. Electrochem. Soc. 158R55–79
[10] ParasuramanA, LimTM, MenictasC and Skyllas-KazacosM
2013 Review of material research and development for
vanadium redox ow battery applications Electrochim. Acta
10127–40
[11] ChakrabartiMH, BrandonNP, HajimolanaSA, TariqF,
YutV, HashimMA, HussainMA, LowCTJ and
AravindPV 2014 Application of carbon materials in redox
ow batteries J. Power Sources 253150–66
[12] WangW, LuoQ, LiB, WeiX, LiL and YangZ 2013 Recent
progress in redox ow battery research and development
Adv. Funct. Mater. 23970–86
[13] ParkM, JungY-J, KimJ, LeeHI and ChoJ 2013 Synergistic
effect of carbon nanober/nanotube composite catalyst on
carbon felt electrode for high-performance all-vanadium
redox ow battery Nano Lett. 134833–9
[14] Huang Ke-LongTN, Liu Su-Qin and ChenLi-Quan 2006
Electrochemical modication of graphite felt electrode for
vanadium redox ow battery Inorg. Mater. 211114–20
[15] LiW, LiuJ and YanC 2011 Multi-walled carbon nanotubes
used as an electrode reaction catalyst for/VO2+ for a
vanadium redox ow battery Carbon 493463–70
[16] HammerE-M, BergerB and KomsiyskaL 2014 Improvement
of the performance of graphite felt electrodes for vanadium-
redox-ow-batteries by plasma treatment Int. J. Renew.
Energy Dev. 37
[17] TsengT-M, HuangR-H, HuangC-Y, LiuC-C, HsuehK-L and
ShieuF-S 2014 Carbon felt coated with titanium dioxide/
carbon black composite as negative electrode for vanadium
redox ow battery J. Electrochem. Soc. 161A1132–8
J. Phys. D: Appl. Phys. 49 (2016) 434002

R Jervis etal
8
[18] SmithRE, DaviesTJ, Baynes N dB and NicholsRJ 2015
The electrochemical characterisation of graphite felts
J. Electroanal. Chem. 74729–38
[19] GouletM-A, Skyllas-KazacosM and KjeangE 2016 The
importance of wetting in carbon paper electrodes for
vanadium redox reactions Carbon 101390–8
[20] HeZ, ShiL, ShenJ, HeZ and LiuS 2015 Effects of nitrogen
doping on the electrochemical performance of graphite
felts for vanadium redox ow batteries Int. J. Energy Res.
39709–16
[21] RabbowTJ, TrampertM, PokornyP, BinderP and
WhiteheadAH 2015 Variability within a single type
of polyacrylonitrile-based graphite felt after thermal
treatment. Part II: chemical properties Electrochim. Acta
17324–30
[22] ShaoY, WangX, EngelhardM, WangC, DaiS, LiuJ, YangZ
and LinY 2010 Nitrogen-doped mesoporous carbon for
energy storage in vanadium redox ow batteries J. Power
Sources 1954375–9
[23] WangWH and WangXD 2007 Investigation of Ir-modied
carbon felt as the positive electrode of an all-vanadium
redox ow battery Electrochim. Acta 526755–62
[24] Nelson WekerJ and ToneyMF 2015 Emerging in situ and
operando nanoscale x-ray imaging techniques for energy
storage materials Adv. Funct. Mater. 251622–37
[25] NelsonGJ, HarrisWM, LombardoJJ, IzzoJR, ChiuWK,
TanasiniP, CantoniM, ComninellisC, AndrewsJC and
LiuY 2011 Comparison of SOFC cathode microstructure
quantied using x-ray nanotomography and focused
ion beam–scanning electron microscopy Electrochem.
Commun. 13586–9
[26] ShearingP, BradleyR, GelbJ, TariqF, WithersP and
BrandonN 2012 Exploring microstructural changes
associated with oxidation in Ni-YSZ SOFC electrodes
using high resolution x-ray computed tomography Solid
State Ion. 21669–72
[27] GuanY, LiW, GongY, LiuG, ZhangX, ChenJ, GelbJ,
YunW, XiongY and TianY 2011 Analysis of the three-
dimensional microstructure of a solid-oxide fuel cell anode
using nano x-ray tomography J. Power Sources 1961915–9
[28] ShearingP, BradleyR, GelbJ, LeeS, AtkinsonA, WithersP
and BrandonN 2011 Using synchrotron x-ray nano-CT
to characterize SOFC electrode microstructures in three-
dimensions at operating temperature Electrochem. Solid
State Lett. 14B117–20
[29] ShearingP, GelbJ, YiJ, LeeW-K, DrakopolousM and
BrandonN 2010 Analysis of triple phase contact in Ni-YSZ
microstructures using non-destructive x-ray tomography
with synchrotron radiation Electrochem. Commun.
121021–4
[30] IzzoJR, JoshiAS, GrewKN, ChiuWK, TkachukA,
WangSH and YunW 2008 Nondestructive reconstruction
and analysis of SOFC anodes using x-ray computed
tomography at sub-50 nm resolution J. Electrochem. Soc.
155B504–8
[31] ShearingP, GelbJ and BrandonN 2010 X-ray nano
computerised tomography of SOFC electrodes using a
focused ion beam sample-preparation technique J. Eur.
