Electrochemical behavior of Bi4B2O9 towards lithium-reversible conversion reactions without nanosizing

Article abstract of DOI:10.1039/c7cp07693b

Conversion type materials, in particular metal fluorides, have emerged as attractive candidates for positive electrodes in next generation Li-ion batteries (LIBs). However, their practical use is being hindered by issues related to reversibility and large polarization. To minimize these issues, a few approaches enlisting the anionic network have been considered. We herein report the electrochemical properties of bismuth oxyborate Bi₄B₂O₉ and show that this compound reacts with lithium via a conversion reaction leading to a sustained capacity of 140 mA h g^-1 when cycled between 1.7 and 3.5 V vs. Li⁺/Li⁰ while having a surprisingly small polarization (～300 mV) in the presence of solely 5% in weight of a carbon additive. These observations are rationalized in terms of charge transfer kinetics via complementary XRD, HRTEM and NMR measurements. This finding demonstrates that borate based conversion type materials display rapid charge transfer with limited carbon additives, hence offering a new strategy to improve their overall cycling efficiency.

Full text of DOI:10.1039/c7cp07693b

View Article Online
View Journal
PCCP
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: F. Strauss, G.
Rousse, D. Batuk, M. Tang, E. Salager, G. Draži, R. Dominko and J. Tarascon, Phys. Chem. Chem. Phys.,
2017, DOI: 10.1039/C7CP07693B.
Volume 18 Number
1
7
January 2016 Pages 1–636
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
PCCP
Physical Chemistry Chemical Physics
www.rsc.org/pccp
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1463-9076
PERSPECTIVE
Darya Radziuk and Helmuth Möhwald
Ultrasonically treated liquid interfaces for progress in cleaning and
separation processes
rsc.li/pccp

Page 1 of 21
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C7CP07693B
Electrochemical behavior of Bi B O towards lithium –
4
2
9
reversible conversion reaction without nanosizing
a,b,c,#
b,c,d,e
b,f
g,e,+
Florian Strauss
, Gwenaëlle Rousse
, Dmitry Batuk , Mingxue Tang , Elodie
g,e
a
a,c
b,c,d,e,
Salager , Goran Dražić , Robert Dominko , Jean-Marie Tarascon
*
a
National Institute of Chemistry, Hajdrihova 19, SI-1000 Ljubljana, Slovenia
b
UMR 8260 “Chimie du Solide et Energie”, Collège de France, 11 Place Marcelin Berthelot, 75231 Paris Cedex 05, France
ALISTORE – European Research Institute, 33 rue Saint-Leu, Amiens 80039 Cedex, France
c
d
e
f
UPMC Univ Paris 06, Sorbonne Universités, 4 Place Jussieu, F-75005 Paris, France
Réseau sur le Stockage Electrochimique de l’Energie (RS2E), FR CNRS 3459, 80039 Amiens Cedex, France
EMAT, University of Antwerp, Groenenborgerlaan 171, B12020, Antwerp, Belgium
g
Université d’Orléans, CNRS, CEMHTI UPR3079, 1D avenue de la recherche scientifique, 45071 Orléans Cedex 2, France
Abstract
Conversion type materials, in particular metal fluorides, have emerged as attractive
candidates for positive electrodes in next generation Li-ion batteries (LIB’s). However, their
practical use is being hindered by issues related to reversibility and large polarization. To
minimize these issues, few approaches enlisting the anionic network have been considered.
We herein report the electrochemical properties of a bismuth oxyborate Bi B O and show
4
2 9
that this compound reacts with lithium via a conversion reaction leading to a sustained
+
0
capacity of 140 mAh/g when cycled between 1.7 and 3.5 V vs. Li /Li while having a
surprisingly small polarization (~300 mV) in presence of solely 5% in weight of carbon
additive. These observations are rationalized in terms of charge transfer kinetics via
complementary XRD, HRTEM and NMR measurements. This finding demonstrates that
borate based conversion type materials display rapid charge transfer with limited carbon
additives, hence offering a new strategy to improve their overall cycling efficiency.
1

Physical Chemistry Chemical Physics
Page 2 of 21
View Article Online
DOI: 10.1039/C7CP07693B
Introduction
Lithium ion batteries (LIB’s) play an important role in powering portable electronics.
Moreover, they are conquering the automotive market and stand as a contender for
1,2
stationary energy storage. Presently, the state of the art materials for LIB’s are based on
intercalation compounds which rely either on cationic (e.g. LiCoO , LiFePO ) or cumulative
2
4
cationic/anionic (Li-rich layered oxides Li1+x
M O
1-x 2
, M = Mn, Co, Ni) redox processes as
2
-
recently reported. This latter finding has enabled the feasibility to reach 2 e per d-metal as
3
for β-Li
2 3
IrO .
Nevertheless, electrodes functioning via conversion reactions were first
4
mentioned back in the 1978’s and fully demonstrated in 2000 for 3d transition metal oxides
5
(
CoO, NiO…). These reactions, demonstrated for binary metal oxides, were latter on
generalized to metal sulfides, nitrides, halides, phosphides. They enlist upon electrochemical
reduction a decomposition of the binary phases into metal nanoparticles embedded into a
6
,7
2 2 3
matrix of Li O, Li S, Li P, according to the reaction scheme below (in the case of oxides).
ꢇ
ꢊ
ꢌ
ꢀꢁ
ꢂ ꢄꢄ+ ꢄꢄꢄ2ꢅꢄꢆ ꢄꢄ+ ꢄꢄꢄ2ꢅꢄꢈꢉ ꢄꢄꢄ↔ ꢄꢄꢄꢋꢄꢀ ꢄꢄ+ ꢄꢄꢄꢅꢄꢈꢉ ꢂ
(1)
ꢃ
ꢍ
Such conversion reactions can deliver capacities exceeding 1000 mAh/g as compared
to 280 mAh/g for the best insertion electrodes, with potentials ranging from 0.2 V to 3 V
depending on the nature of the anion. The more electronegative the anion is, the higher the
redox potential. This explains the use of 3d-metal based fluorides as positive electrodes.
However, the capacity benefits offered by these conversion electrodes are negated by a
large irreversible capacity (~ 25%) during the first cycle and poor energy efficiency associated
with a large polarization between charge and discharge. Additionally, in presence of highly
insulating compounds (fluorides), there is a need to prepare nano-composite electrodes
heavily loaded in carbon (~ 30%) to ensure a proper electronic conduction. This
electrochemically dead carbon weight penalizes the overall electrode capacity. Nevertheless,
positive electrodes based on either FeF
2
, FeF
3
or BiF
3
were intensively studied together with
II
III
8–12
their corresponding oxyfluorides FeOF
,
Fe (1-x)Fe
O
x x
F2-x and BiO
x
F
3-2x
.
The partial
substitution of fluorine for oxygen was purposely done to enhance electronic conduction.
Nevertheless, the large voltage hysteresis, which is known to increase from hydrides to
7
,13
phosphides, oxides and fluorides,
still remained an issue in terms of round trip energy
efficiency. Herein, to further explore the effect of the anion electronegativity on the voltage
3
-
polarization, we decided to study Bi compounds based on polyanions (BO
2
3
) . Thus, our

