Achieving highly stable Li-O2 battery operation by designing a carbon nitride-based cathode towards a stable reaction interface

Article abstract of DOI:10.1039/c7ta05009g

Aprotic Li-O₂ batteries have attracted a great deal of attention because of their potential to offer much higher energy density than those provided by commercialized lithium-ion batteries. However, their reversible operation is plagued by serious side-reactions from liquid electrolytes and/or carbon-based materials. Recently, carbon-free materials have been proposed and utilized to construct stable cathodes for Li/O₂ chemistry. Different from most of the previously reported metal-based carbon-free cathodes, herein we report a non-metal-based carbon-free cathode support consisting of mesoporous boron-doped carbon nitride (m-BCN) and demonstrate its excellent stability and activity for Li/O₂ chemistry. Benefiting from the introduction of evenly distributed RuO₂ nanoparticles (1-2 nm) in the pores of the boron-doped carbon nitride support, excellent cycle stability with a low overpotential (141 cycles with a pristine Li anode which is extended to 227 cycles after replacing it with a new Li anode at 0.5 mA cm^-2) and superior rate capability (1.28 mA h cm^-2 at 1 mA cm^-2) are obtained. This impressive performance is ascribed to the enhanced stability and activity of such designed cathodes, which is supported by the fact that reversible formation and decomposition of Li₂O₂ with no accumulation of Li₂CO₃ is detected during cycling. These results demonstrate that manipulating cathode materials towards stable reaction interfaces is essential for alleviating the formation of by-products and improving the performance of Li-O₂ batteries.

Full text of DOI:10.1039/c7ta05009g

View Article Online
View Journal
Journal of
M at erials Chemistry A
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: P. Lou, Z. Cui and
X. Guo, J. Mater. Chem. A, 2017, DOI: 10.1039/C7TA05009G.
Volume
4
Number
1
7
January 2016 Pages 1–330
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Journal of
Materials Chemistry A
Accepted Manuscripts are published online shortly after
Materials for energy and sustainability
www.rsc.org/MaterialsA
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 2050-7488
PAPER
Kun Chang, Zhaorong Chang et al.
Bubble-template-assisted synthesis of hollow fullerene-like
MoS nanocages as a lithium ion battery anode material
2
rsc.li/materials-a

Page 1 of 7
Jo uP rl ne aa sl eo fd oM na ot et r ai ad l jsu sCt hme am r gi si nt rs y A
View Article Online
DOI: 10.1039/C7TA05009G
Journal Name
ARTICLE
Achieving Highly Stable Li-O Battery Operations by Designing
2
Carbon Nitride-Based Cathode towards Stable Reaction Interface
Peili Lou,a,b Zhonghui Cuia,* and Xiangxin Guoa,*
Received 00th January 20xx,
Accepted 00th January 20xx
The aprotic Li-O batteries have attracted a great deal of attention for their potential to offer much higher energy density
2
DOI: 10.1039/x0xx00000x
than that provided by commercialized lithium-ion batteries. However, their reversible operation is plagued by serious side-
reactions from liquid electrolytes and/or carbon-based materials. Recently, carbon-free materials have been proposed and
www.rsc.org/
attempted to construct stable cathode for Li/O
carbon-free cathodes, herein we report a non-metal-based carbon-free cathode support consisting of mesoporous boron-
doped carbon nitride (m-BCN), and demonstrate its excellent stability and activity upon Li/O chemistry. Benefited from the
introduction of evenly distributed RuO nanoparticles (1-2 nm) in the pore of the boron-doped carbon nitride support,
excellent cycle stability with low overpotential (141 cycles with pristine Li anode and extended to 227 cycles after replacing
2
chemistry. Different with most of the previously reported metal-based
2
2
-
2
-2
-2
a new Li anode at 0.5 mA cm ) and rate capability (1.28 mAh cm at 1 mA cm ) are obtained. This impressive performance
is ascribed to the enhanced stability and activity of such designed cathodes, which is supported by the fact that the reversible
formation and decomposition of Li
that manipulating cathode materials towards stable reaction interfaces is essential for alleviating the formation of by-
products and improving the performance of Li-O batteries.