Ceram. Soc. 301809–14
[32] ShearingPR, EastwoodDS, BradleyRS, GelbJ, CooperSJ,
TariqF and LeeP 2013 Exploring electrochemical devices
using x-ray microscopy: 3D micro-structure of batteries and
fuel cells Microsc. Anal. 2719–22
[33] HartnigC, MankeI, SchloesserJ, KrügerP, KuhnR,
RiesemeierH, WippermannK and BanhartJ 2009 High
resolution synchrotron x-ray investigation of carbon
dioxide evolution in operating direct methanol fuel cells
Electrochem. Commun. 111559–62
[34] HartnigC, MankeI, KuhnR, KardjilovN, BanhartJ and
LehnertW 2008 Cross-sectional insight in the water
evolution and transport in polymer electrolyte fuel cells
Appl. Phys. Lett. 92134106
[35] MankeI, HartnigC, GrünerbelM, LehnertW, KardjilovN,
HaibelA, HilgerA, BanhartJ and RiesemeierH 2007
Investigation of water evolution and transport in fuel cells
with high resolution synchrotron x-ray radiography Appl.
Phys. Lett. 90174105
[36] HartnigC, MankeI, KuhnR, KleinauS, GoebbelsJ and
BanhartJ 2009 High-resolution in-plane investigation of the
water evolution and transport in PEM fuel cells J. Power
Sources 188468–74
[37] FineganDP, ScheelM, RobinsonJB, TjadenB, HuntI,
MasonTJ, MillichampJ, Di MichielM, OfferGJ and
HindsG 2015 In-operando high-speed tomography
of lithium-ion batteries during thermal runaway Nat.
Commun. 6 6924
[38] FineganDP etal 2016 Quantifying bulk electrode strain and
material displacement within lithium batteries via high-
speed operando tomography and digital volume correlation
Adv. Sci. 3 1500332
[39] CooperSJ etal 2014 Image based modelling of
microstructural heterogeneity in LiFePO4 electrodes for
Li-ion batteries J. Power Sources 2471033–9
[40] Di MichielM, MerinoJM, Fernandez-CarreirasD, BuslapsT,
HonkimäkiV, FalusP, MartinsT and SvenssonO 2005 Fast
microtomography using high energy synchrotron radiation
Rev. Sci. Instrum. 76043702
[41] Chen-WiegartY-CK, ShearingP, YuanQ, TkachukA and
WangJ 2012 3D morphological evolution of Li-ion battery
negative electrode LiVO2 during oxidation using x-ray
nano-tomography Electrochem. Commun. 2158–61
[42] BrunkeO, BrockdorfK, DrewsS, MüllerB, DonathT,
HerzenJ and BeckmannF 2008 Comparison Between
X-Ray Tube-Based and Synchrotron Radiation-Based µCT
707801–12
[43] TrogadasP, TaiwoOO, TjadenB, NevilleTP, YunS,
ParrondoJ, RamaniV, CoppensM-O, BrettDJL and
ShearingPR 2014 X-ray micro-tomography as a diagnostic
tool for the electrode degradation in vanadium redox ow
batteries Electrochem. Commun. 48155–9
[44] ViswanathanV, CrawfordA, StephensonD, KimS, WangW,
LiB, CoffeyG, ThomsenE, GraffG and BalducciP 2014
Cost and performance model for redox ow batteries
J. Power Sources 2471040–51
[45] TangA, BaoJ and Skyllas-KazacosM 2014 Studies on
pressure losses and ow rate optimization in vanadium
redox ow battery J. Power Sources 248154–62
[46] AlottoP, GuarnieriM and MoroF 2014 Redox ow batteries
for the storage of renewable energy: a review Renew.
Sustain. Energy Rev. 29325–35
[47] ZieglerA, KöhlerT and ProksaR 2007 Noise and resolution
in images reconstructed with FBP and OSC algorithms for
CT Med. Phys. 34585–98
[48] SunB and Skyllas-KazacosM 1992 Modication of graphite
electrode materials for vanadium redox ow battery
application—I. Thermal treatment Electrochim. Acta
371253–60
[49] StampanoniM, GrosoA, IseneggerA, MikuljanG, ChenQ,
BertrandA, HeneinS, BetempsR, FrommherzU
and BöhlerP 2006 Trends in synchrotron-based
tomographic imaging: the SLS experience SPIE
Optics + Photonics: International Society for Optics
and Photonics 63180 63180M
[50] MaroneF and StampanoniM 2012 Regridding reconstruction
algorithm for real-time tomographic imaging J. Synchrotron
Radiat. 191029–37
J. Phys. D: Appl. Phys. 49 (2016) 434002

R Jervis etal
9
[51] PaganinD, MayoSC, GureyevTE, MillerPR and
WilkinsSW 2002 Simultaneous phase and amplitude
extraction from a single defocused image of a homogeneous
object J. Microsc. 20633–40
[52] QiuG, JoshiAS, DennisonCR, KnehrKW, KumburEC
and SunY 2012 3D pore-scale resolved model for coupled
species/charge/uid transport in a vanadium redox ow
battery Electrochim. Acta 6446–64
[53] TaiwoOO, FineganDP, GelbJ, HolznerC, BrettDJL
and ShearingPR 2016 The use of contrast enhancement
techniques in x-ray imaging of lithium-ion battery
electrodes Chem. Eng. Sci. 154 27–33
J. Phys. D: Appl. Phys. 49 (2016) 434002