Page 3 of 21
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C7CP07693B
interest for bismuth oxyborate (Bi
4
B
2
O
9
) which has been intensively studied for its optical
1
4,15
properties.
We investigate for the first time the reaction mechanism of Bi versus Li
4 2 9
B O
and found that this compound can deliver reversible capacities of 140 mAh/g with an
average redox potential of 2.3 V and a surprisingly small polarization (~300 mV), while solely
using 5 wt.% of carbon additive. Complementary galvanostatic measurements combined
with in- and ex situ X-ray diffraction (XRD), high resolution transmission electron microscopy
7
(
HRTEM) and Li nuclear magnetic resonance (NMR) spectroscopy are used to rationalize
such findings.
Experimental section
a) Synthesis: Bismuth oxyborate Bi B O was obtained by mixing stoichiometric amounts of
4
2 9
bismuth oxide Bi O (Alfa Aesar) and boric acid H BO (Alfa Aesar 99+%) with a mortar and
2
3
3
3
pestle and annealing the mixture at 600°C for 36 h in air inside an alumina crucible with one
intermediate grinding. Prior to usage, Bi O has been annealed at 700°C for 12h in air in
2
3
order to decompose bismuth carbonates which partially form during storage of Bi O in
2
3
ambient atmosphere. After the heat treatment the purity of Bi O was checked through XRD
2
3
with the corresponding Rietveld refinement shown in Figure S1 indicating that Bi O is single
2
3
phase and presents a good cristallinity.
b) Electrochemical characterization: Electrochemical test were carried out in Swagelok-type
cells. The active material was prepared by mixing 300 mg Bi B O powder with 5 wt.%
4
2 9
3
Carbon Super P (Alfa Aesar 99+%, BET surface area 62 m /g) in a stainless steel jar (10 mL
volume) and one stainless steel ball (4 g) for 10 min using a SPEX high energy miller. The cells
(
Swagelok type) were assembled in an argon filled glove box (MBraun, Germany, O
.1 ppm) using a metallic lithium disc as the anode. Approximately 10 mg of the active
material in powdered form was used as the cathode, separated by two sheets of glassy fiber
separator soaked with LP100 electrolyte (1M LiPF in a mixture of ethylene carbonate
2 2
/ H O <
0
6
EC/propylene carbonate PC/dimethyl carbonate DMC 1/1/3 w/w/w). Galvanostatic charge/
discharge cycling was performed at room temperature on a VMP3 potentiostat (BioLogic
S.A., Claix, France). The current was set to a specific C-rate, where a rate of 1C corresponds
+
to a current which inserts/removes 1 Li per formula unit during 1 hour.
3

Physical Chemistry Chemical Physics
Page 4 of 21
View Article Online
DOI: 10.1039/C7CP07693B
c) Physical characterization:
X-ray powder diffraction patterns were recorded in Bragg-Brentano geometry with a Bruker
D8 Advance diffractometer equipped with a copper source (λCu-Kα1 = 1.54056 Å, λCu-Kα2
.54439 Å) and a LynxEye detector. An in-house built sample holder equipped with a Be
=
1
1
6
window transparent to X-rays was used for collecting the XRD patterns of the air sensitive
samples. Prior to examination, the cycled powders were recovered inside the glove box,
washed 3 times with DMC, dried in vacuum and filled into the sample holder without
exposing them to ambient atmosphere at any time.
For in situ XRD experiments a similar setup as described above was used and the cell was
discharged/charged using a C/10 rate corresponding to the uptake/removal of 0.1 Li per 1
hour. Patterns were continuous recorded for a 2θ range from 5 to 60° with a collection time
of 1 hour for each scan.
1
7
Powder patterns were refined using the Rietveld method as implemented in the FullProf
1
8
4 2 9
program. Pristine samples of Bi B O were investigated through combined HRTEM and
electron energy loss spectroscopy (EELS) analysis using a Cs probe corrected Scanning
Transmission Electron Microscope Jeol ARM 200 CF equipped with Centurio EDXS system
2
from Jeol with 100 mm SDD detector and Gatan Quantum ER Dual EELS system.
4 2 9
To examine cycled samples of Bi B O /C, the recovered powder was grinded in a mortar and
holey carbon TEM grids were dipped into the powder. To avoid exposure to air, the samples
were stored and prepared for TEM in an Ar filled glovebox. A special Gatan vacuum holder
was used for the transfer of the samples from the glovebox to the TEM in order to avoid air
exposure. Electron diffraction (ED) data, EDX spectroscopy elemental maps and high
resolution transmission electron microscopy (HRTEM) images were acquired on an FEI Osiris
microscope operated at 200 kV and equipped with a Super-X EDX detector.
7
Li magic-angle-spinning (MAS) NMR experiments were performed on a Bruker Avance III
7
spectrometer in a 4.7 T magnetic field with a Li Larmor frequency of 77.8 MHz. Electrodes
were packed into 1.3 mm rotors inside an Ar-filled glovebox and spun at a MAS rate of 62.5
kHz (unless specifically stated elsewhere) under N
2
protection. The recycle delay was varied
o
from 8 to 20 s to make sure that the excited spins were fully relaxed. The 90 pulse length
7
was calibrated as 1.38 µs. LiCl(s) with a Li chemical shift at 0 ppm was used as the shift
reference. The spectra were calibrated by samples weight and number of scans.
4