2
O
2
but no accumulation of Li
CO
2 3
is detected during cycling. These results demonstrate
2
In regard to the cathode, carbon-free materials have been
Introduction
In recent years, identifying new energy storage systems that
could enable higher energy density than that of currently
commercialized lithium-ion batteries (LIBs) has been becoming
the focus of battery research. Accidentally introducing some
oxygen into LIB brings incredible increase in discharge capacity,
proposed to offer stable reaction interfaces for Li-O
2
12-26
12, 13
Ru,14, 15
chemistry.
For example, noble metals (e.g., Au,
16
17
2 4
MnCo O
,18
RuO
Ti
2
), transition metal oxides (e.g., Co
) and carbides (e.g., TiC ), composites of noble metals
3
4
O ,
1
9
20
4
O
7
2
1
and transition metal compounds (e.g., RuO
2
/Mn
RuO @TiN
), as well as a few non-metal compounds (e.g.,
C ) have been attempted in recent years. For these reported
2
O
3
,
TiO
2
-
1
22
23
24
nanowires@porous-RuO
2
,
Ru/ITO,
2
and
25
Au@MnO
2
leading to the discovery of the aprotic lithium-oxygen (Li-O
2
)
26
B
4
2
battery, which is able to deliver several times higher energy
density than that of state-of-the-art LIBs.2-5 The rechargeable Li-
carbon-free cathodes (hosts), one can find that almost all of
them are metal-based materials, which are of high densities and
will incur compromise to the energy density of their resulted Li-
O batteries. Comparatively, non-metal compounds consisting
2
of elemental C, B and N are indeed with low densities; smart
designs based on these materials are promising to show good
O
+
2
battery theoretically works based on a simple reaction of 2Li
= Li , with Li as the reversible mediator formed on
O
2
2
O
2
2 2
O
2-6
discharge and decomposed on recharge. However, their
reversible operation is plagued by serious side-reactions from
organic electrolytes and/or carbon-based cathode materials.5-8
For example, the decomposition of organic electrolytes and
carbon-based cathode materials caused by oxygen radicals like
3 4
battery performance. Graphitic carbon nitride (g-C N ) is a
typical one of non-metal compounds, which has attracted
extensive attentions due to its unexpected catalytic activity for
−
2−
푂
and
푂
incurs the formation of carbonates, LiOH and other
2
2
27
a variety of reaction. With this material, a free-standing
oxygen cathode consisting of mesoporous graphene@g-C
was developed, showing very large discharge capacity and
irreversible species, leading to high overpotential, low energy
efficiency and poor cyclability. This indicates that the stability of
cell components is of prime importance, motivating efforts to
3 4
N
28
−
2
2− 9-11
enhanced cyclability. Another similar design was reported by
Yi et al. who composite the g-C with carbon paper and
demonstrate an improved battery performance. Besides g-
, a sole B N based cathode was designed for Li-O battery,
search for stable materials that are inert to the 푂 /푂
.
2
3 4
N
2
9
C
3
N
4
3
2
a.State Key Laboratory of High Performance Ceramics and Superfine
26
Microstructure, Shanghai Institute of Ceramics, Chinese Academy of Sciences,
Shanghai, 200050 China.
University of Chinese Academy of Sciences, Beijing, 100039 China.
Corresponding author: cuizhonghui@mail.sic.ac.cn (Z. Cui) or
which also shown enhanced cyclability. These reports imply
that such non-metal based carbon-free materials have better
stability against oxygen reduction/evaluation reactions
b.
*
xxguo@mail.sic.ac.cn (X. Guo)
Electronic Supplementary Information (ESI) available: Experimental details and
Figure S1-S7 are given. See DOI: 10.1039/x0xx00000x
(
ORR/OER) than carbon materials, mainly due to the absence of
†
fragile C-C bond. Nevertheless, the available reports concerning
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Jo uP rl ne aa sl eo fd oM na ot et r ai ad l jsu sCt hme am r gi si nt rs y A
Page 2 of 7
ARTICLE
Journal Name
such non-metal carbon-free materials appear to be numbered.
Searching for new non-metal based carbon-free materials,
optimizing their structure-performance and exploring their full
advantage and potential are still open for further studies.