Page 5 of 21
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C7CP07693B
Results and Discussion
a) Structural characterization
4 2 9
The resulting yellowish Bi B O powder made at 600°C consists of agglomerated
crystals whose size ranges from 5 to 10 μm (Figure 1, inset). The corresponding XRD pattern
and the Rietveld refinement recorded for a 2θ ranging from 5 to 100° are shown in Figure 1.
All peaks could be indexed in a monoclinic space group P 2
1
/c with the unit cell parameters a
3
=
11.1153(8) Å, b = 6.6324(1) Å, c = 11.0447(3) Å, β = 91.0465(3)° and V = 814.009(4) Å , as
1
4,19
reported in literature.
All atoms are placed in general positions 4e with four
crystallographic sites for bismuth, two for boron and nine for oxygen atoms. The final atomic
positions are given in Table S1.
Figure 1: Rietveld refinement of Bi B O and representative SEM images of the powder
4
2 9
obtained after the solid-state synthesis. The red crosses, black continuous line and bottom
green line represent the observed, calculated, and difference patterns respectively. Upper
and lower vertical blue tick bars are the Bragg positions of Bi
39 (0.4 wt.%).
4 2 9
B O and the impurity phase
2
Bi24B O
The atomic arrangement within the unit cell is shown in Figure 2. Boron is trigonal
3
-
planar coordinated by oxygen forming regular BO
5
3
triangles (average B-O bond distance

Physical Chemistry Chemical Physics
Page 6 of 21
View Article Online
DOI: 10.1039/C7CP07693B
1
.3396 Å) which are stacked in parallel along the b-axis (Figure 2). Bismuth is four- and five-
fold coordinated by oxygen forming asymmetric truncated BiO and BiO polyhedra with Bi-
4
5
O distances ranging from 2.150 to 2.492 Å. Note that three crystallographic independent
oxygen atoms (denoted O1, O2 and O4) are solely connected to Bi atoms and are not part of
4 2 9
borate groups, hence Bi B O can be called an oxyborate.
Figure 2: Structure of Bi
4 2 9
B O projected along the b-axis. Bismuth, boron and oxygen are
drawn in purple, green and red respectively.
Figure S2a shows a typical XRD pattern of the Bi
carbon black for 10 minutes. There is a slight broadening of the Bragg peaks pertaining to
Bi , indicative of a decrease in the sample particle size. Rietveld refinements were
4 2 9
B O /C composite after ball milling with
4 2 9
B O
undertaken with an isotropic size parameter (Scherrer formula) as a refinable parameter.
Results indicate a decrease in the average crystallite size from ~6 µm to ~680 nm for the
4 2 9
pristine and milled Bi B O , respectively. This was further confirmed by electron microscopy
of the carbon/active material composite as shown in Figure S2b.
6

Page 7 of 21
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C7CP07693B
b) Electrochemistry
A typical voltage composition curve in galvanostatic mode between 1.0 and 3.5 V together
with the corresponding derivative plot is shown in Figure 3 and capacity retention is shown
+
in Figure 4. The initial reduction corresponds to the reaction of 12 Li with one mole of
Bi
4
B
2
O
9
. Such a Li uptake is associated with two different voltage regimes. At the beginning
+
of the discharge, a flat plateau at 1.8 V is visible concomitant with the uptake of 8 Li ,
followed by an S-shape domain centered around 1.4 V and corresponding to the uptake of 4
+
+
Li . On following charge, 8 Li can be extracted via two different voltage regimes located at
~
1.6 and ~2.5 V (Figure 3b). On subsequent cycling, the initial reduction processes around
.8 and 1.4 V become reversible with the former one shifted up to 2.2 V through the
1
subsequent discharges (Figure 3b). This potential difference between the initial and
following discharge is typically observed for conversion type materials since different
reaction pathways occur during the initial and subsequent reductions. The potential drop is
larger during the first discharge because of the higher activation energy needed to trigger
5
the conversion reaction as explained in the first seminal paper on this topic. After this
formation step, the second and subsequent discharges will be kinetically less limited owing
2
0
to the nano character of the formed composite, hence the discharge potential is increased.
An irreversible loss in capacity between the first and second discharge (about 33%) is
observed and the initial capacity decays of about 50% within the first 3 cycles (Figure 4).
+
0
Figure 3: (a) Voltage-composition curve of Bi
4 2 9
B O versus Li /Li cycled at a C/10 rate
between 1.0 and 3.5 V and (b) the corresponding derivative plot.
7

Physical Chemistry Chemical Physics
Page 8 of 21
View Article Online
DOI: 10.1039/C7CP07693B
Figure 4: Capacity retention of Bi B O /Li half-cell cycled between 1.0 and 3.5 V with a C/10
4
2 9
rate.
The complete reduction of Bi
4
B
2
O
9
via a conversion reaction would require, according
+
-
to reaction 2, the uptake of 12 Li and 12 e to form elemental Bi, lithium oxide Li
2
O and
3 3
lithium borate Li BO . This number is in good agreement with the amount of reacted Li
deduced from the electrochemical trace down to 1.0 V (Figure 3a), however the reaction is
only partially reversible as not the entire Li could be extracted upon subsequent charge to
3
.5 V; a main characteristic of conversion reaction which always shows large irreversibility
5
between first discharge and first charge.
ꢊ
ꢇ
ꢌ
ꢎꢉ ꢎ ꢂ ꢄꢄ+ ꢄꢄꢄ12ꢄꢈꢉ ꢄꢄ+ ꢄꢄꢄ12ꢄꢆ ꢄꢄꢄ↔ ꢄꢄꢄ3ꢄꢈꢉ ꢂꢄꢄ + ꢄꢄꢄ2ꢄꢈꢉ ꢎꢂ ꢄ+ ꢄꢄ4ꢄꢎꢉ
(2)
ꢏ
ꢍ
ꢐ
ꢍ
ꢑ
ꢑ
X-ray diffraction measurements upon cycling were realized to examine Li-driven
structural changes in Bi using a homemade cell equipped with a Be window transparent
4 2 9
B O
16
to X-rays. We initially scrutinized the sloping voltage region at the beginning of the initial
reduction centered around 1.9 V (green rectangle, Figure 3a). The collected in situ XRD
patterns together with the corresponding electrochemical trace are presented in Figure 5.
One can notice constancy in the intensity of the reflections (zoom Fig. 5b) and that neither a
shift of the pristine reflections nor an appearance of a new phase was observed. However
the onset of an insertion prior to a conversion reaction has been observed in literature for
2
1–23
several other conversion materials.
Although a possible Li contribution in the XRD
8