Results and Discussion
Synthesis and Characterization of the RuO
View Article Online
DOI: 10.1039/C7TA05009G
2
@m-BCN.
The RuO @m-BCN was synthesized via two facile steps. Firstly,
2
mesoporous boron-doped carbon nitrides (m-BCN) as stable
cathode supports were prepared by a thermal-induced
condensation of the mixture of dicyanodiamide and
tetrafluoroborate-based
tetrafluoroborate, BmimBF
(1-Butyl-3-methylimidazolium
) ionic liquid (IL).33 Herein, the
4
4
BmimBF is employed not only as the soft template to construct
porous-structured carbon nitride, but also as the source of
elemental boron for heteroatom-doping the carbon nitride
Figure 1. The key features of a desired cathode for Li-O
batteries.
2
matrix. Secondly, the decoration of the m-BCN with RuO
nanoparticles was completed via a simple solution process, in
2
3+
which the positive charged Ru ions were first immobilized
onto the surface of the m-BCN supports for the electrostatic
Different with the well-known LIBs, in which the cathodes are
active materials, the cathodes for Li-O batteries are the places
for Li/O reaction, the hosts for reaction products, the mediums
for electron transfer, and the catalysts or promoters for Li/O
reaction. Thus, several basic factors should be taken into
account when searching for a suitable cathode for Li-O battery
Figure 1): (1) Mass (Density) The cathodes for Li-O batteries
are not the active materials but rather the hosts for the reaction
products of Li/O chemistry, thus lighter materials providing
higher specific energy densities. (2) Porosity The products of Li-
batteries whether the desired Li or unwanted by-
2
interaction, and then in situ transformed into RuO
nanoparticles and embedded on the surface of m-BCN supports
when introducing the NaBH into the solution.
2
2
2
4
2
(
2
2
O
2
2 2
O
products like carbonates are solid and need to be stored within
the pore of the cathodes, thus the cathode with higher porosity
+
providing higher capacity if the supply of O
2
and Li can be
maintained. (3) Electronic conductivity As the sole medium to
allow favourable electron transport from external circuit to
Li/O
conductivity promise more efficient and immediate supply of
electrons and fast kinetics of Li/O reaction. (4) Stability and
catalytic activity upon ORR/OER As the reaction sites of Li/O
2
reaction sites, the cathodes with better electronic
2
2
chemistry, the cathodes with enhanced stability and catalytic
activity upon ORR/OER are of strong promise to diminish the
formation of by-products and promote the formation and
Figure 2. Structure and morphology of the as-prepared
RuO @m-BCN. (a) SEM, (b) XRD, (c) B 1s, (d-e) TEM, (f) HRTEM,
g) STEM, and (h-k) corresponding EDS mapping of C, N, Ru and
O, respectively. The scale bars in Figure 2g-k are equivalent to
00 nm.
2
decomposition of Li
Keeping these factors in mind, in this work, we demonstrate a
ruthenium oxide (RuO ) nanoparticles decorated mesoporous
boron-doped carbon nitride (RuO @m-BCN) composite as a
stable carbon-free cathode for Li-O
2
battery. In this material, The structure and morphology of as-prepared RuO @m-BCN
2 2
O , leading to better battery performance.
(
2
1
2
2
firstly, the labile “carbon” is avoided, thus promising to negate composites were investigated by X-ray diffraction (XRD),
the formation of carbon-decomposition induced carbonates scanning electron microscopy (SEM), transmission electron
and the carbon catalytic effect on the electrolyte decomposition microscopy (TEM) and X-ray photoelectron spectroscopy (XPS).
forming extra carbonates. Secondly, the evenly distributed The SEM image indicates that the as-prepared RuO
RuO
nanoparticles (1-2 nm) provide abundant active sites for consists of porous and non-uniform agglomerates (Figure 2a).
ORR/OER, leading to greatly enhanced catalytic effect.
Thirdly, the evenly distributed RuO
nanoparticles significantly transmission and storing of solid reaction products. Figure 2b
improve the electronic conductivity of RuO
@m-BCN shows the XRD patterns of the as-prepared m-BCN and
composites, guaranteeing fast electron supply. Fourthly, more RuO @m-BCN. The m-BCN supports show similar pattern to
molecules can be absorbed onto the electrode surface for that of the standard graphitic carbon nitride (g-C , JCPDS No.
the existence of oxygen vacancy in the RuO
lattices, which 87-1526) with two characteristic peaks arising at 13.0 and 27.4
provides fast oxygen supply for ORR and contributes to a good degree, which can be indexed as the 100 peak characteristic for
2
@m-BCN
2
30-32
The porous structure is of key importance for oxygen
2
2
2
O
2
3 4
N
2
rate performance. When subjected this material to Li-O
2
the interplanar separation and the 002 peak representing for
batteries, improved cyclability and rate performance with high the interlayer stacking of aromatic systems, respectively.3
3, 34
energy efficiency are characterized.
3 2
After decorated in the RuCl solution, RuO (Rutile, JCPDS No.
2
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 7
Jo uP rl ne aa sl eo fd oM na ot et r ai ad l jsu sCt hme am r gi si nt rs y A
Journal Name
ARTICLE
4
RuO
0-1290) -like peaks appear, as shown in Figure 1a, indicating introduction of evenly distributed RuO
2
nanoparticles. The
View Article Online
DOI: 10.1039/C7TA05009G
2 2
nanoparticles being successfully embedded onto the m- conductance value of RuO @m-BCN and m-BCN can be
BCN supports. One can see clearly that the peaks of the as- determined from the slope of I-V plots. Then the conductivity of
synthesized RuO
2
slightly shift to the lower angles, which is the RuO
2
@m-BCN and m-BCN are calculated to be 11.7 and 5.8
-7
-1
ascribed to their non-stoichiometry feature. In order to × 10 S cm , respectively. This value is much lower than that of
4
-1 37
determine the Ru/O ratio and the state of elemental doping, the bulk RuO
XPS measurements were performed. The Ru/O ratio of the nanosheets (~ 10 S cm ) , owing to the existence of the
RuO @m-BCN is about 1.82, which is smaller than that of the insulator component of m-BCN and voids between every
perfect RuO crystal. This small Ru/O ratio indicates the RuO @m-BCN particles. However, this value is comparable with
existence of oxygen vacancies in RuO lattices, which favours that of previously reported cathodes like Co
2 2
(~ 10 S cm ) and two-dimensional RuO
3
-1 16
2
2
2
2
3
O
4
nanofibers
the oxygen transport on the electrode surface.24 The non- immobilized on nonoxidized graphene nanoflakes
stoichiometry of the RuO is further supported by the energy metallic Magnéli phase Ti
dispersive spectroscopy (EDS) results (Figure S1). cyclability and large capacities. This means that the electrical
A peak centred at 192.4 eV is detected from the as-prepared conductivity of the RuO @m-BCN is sufficient to afford fast
RuO @m-BCN (Figure 2c), which is consistent well with the electron transport for the formation and decomposition of
binding energy of the C-NB group,3
boron being successfully incorporated into the g-C
The content of boron is determined to be 4.34 mol %. B enters Batteries
the C lattices by replacing the C-atoms forming a π-bonded, The electrochemical performance of the RuO
planar layered configuration and acts as electron acceptor composites was examined by Swagelok-type Li-O
because it has three valence electrons, inducing surface states using 0.5 M LiClO
near the valence band top and hence modifying the electronic electrolyte. The use of LiClO
38
and
2
4 7
O
19, both of which provide good
2
2
5, 36
indicating the element desired Li
2 2
O .
3
N
4
lattices. Electrochemical Performance Characterization in Li-O
2
.
3
N
4
2
@m-BCN
batteries
2
4
dissolved in dimethyl sulfoxide (DMSO) as the
/DMSO-based electrolytes is
The absence of distinct peak in the range because their excellent stability upon ORR/OER, guaranteeing
4
3
5, 36
structure of C
of 680 to 696 eV reveals that no element F enters the C
lattices (Figure S2). The mesoporous structure of the RuO @m- shows the initial discharge-charge profile of the RuO
BCN and uniform decoration of the RuO nanoparticles with a and m-BCN at a current density of 0.2 mA cm . Note that there
3 4
N .