Page 9 of 21
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C7CP07693B
patterns should be heavily masked by Bi (Z = 83), Li intercalation would lead to a change in
lattice parameters. However the refined lattice parameters based on the in situ recorded
XRD patterns are shown to be constant (Figure S3) for a Li uptake for x = 1, hence a Li
insertion process into Bi
4
B
2
O
9
can be excluded.
+
Figure 5: In situ XRD of a Bi
4
B
2
O
9
/Li half-cell cycled at C/10 for the uptake of 1 Li . (a) Global
view of the overall scanned 2θ angle, whereas in (b) a magnification of the main reflections
is shown and in (c) the corresponding electrochemical trace is presented.
To pin down the origin of the process occurring at the beginning of the initial discharge,
4 2 9
Bi B O /Li cells were discharged with different C-rates, ranging from 1C to C/100 (Figure
S4a). When the C-rate decreases, we note a more defined pseudo-plateau around 1.95 V
which disappears for rates higher than C/5 indicating kinetic limitations of this process. Since
+
in situ XRD ruled out a Li insertion process, we hypothesize that the initial pseudo-plateau
region is most likely related to surface reactions. This hypothesis is confirmed by comparing
4 2 9
the initial discharge between a ball-milled and handground Bi B O /C composite with
namely an increase (Figure S4b) of the initial sloping region from x = 0.25 to x = 1 for the ball
milled samples which present the greatest surface area.
Interestingly, HRTEM (Figure 6a) provides evidence for the presence of a ~3 nm thick
amorphous surface layer on pristine Bi
amorphous bismuth oxide Bi (Figure 6b, c). This observation suggests the removal of
some oxygen from the amorphous surface layer during the sloping initial voltage region to
4 2 9
B O particles, which is mainly made up of
x y
O
lead to the formation of Li
2
O and an oxygen deficient surface Bi
9
x
Oy-z. In favor of such an

Physical Chemistry Chemical Physics
Page 10 of 21
View Article Online
DOI: 10.1039/C7CP07693B
4 2 9
explanation let us recall that the voltage-composition profile of Bi B O versus Li presents a
profile alike the one obtained during the electrochemical reduction of BIMEVOX phases
2
4,25
(
Bi
4
V
2
O11, Bi3.6Pb0.4
V
2
O11-y, Bi
4
V1.8Cu0.2
O11-y),
which has been previously reported as being
nested in an oxygen removal process of the pristine material upon reduction, yielding to
2
4
Bi
a similar reaction takes place in the case of Bi
able to any further insights through TEM analysis, due to the beam sensitivity of the milled
Bi particles once discharged.
4
V
2
O10.66 and an amorphous Li
2
O shell formed around the particles. Thus we assume that
4 2 9
B O
. Note that we were unfortunately not
4 2 9
B O
4 2 9
Figure 6: (a) STEM picture of a Bi B O particle as synthesized where an amorphous surface
layer is highlighted between the two white dashed lines. (b) EELS analysis along the orange
arrow indicated in (a) where the amorphous surface layer is indicated with the grey shaded
area and (c) shows a zoom of that region.
The in situ XRD patterns and the corresponding electrochemical trace when the
discharge is further continued down to 1.7 V are shown in Figure S5. There is a decrease in
the intensity of the pristine reflections to the expense of a broad bumpy background for
angles greater than 22.5° 2θ (Figure S6a). This is indicative of the appearance of some diffuse
x 4 2 9
scattering, implying a Li-driven amorphization of the “Li Bi B O ” material.
Additional ex situ XRD patterns were recorded for two different samples that have
been discharged to 1.0 V and charged back to 3.5 V (Fig. S6b). The powders, prepared by
disassembling cycled cells inside the glove box, were recovered, washed twice with DMC and
dried under vacuum prior to measurement. The X-ray powder pattern collected at 1.0 V is
featureless with the exception of some weak reflections over the 28°- 32° 2θ region,
implying a pronounced amorphization of the “Li
x 4 2 9
Bi B O ” material. These weak features are
still present in the XRD pattern fully charged (3.5 V) sample, hence indicating that they are
1
0

Page 11 of 21
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C7CP07693B
likely due to traces of non-electrochemically connected pristine materials. On subsequent
cycling, we noticed that the amorphous nature of the electrode composite remains.
To get further insights into the Li-driven structural/morphological changes on the
partly reduced or oxidized samples, we performed TEM measurements. A representative
4 2 9
image of Bi B O reduced to 1.0 V (Figure 7) shows the appearance of nanoparticles
embedded in an amorphous matrix. The selected area electron diffraction (SAED) analysis of
the nanocomposite (Fig. 7b) shows characteristic diffraction rings/spots characteristic of
elemental Bi. Note that due to their very small size (~5 nm) and the resulting large peak
broadening, we could not identify Bi reflections in the corresponding XRD pattern at 1.0 V
(
Figure S6b) that shows a high background due to the presence of amorphous phases. The
identification of Bi nanoparticles was further confirmed by EDXS. However, we could not
probe the distribution of Li and B, since EDX analysis cannot spot light elements.
Additionally, we observed a strong agglomeration of the Bi nanoparticles (Figure 7c) after
discharge to 1.0 V. Lastly, Bi particles could be also be spotted with SAED in the recharged
3
.5 V sample (Figure S7a), even though the diffraction rings are less pronounced as
compared to the discharge sample indicative of a lower amount of Bi present in the charged
sample. Overall the results obtained from TEM at this stage explain the poor reversibility on
subsequent cycling due to kinetic limitations associated to particles coarsening and
4 2 9
agglomeration of Bi. However, we could not reconfirm the formation of Bi B O since the
collected SAED patterns reveal barely detectable rings besides the one denoted to Bi (Figure
S7b), indicative of the amorphous nature of the recharged phase.
1
1

Physical Chemistry Chemical Physics
Page 12 of 21
View Article Online
DOI: 10.1039/C7CP07693B
Figure 7: (a) TEM image of a sample discharge to 1.0 V. The orange arrow in the inset is
indicating the bright amorphous matrix consisting of Li O/ Li BO . (b) SAED image of the
2
3
3
composite. (c) TEM image and the corresponding elemental maps for Bi, O and C given in
red, blue and green respectively.
In short, we could demonstrate through TEM the partial reversible formation of
elemental Bi upon the electrochemical reduction of Bi
formation of Li O and Li BO as stated in the reaction scheme (equation 2) could not be
evidenced yet due to their most likely nanocrystalline/amorphous nature, similar as
4 2 9
B O versus Li. However the
2
3
3
2
6
observed in the case of transition metal borates. As an attempt to circumvent this issue,
2
7–30
Nuclear Magnetic Resonance (NMR), which is a suitable tool for probing local structures
7
and their evolvement upon electrochemical cycling was used taking advantage of the Li
3
1,32
7
nuclei.
4 2 9
Well-resolved Li NMR measurements were performed for Bi B O samples
discharged to 1 V and 1.7 V prior to recharge each of them to 3.5 V as shown in Figure 8.
Only one sharp resonance locating at 0 ppm is observed for all the chosen electrodes,
indicating that both of the formed products Li
2 3 3
O and Li BO are diamagnetic. The residual
interaction are observed by obvious spinning side bands (SSBs) cover over ~1600 ppm even
1
2

Page 13 of 21
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C7CP07693B
under fast spinning, indicating partially disordered structure and strong dipolar interaction
within these electrodes, in accordance with the amorphous/nanocrystalline nature of the
7
Li
1
2
O/Li
3
BO
3
matrix (Figure 8a, inset). Moreover, by integration of the Li NMR signal for both
7
and 3.5 V samples, we could deduce a Li signal 3 times larger for 3.5 V sample as
compared to the 1 V sample, which is consistent with the corresponding Li content of ~12
and ~4 in the 1.0 and 3.5 V samples, respectively (Figure 3a). Altogether, these observations
confirm the overall reaction scheme proposed in equation 2.
7
Figure 8: (a) Li MAS NMR spectra of Li
x
Bi
4
B
2
O
9
electrodes at different states of charge/ discharge
and (b) the enlargement of the framed region are shown in (a). The asterisks marked in (a)
demonstrate the SSBs. Spectra are calibrated to mass weight of electrodes and number of
scans.
Finally, the existence of two plateaus at 1.8 and 1.4 V in the discharge curve (Figure
) suggests step-wise reactions, which are difficult to fully study because of the
3
amorphous/nanocrystalline nature of the formed products. Nevertheless, previous studies
on 3d-metal borates have indicated the formation of Li
3
BO
3
at redox potentials depending
and α-Bi vs. Li occurs via
2
6
upon the nature of the 3d-metal, while reactivity of δ-Bi
2
O
3
2 3
O
8
,33
conversion reactions at 1.8 and 1.2/1.4V respectively.
Note that this is nearly the same
vs. Li. Although quite
redox potentials we have measured when discharging Bi
4 2 9
B O
1
3

Physical Chemistry Chemical Physics
Page 14 of 21
View Article Online
DOI: 10.1039/C7CP07693B
4 2 9
speculative, it is thus tempting to hypothesize that during the conversion of Bi B O there is
first the conversion of the borate leading to the formation Li
3
3
BO and Bi, and simultaneously
to the growth of the two δ and α Bi
conversion reactions at 1.8 and 1.4 V respectively (accompanied with the formation of Li
and Bi), before all Bi is converted, with the length of the different plateaus
corresponding to the relative ratio of these polymorphs. Although we do not have evidence
for the existence of Bi from electron microscopy at present, our hypothesis of starting
with the formation of Li BO and then Li O seems to be reasonable, and similar observations
2 3
O polymorphs. Both polymorphs will undergo
2
O
4 2 9
B O
2
O
3
3
3
2
have been made for the bismuth oxyfluoride system. This system was shown to react with Li
via a conversion reaction, but the conversion of the fluoride occurs first, followed by the
8
conversion of the as-formed bismuth oxide.
At this stage a legitimate question deals with the poor reversibility of the reduction
process when the cycling spans over the 1 to 3.5 V voltage range and whether it could be
4 2 9
improved by limiting the potential window. To check this point, Bi B O /Li half cells were
cycled between 1.7 and 3.5 V (Fig. 9a). The voltage-composition curve shows, aside the
initial step voltage previously discussed, a discharge plateau at 1.8 V corresponding to the
+
uptake of 8 Li . On the subsequent charge, which shows an S-type voltage profile, only 5 out
+
of the 8 uptake Li ions can be released, hence leading to an irreversible capacity of ~35 %.
Note that the following discharge does not follow the first one and this is indicative of a
conversion reaction. Once the first cycle is achieved, the average potential of the cell is
about 2.3 V with a sustained reversible capacity of ~140 mAh/g for several cycles prior the
appearance of a noticeable fading (Figure 9b). This strongly improved capacity retention by
limiting the voltage window from 1.0 – 3.5 V to 1.7 – 3.5 V, does not come as a total surprise
as it is known that by increasing the discharge cutoff potential Bi agglomeration is minimized
3
4
(
4 2 9
Figure S8a versus Figure S8b, and Figure 7). Moreover by cycling Bi B O over a high
potential window, parasitic side reactions such as reductive electrolyte decomposition are
not as pronounced as for potentials < 1.7 V. This continuous decomposition most likely leads
to the formation of an insulating organic layer, propagating particle separation/ coarsening
and thus deteriorating charge transport throughout the composite leading to an increase in
3
5,36
7
voltage polarization upon cycling (Figure 10a).
electrode samples discharged 1.7 and recharged to 3.5 V (Figure 8) reveal the existence of
Li O and Li BO . Again the signal integration for the discharged sample (1.7 V) is about 3
Interestingly, Li NMR measurements for
2
3
3
1
4

Page 15 of 21
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C7CP07693B
times greater than for the charged one (3.5 V), consistent with the Li content deduced from
the voltage composition curve (Figure 9a).
Figure 9: (a) Voltage-composition curve of Bi
4 2 9
B O versus Li cycled at a C/10 rate between
1
.7 and 3.5 V and (b) the corresponding capacity retention.
Two other important experimental findings are worth mentioning regarding the
4 2 9
cycling performances of Bi /Li cells. They are i) the ability to reversibly cycle Bi B O
4
B
2
O
9
powders made of macroscopic particles with only 5 wt.% carbon additive and ii) the small
cycling hysteresis around 300 mV, while usually conversion reaction electrodes require at
least 20% of carbon and show voltage polarization greatly larger. The origin of such a
difference is most likely rooted in the presence of a ~3 nm thick amorphous surface layer
4 2 9
onto the Bi B O particles, as deduced by HRTEM (Figure 6a), which is mainly made up of
amorphous bismuth oxide (Figure 6b, c). Since it is well known that non-stoichiometric metal
3
7–39
oxides, namely Bi-oxides can promote electronic conductivity,
we can hypothesize that
this layer is responsible for enhanced charge transfer kinetics, enabling electronic
conducting, hence accounting for the use of solely 5 wt.% of conductive carbon additive. This
4 2 9
hypothesis is supported by the fact that Bi B O as synthesized could be electrochemically
cycled without any conducting additive between 1.0 and 3.5 V (Figure S10), although the
reversibility is very poor once the material is electrochemically grinded, due to the loss of
electronic percolation throughout the formed composite composed of insulating Li
Li BO phases. The positive effect of this amorphous surface layer on the electrochemical
activity of Bi is raising questions about its origin and how to control its formation.
2
O and
3
3
4 2 9
B O
1
5