12
3
N
4
the formation of Li
2
O
2
even for hundreds of cycles. Figure 3
a
2
2
@m-BCN
-2
2
size of about 1-2 nm can be clearly observed from the TEM are two testing protocols present for evaluating electrode
images with different magnifications (Figure 2d-f). The porous performance: the voltage-limited and capacity-limited method.
2
-1
RuO
a pore size distribution centred around 75 nm in the range of favours the exploration of the full potential of a new electrode.
1-125 nm, which is obtained from the nitrogen absorption- In addition, the discharge cut-off was elaborately set at 2.5 V to
desorption isotherms (Figure S3). The substantial mesopores avoid the lithium intercalation into the RuO
lattices.39 As
and macropores within the RuO @m-BCN composites favour shown in Figure 3a, for the bare m-BCN, only a small capacity is
rapid transport of the electrolyte and O , thus leading to an delivered during initial discharge, while a large capacity with a
enhanced rate capability of the Li–O batteries. The crystalline voltage plateau at 4.44 V is obtained during recharge. It is
nature of the RuO nanoparticles is confirmed by the noteworthy that the RuO @m-BCN/LiClO -DMSO has excellent
2
@m-BCN offers a specific area of 136.5 m g RuO
2
@m-BCN and Herein the voltage-limited protocol is adopted, because it
1
2
2
2
2
2
2
4
appearance of identifiable lattice fringes in HRTEM (Figure 2f electrochemical stability in a wide voltage window of 2.5- 4.5 V,
and S4). From Figure 2f, a spacing of 3.26 Å can be identified which is supported by the absence of the obvious current peak
clearly from a marked particle, which is slightly larger than the in cyclic voltammetry (CV) collected under Ar atmosphere
spacing of the (110) plane of standard RuO nanocrystals (3.18 (Figure 3b). These results indicate that some oxygen radical
2
Å). This larger interplanar spacing observed here is consistent induced side-reactions occurred at this voltage, meaning that
with the results indicated from the XRD pattern that shifts to the upper cut-off for the DMSO/m-BCN-based system should
lower angles (Figure 2b). Furthermore, the high-angle annular lower than the voltage of 4.44 V. In contrast, the RuO
2
@m-BCN
dark field (HADAF) image and the corresponding EDS elemental shows a well-defined plateau at 2.78 V on initial discharge and
mappings confirm the uniform distribution of the RuO
nanoparticles on the m-BCN supports (Figure 2g-k).
2
a slope ranging from 3 to 3.48 and a plateau at 3.55 followed by
an upward slope at around 4 V on recharge (Figure 3a). This
As discussed in Introduction, the conductivity is of important leads to a much larger discharge capacity as well as smaller
influence on the performance of Li-O batteries. The electrical overpotential than that of the sole m-BCN cathode (Figure 3a),
conductivity of the RuO @m-BCN and m-BCN were indicating an enhanced activity upon ORR/OER. Specifically, the
characterized by I-V measurement. The testing samples were discharge capacity of the RuO
2
2
2
@m-BCN increases more than 12
prepared by pressing their powders into ~ 1 mm-thick pallet (8 times to 2.57 mAh cm (i.e., 512 mAh g based on the mass
-2
-1
mm in diameter) under a pressure of 6 MPa (Figure S5). Figure loading of the RuO
S6 shows the I-V plots of the RuO @m-BCN and m-BCN under a the highly porous structure of RuO
given voltage. For the m-BCN, its resistance is so high that only storage of more discharge products and fast mass transport.
a very small value of current feedback can be collected. In This tailored porous structure of RuO @m-BCN significantly
contrast, a much larger value of current collected for RuO @m- increases the contact areas between RuO catalysts and
BCN, indicating a significantly improved conductivity for the discharge products, thus lowering the overpotentials to 0.18 V
2
@m-BCN), which is indeed beneficial from
2
2
@m-BCN that favours the
2
2
2
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Jo uP rl ne aa sl eo fd oM na ot et r ai ad l jsu sCt hme am r gi si nt rs y A
Page 4 of 7
ARTICLE
Journal Name
for discharge and 0.54 V for recharge. Herein the discharge previously, and favour a large discharge capacity.5, 43-48 Here,
View Article Online
DOI: 10.1039/C7TA05009G
capacity for the RuO
carbon-based electrodes,
previously reported carbon-free cathodes (Table S1).