Physical Chemistry Chemical Physics
Page 16 of 21
View Article Online
DOI: 10.1039/C7CP07693B
Boron species (in particular boron oxide B
2
O
3
) are well known to be volatile at elevated
4
0–42
temperatures,
hence the feasibility to form a boron deficient surface layer during the
prolonged annealing treatment.
4 2 9
Turning the relative small voltage polarization observed for Bi B O , it is of the same
amplitude as displayed by other borates which react towards Li via a conversion reaction.
2
6,43
Although the origin of such polarization is not fully understood, theoretical studies and
4
4
3 4
recent experimental work on the conversion reaction of Co O with Na, suggest that it is
associated to different thermodynamic paths for charge and discharge that are correlated to
the difference in ionic mobilities between Li and displaced ionic species (herein B, O,
2
3,44,45
Bi).
Borate species may thus play a crucial role as network former leading to a
continuous amorphous matrix promoting enhanced ionic conductivity as observed for a wide
4
6–49
range of borate glasses.
Conclusion
We have reported that Bi B O reversibly reacts versus Li via a conversion-type
4
2 9
reaction leading to the formation of Bi nanoparticles surrounded by an amorphous matrix
7
composed of Li O and Li BO , as deduced by combined XRD, TEM imaging and Li NMR
2
3
3
spectroscopy. We found this process to be reversible by limiting the cycling voltage range to
1
.7 – 3.5 V so that capacities of 140 mAh/g can be achieved at an average potential of 2.3 V.
3
Moreover, bearing in mind that the density of Bi B O is large (8.17 g/cm ), we can reach a
4
2 9
volumetric energy density of 2819 Wh/L. Remarkably such values can be achieved with both
a small polarization (300 mV) and the addition of solely 5 wt.% carbon, which is quite
unusual for conversion type cathode materials. Such a finding is rationalized through the
appearance of an amorphous nanometer thick Bi-oxide rich surface layer during the
4 2 9
synthesis of Bi B O which in particular enhance electronic transport. Having point out the
importance of surface morphology and benefit of the borate species towards the
electrochemical properties of conversion type electrode materials, it now pertains to see if
this concept can be applied to other borate based conversion materials.
1
6

Page 17 of 21
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C7CP07693B
Author information
Corresponding Author
*
E-mail: jean-marie.tarascon@college-de-france.fr
Notes
The authors declare no competing financial interest.
Present addresses
#
Battery and Electrochemistry Laboratory (BELLA), Institute of Nanotechnology (INT),
Karlsruhe Institute of Technology (KIT), Hermann-von-Helmholtz-Platz 1, D-76344
Eggenstein-Leopoldshafen, Germany
+
National High Magnetic Field Laboratory, 1800 East Paul Dirac Drive, Tallahassee, Florida
3
2310, United States
Acknowledgement
F.S. acknowledges ALISTORE-ERI for his PhD grant. F.S. and R.D. acknowledge the financial
support from the Slovenian Research Agency (research core funding No. P2-0393).
References
(
(
1)
2)
Armand, M.; Tarascon, J.-M. Building Better Batteries. Nature 2008, 451 (7179), 652–657.
Sathiya, M.; Rousse, G.; Ramesha, K.; Laisa, C. P.; Vezin, H.; Sougrati, M. T.; Doublet, M.-L.;
Foix, D.; Gonbeau, D.; Walker, W.; Prakash, A. S.; Ben Hassine, M.; Dupont, L.; Tarascon, J.-M.
Reversible Anionic Redox Chemistry in High-Capacity Layered-Oxide Electrodes. Nat. Mater.
2
013, 12 (9), 827–835.
(3)
Pearce, P. E.; Perez, A. J.; Rousse, G.; Saubanère, M.; Batuk, D.; Foix, D.; McCalla, E.;
Abakumov, A. M.; Van Tendeloo, G.; Doublet, M.-L.; Tarascon, J.-M. Evidence for Anionic
Redox Activity in a Tridimensional-Ordered Li-Rich Positive Electrode β-Li2IrO3. Nat. Mater.
2
017, 16 (5), 580–586.
(4)
(5)
(6)
(7)
Besenhard Jürgen O. Chalkogenide Des As, Sb Und Bi Als Positive Elektroden in Lithium-
Batterien / Chalcogenides of As, Sb and Bi as Positive Electrodes in Lithium Cells. Z. Für
Naturforschung B 2014, 33 (3), 279.
Poizot, P.; Laruelle, S.; Grugeon, S.; Dupont, L.; Tarascon, J.-M. Nano-Sized Transition-Metal
Oxides as Negative-Electrode Materials for Lithium-Ion Batteries. Nature 2000, 407 (6803),
4
96–499.
Cabana, J.; Monconduit, L.; Larcher, D.; Palacín, M. R. Beyond Intercalation-Based Li-Ion
Batteries: The State of the Art and Challenges of Electrode Materials Reacting Through
Conversion Reactions. Adv. Mater. 2010, 22 (35), E170–E192.
Wu, F.; Yushin, G. Conversion Cathodes for Rechargeable Lithium and Lithium-Ion Batteries.
Energy Env. Sci 2017.
1
7

Physical Chemistry Chemical Physics
Page 18 of 21
View Article Online
DOI: 10.1039/C7CP07693B
(
8)
9)
Bervas, M.; Klein, L. C.; Amatucci, G. G. Reversible Conversion Reactions with Lithium in
Bismuth Oxyfluoride Nanocomposites. J. Electrochem. Soc. 2006, 153 (1), A159.
Wiaderek, K. M.; Borkiewicz, O. J.; Castillo-Martínez, E.; Robert, R.; Pereira, N.; Amatucci, G.
G.; Grey, C. P.; Chupas, P. J.; Chapman, K. W. Comprehensive Insights into the Structural and
Chemical Changes in Mixed-Anion FeOF Electrodes by Using Operando PDF and NMR
Spectroscopy. J. Am. Chem. Soc. 2013, 135 (10), 4070–4078.
(
(
10) Kim, S.-W.; Pereira, N.; Chernova, N. A.; Omenya, F.; Gao, P.; Whittingham, M. S.; Amatucci, G.
G.; Su, D.; Wang, F. Structure Stabilization by Mixed Anions in Oxyfluoride Cathodes for High-
Energy Lithium Batteries. ACS Nano 2015, 9 (10), 10076–10084.
(11) Chevrier, V. L.; Hautier, G.; Ong, S. P.; Doe, R. E.; Ceder, G. First-Principles Study of Iron
Oxyfluorides and Lithiation of FeOF. Phys. Rev. B 2013, 87 (9), 094118.
(
12) Pereira, N.; Badway, F.; Wartelsky, M.; Gunn, S.; Amatucci, G. G. Iron Oxyfluorides as High
Capacity Cathode Materials for Lithium Batteries. J. Electrochem. Soc. 2009, 156 (6), A407–
A416.
(13) Taberna, P. L.; Mitra, S.; Poizot, P.; Simon, P.; Tarascon, J.-M. High Rate Capabilities Fe3O4-
Based Cu Nano-Architectured Electrodes for Lithium-Ion Battery Applications. Nat. Mater.
2
006, 5 (7), 567–573.
(14) Hyman, A.; Perloff, A. The Crystal Structure of Bismuth (2: 1) Borate, 2Bi2O3. B2O3. Acta
Crystallogr. B 1972, 28 (7), 2007–2011.
(
15) Huang, H.; He, Y.; Lin, Z.; Kang, L.; Zhang, Y. Two Novel Bi-Based Borate Photocatalysts: Crystal
Structure, Electronic Structure, Photoelectrochemical Properties, and Photocatalytic Activity
under Simulated Solar Light Irradiation. J. Phys. Chem. C 2013, 117 (44), 22986–22994.
16) Leriche, J. B.; Hamelet, S.; Shu, J.; Morcrette, M.; Masquelier, C.; Ouvrard, G.; Zerrouki, M.;
Soudan, P.; Belin, S.; Elkaïm, E.; Baudelet, F. An Electrochemical Cell for Operando Study of
Lithium Batteries Using Synchrotron Radiation. J. Electrochem. Soc. 2010, 157 (5), A606.
17) Rietveld, H. A Profile Refinement Method for Nuclear and Magnetic Structures. J. Appl.
Crystallogr. 1969, 2 (2), 65–71.
(
(
(
(
(
18) Rodríguez-Carvajal, J. Recent Advances in Magnetic Structure Determination by Neutron
Powder Diffraction. Phys. B Condens. Matter 1993, 192 (1), 55–69.
19) Filatov, S. K.; Shepelev, Y. F.; Aleksandrova, Y. V.; Bubnova, R. S. Structure of Bismuth
Oxoborate Bi4B2O9 at 20, 200, and 450°C. Russ. J. Inorg. Chem. 2007, 52 (1), 21–28.
20) Gillot, F.; Boyanov, S.; Dupont, L.; Doublet, M.-L.; Morcrette, M.; Monconduit, L.; Tarascon, J.-
M. Electrochemical Reactivity and Design of NiP2 Negative Electrodes for Secondary Li-Ion
Batteries. Chem. Mater. 2005, 17 (25), 6327–6337.
(21) Bezza, I.; Kaus, M.; Riekehr, L.; Pfaffmann, L.; Doyle, S.; Indris, S.; Ehrenberg, H.; Solhy, A.;
Saadoune, I. Electrochemical Lithiation/Delithiation of SnP O Observed by in Situ XRD and
2
7
7
31
119
Ex Situ Li/ P NMR, and Sn Mössbauer Spectroscopy. Phys Chem Chem Phys 2016, 18
15), 10375–10382.
(
(22) Hwang, H.; Kim, M. G.; Kim, Y.; Martin, S. W.; Cho, J. The Electrochemical Lithium Reactions of
Monoclinic ZnP2 Material. J. Mater. Chem. 2007, 17 (30), 3161.
(
23) Doe, R. E.; Persson, K. A.; Meng, Y. S.; Ceder, G. First-Principles Investigation of the Li−Fe−F
Phase Diagram and Equilibrium and Nonequilibrium Conversion Reactions of Iron Fluorides
with Lithium. Chem. Mater. 2008, 20 (16), 5274–5283.
(
24) Patoux, S.; Vannier, R.-N.; Mairesse, G.; Nowogrocki, G.; Tarascon, J.-M. Lithium- and Proton-
Driven Redox Reactions in BIMEVOX-Type Phases. Chem. Mater. 2001, 13 (2), 500–507.
25) Arroyo y de Dompablo, M. E.; García-Alvarado, F.; Morán, E. Bi4V2O11 and Related
Compounds as Positive Electrode Materials for Lithium Rechargeable Batteries. Solid State Ion.
(
1
996, 91 (3), 273–278.
(
26) Débart, A.; Revel, B.; Dupont, L.; Montagne, L.; Leriche, J.-B.; Touboul, M.; Tarascon, J.-M.
Study of the Reactivity Mechanism of M3B2O6 (with M= Co, Ni, and Cu) toward Lithium.
Chem. Mater. 2003, 15 (19), 3683–3691.
1
8