2 2 2
@m-BCN is smaller than that of many the formation of Li O toroid may benefit from the non-
10, 40
but is comparable with that of stoichiometric feature of RuO
2
along with the contribution form
After recharge, these discharge
products are completely removed even the voltage just above 4
V (Figure 3d). These reversible morphology changes indicate the
12, 19, 20, 23,
5, 24
DMSO-based electrolytes.
26
superior ability of RuO
of discharge products during cycling, which is essential for the
stable operation of Li-O batteries.
2
@m-BCN to promote the decomposition
2
Figure 4. Electrochemical behaviour of RuO
2 2
@m-BCN in Li-O
batteries. (a) cycle performance with a current density of 0.3
-
2
mA cm and (b) rate performance.
The cyclability of the RuO @m-BCN cathodes were investigated
in a full discharge-charge protocol between 2.5 and 4.2 V.
Remarkably, the RuO @m-BCN cathode exhibits an impressive
2
Figure 3. (a) The first discharge-charge profiles of the m-BCN
and the RuO @m-BCN cathode with a current density of 0.2 mA
cm ; (b) CV curves of RuO @m-BCN collected under Ar and O
saturated DMSO electrolytes, respectively; (c-d) the
morphology of the 1st discharged and recharged electrode of
2
cycle stability up to 126-cycle with almost overlapped
discharge-charge curves and stable reversible capacities around
2
-
2
2
2
-
2
-2
1
.8 mAh cm at 0.3 mA cm (Figure 4a and S9). In the course of
cycling, the overpotentials for the RuO @m-BCN remain stable
2
with a low value of ~ 0.6 V, which is completely different with
the phenomenon that the overpotential increases rapidly upon
the RuO
2
@m-BCN, respectively.
Figure 3b shows the CV curve of the RuO
2
@m-BCN collected in
10, 11, 43, 44, 47, 48
cycling observed in many carbon electrodes.
oxygen saturated DMSO electrolytes. During the first cathodic
scan, one profound reduction peak appears at around 2.72 V,
which corresponds to the formation of Li O . Upon anodic scan,
2 2
Furthermore, discharging/charging tests of this RuO
2
@m-BCN
cathode at a series of rates clearly show its superior rate
capability. As shown in Figure 4b, the RuO
2
@m-BCN deliver a
the oxidation of discharge products gives an obvious OER peak
at 3.16 V followed by a flat peak at around 3.8 V, which is well
consistent with the two oxidation plateaus appeared in Figure
-2
reversible capacity of 1.90, 1.68, and 1.28 mAh cm at the
-2
current density of 0.3, 0.5 and 1 mA cm , respectively. The
2
excellent performance of RuO @m-BCN is ascribed to the
following advantages: (1) the use of m-BCN, which excludes the
fragile C-C bond and thus enhances the structure stability upon
19, 23, 30
3
a and previous reports.
It can be seen that the first
oxidation peak contributes more than the second one to the
total oxidation current, meaning that most of the discharge
products are oxidized below 3.8 V. This indicates that the
cycling; (2) the introduction of evenly distributed RuO
2
nanoparticles, which provide substantial active sites for
ORR/OER and significantly improved electronic wiring with
discharge products; (3) the tailored porous structure, which
RuO
For the second cycle, the CV curve remains unchanged,
indicating good stability and activity of the RuO @m-BCN upon
repeated ORR/OER. Besides the two oxidation peaks located at
.16 and 3.8 V, an oxidation peak appears around the voltage
cap of 4.3 V, which is attributed to the decomposition of DMSO
2 2 2
@m-BCN can promote the early decomposition of Li O .
2
favours good electrolyte wetting and fast O
In addition, the high performance of RuO
2
transport.
@m-BCN, in some
2
3
extent, may benefit from the cycle protocol, in which a higher
charging cutoff is used. We previously found that charging
vertical aligned carbon nanotubes based Li-O cells to a slightly
2
4
1, 42
solvents and/or some by-products.
During initial cycling, the morphology of the discharge products
deposited on the RuO @m-BCN electrodes were visualized by
higher voltage favours the electrochemical splitting of discharge
products more completely and renews the cathodes for
2
SEM (Figure 3c-d). Compared with the pristine electrode (Figure
S7), toroid particles meeting the characteristic morphology of
49
subsequent discharging, leading to an improved cyclability.