Page 19 of 21
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C7CP07693B
(27) Grey, C. P.; Dupré, N. NMR Studies of Cathode Materials for Lithium-Ion Rechargeable
Batteries. Chem. Rev. 2004, 104 (10), 4493–4512.
(28) Jiang, M.; Key, B.; Meng, Y. S.; Grey, C. P. Electrochemical and Structural Study of the Layered,
“Li-Excess” Lithium-Ion Battery Electrode Material Li[Li1/9Ni1/3Mn5/9]O2. Chem. Mater.
2
009, 21 (13), 2733–2745.
(
29) Rosina, K. J.; Jiang, M.; Zeng, D.; Salager, E.; Best, A. S.; Grey, C. P. Structure of Aluminum
Fluoride Coated Li[Li1/9Ni1/3Mn5/9]O2 Cathodes for Secondary Lithium-Ion Batteries. J.
Mater. Chem. 2012, 22 (38), 20602–20610.
(
30) Trease, N. M.; Seymour, I. D.; Radin, M. D.; Liu, H.; Liu, H.; Hy, S.; Chernova, N.; Parikh, P.;
Devaraj, A.; Wiaderek, K. M.; Chupas, P. J.; Chapman, K. W.; Whittingham, M. S.; Meng, Y. S.;
Van der Van, A.; Grey, C. P. Identifying the Distribution of Al3+ in LiNi0.8Co0.15Al0.05O2.
Chem. Mater. 2016, 28 (22), 8170–8180.
(
31) Key, B.; Bhattacharyya, R.; Morcrette, M.; Seznéc, V.; Tarascon, J.-M.; Grey, C. P. Real-Time
NMR Investigations of Structural Changes in Silicon Electrodes for Lithium-Ion Batteries. J. Am.
Chem. Soc. 2009, 131 (26), 9239–9249.
(
(
(
32) Grey, C. P.; Tarascon, J. M. Sustainability and in Situ Monitoring in Battery Development. Nat.
Mater. 2017, 16 (1), 45–56.
33) Li, Y.; Trujillo, M. A.; Fu, E.; Patterson, B.; Fei, L.; Xu, Y.; Deng, S.; Smirnov, S.; Luo, H. Bismuth
Oxide: A New Lithium-Ion Battery Anode. J. Mater. Chem. A 2013, 1 (39), 12123.
34) Park, C.-M.; Yoon, S.; Lee, S.-I.; Sohn, H.-J. Enhanced Electrochemical Properties of
Nanostructured Bismuth-Based Composites for Rechargeable Lithium Batteries. J. Power
Sources 2009, 186 (1), 206–210.
(35) Chagnes, A.; Swiatowska, J. Electrolyte and Solid-Electrolyte Interphase Layer in Lithium-Ion
Batteries. In Lithium Ion Batteries-New Developments; InTech, 2012.
(
36) Renman, V.; Hu, S.; Eriksson, R.; Maibach, J.; Johnsson, M.; Pay Gómez, C.; Edström, K. Ni Sb 4
3
6 6
O F and Its Electrochemical Behavior toward Lithium—A Combination of Conversion and
Alloying Reactions. Chem. Mater. 2016, 28 (18), 6520–6527.
(37) Sammes, N. M.; Tompsett, G. A.; Näfe, H.; Aldinger, F. Bismuth Based Oxide Electrolytes—
Structure and Ionic Conductivity. J. Eur. Ceram. Soc. 1999, 19 (10), 1801–1826.
(
(
38) Skinner, S. J.; Kilner, J. A. Oxygen Ion Conductors. Mater. Today 2003, 6 (3), 30–37.
39) Kitada, A.; Hasegawa, G.; Kobayashi, Y.; Kanamori, K.; Nakanishi, K.; Kageyama, H. Selective
Preparation of Macroporous Monoliths of Conductive Titanium Oxides TinO2n–1 (N = 2, 3, 4,
6
). J. Am. Chem. Soc. 2012, 134 (26), 10894–10898.
(
(
(
(
(
40) Forni, L.; Fornasari, G.; Tosi, C.; Trifirò, F.; Vaccari, A.; Dumeignil, F.; Grimblot, J. Non-
Conventional Sol–gel Synthesis for the Production of Boron-Alumina Catalyst Applied to the
Vapour Phase Beckmann Rearrangement. Appl. Catal. Gen. 2003, 248 (1–2), 47–57.
41) Chen, K.; Ai, N.; Lievens, C.; Love, J.; Jiang, S. P. Impact of Volatile Boron Species on the
Microstructure and Performance of Nano-Structured (Gd,Ce)O2 Infiltrated (La,Sr)MnO3
Cathodes of Solid Oxide Fuel Cells. Electrochem. Commun. 2012, 23, 129–132.
42) Chen, K.; Hyodo, J.; O’Donnell, K. M.; Rickard, W.; Ishihara, T.; Jiang, S. P. Effect of Volatile
Boron Species on the Electrocatalytic Activity of Cathodes of Solid Oxide Fuel Cells III.
Ba0.5Sr0.5Co0.8Fe0.2O3-δ Electrodes. J. Electrochem. Soc. 2014, 161 (12), F1163–F1170.
43) Ibarra-Palos, A.; Darie, C.; Proux, O.; Hazemann, J. L.; Aldon, L.; Jumas, J. C.; Morcrette, M.;
Strobel, P. Electrochemical Reactions of Iron Borates with Lithium: Electrochemical and in Situ
Mössbauer and X-Ray Absorption Studies. Chem. Mater. 2002, 14 (3), 1166–1173.
44) Kim, H.; Kim, H.; Kim, H.; Kim, J.; Yoon, G.; Lim, K.; Yoon, W.-S.; Kang, K. Understanding Origin
of Voltage Hysteresis in Conversion Reaction for Na Rechargeable Batteries: The Case of
Cobalt Oxides. Adv. Funct. Mater. 2016, 26 (28), 5042–5050.
(
45) Yu, H.-C.; Ling, C.; Bhattacharya, J.; Thomas, J. C.; Thornton, K.; Van der Ven, A. Designing the
next Generation High Capacity Battery Electrodes. Energy Environ. Sci. 2014, 7 (5), 1760.
46) Kamitsos, E. I.; Chryssikos, G. D. Borate Glass Structure by Raman and Infrared Spectroscopies.
J. Mol. Struct. 1991, 247, 1–16.
(
1
9

Physical Chemistry Chemical Physics
Page 20 of 21
View Article Online
DOI: 10.1039/C7CP07693B
(
(
(
47) Konijnendijk, W. L.; Stevels, J. The Structure of Borate Glasses Studied by Raman Scattering. J.
Non-Cryst. Solids 1975, 18 (3), 307–331.
48) Tuller, H. L.; Button, D. P.; Uhlmann, D. R. Fast Ion Transport in Oxide Glasses. J. Non-Cryst.
Solids 1980, 40 (1–3), 93–118.
49) SClIUTT, H.-J. Wolfgang MULLER, Detlef KRUSCHKE, Manfred TORGE, Arnd-Rtidiger GRIMMER.
Solid State Ion. 1987, 23, 53–58.
2
0

Page 21 of 21
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C7CP07693B
TOC graphic
Galvanostatic charge-discharge curve for a Bi B O /C macrocomposite, highlighting the small
4
2
9
voltage hysteresis for the conversion reaction around 300 mV.