Figure S10 and S11 show the cycle performance and CV curves
collected from cells operated between 2.5 and 4 V, respectively.
Compared with the results shown in Figure 4a, this cell suffers a
poor cycle stability. The main cause should be the accumulation
of by-products, leading to the bury of active sites and loss of
6
, 43, 44
Li
RuO
2
O
2
with a size around 200 nm formed and fill the
@m-BCN cathode during initial discharge (Figure 3c).
2
Excellent uniformity of such discharge products is further
confirmed by a SEM image with lower magnification (Figure S8).
The toroidal morphology indicates
a
solution growth
50
electronic conductivity.
mechanism of Li formation, which has been well studied
2 2
O
4
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 7
Jo uP rl ne aa sl eo fd oM na ot et r ai ad l jsu sCt hme am r gi si nt rs y A
Journal Name
ARTICLE
Electrode
Characterization
To demonstrate the excellent stability of the RuO @m-BCN and
2 2
the reversibility of formation and oxidation of Li O , the cycled
electrodes were analysed by ex-situ XRD, XPS and Fourier
transform infrared spectroscopy (FTIR). The commercialized
Stability
and
Electrochemical
Products pristine, 5th discharged, 5th charged, 20th discharged, and 20th
View Article Online
DOI: 10.1039/C7TA05009G
.
charged electrodes.
2
Furthermore, the cycled electrodes were investigated by ex situ
FTIR to study whether the LiOH is involved during cycling.
Compared to the spectrum of the pristine electrode saturated
with DMSO-based electrolytes, a new broad peak ranging from
2 2 2 3
Li O , LiOH and Li CO powders were used as the references for
-
1
3
50 to 650 cm appears in the 5th discharged electrode and
identifying the chemistry of reaction products. As shown in
Figure S12, the 5th discharged and pristine electrodes show
almost identical XRD patterns. The main reasons for this
observation may be the absence of crystalline products or their
amount too small to be detected. Then the XPS analyses, a
surface sensitive technique that can detect the amorphous
and/or trace products formed on electrodes, were conducted.
For the Li 1s photoemission spectra of the 5th cycled electrodes,
a broad peak centred at 54.7 eV meeting the binding energy of
disappears after recharge, which is also reproducible for the
2
5
0th cycle (Figure 5c). The broad characteristic peak centred at
45 cm can be assigned to Li O , which is consist with the
2 2
-
1
results from XPS (Figure 5a). Another characteristic peak
-1
appeared at 460 cm meets the peak position of standard
5
3
LiOH, indicating the existence of LiOH, which has been
reported several times and is ascribed to the decomposition of
54
DMSO solvent during cycling. After recharge to 4.1 V, the
disappearance of this two characteristic peaks indicates that
these discharge products including LiOH have been removed
3
0, 51
Li
Figure 5a). Moreover, such peak appears further even in the
0th and 40th discharged electrodes, indicating the excellent
repeatability of the reaction occurred on the RuO @m-BCN
cathodes. We cannot rule out the formation of the irreversible
product of LiOH from the reversible product of Li only from
the Li 1s photoemission spectra, due to their near-identical
O
2 2
appears upon discharge and disappears after recharge
(
2
(Figure 5c). This results a “clear” electrode surface as showed in
Figure 3d, which is crucial for subsequent operation. Here, the
splitting of LiOH at around 4 V with an upward slope (Figure 4a
and c) is attributed to the tailored porous structure, which
2
2 2
O
guarantees the good contact of discharge products with RuO
catalysts.
2
55
52
binding energies. However, the excellent reversibility (Figure
a) and previous demonstrated rechargeability (Figure 4) lead
us to tentatively conclude that the main discharge product
deposited on the RuO @m-BCN electrodes is Li . This
conclusion is further supported by the 1s/Ru 3d
photoemission spectra (Figure 5b), in which no obvious peaks
5
2
2 2
O
C
2
−
corresponding to the CO3 can be observed. One can see only
a very small peak in the discharged electrode (e.g., the 5th, 20th
and 40th) at around 290 eV (Figure 5b),3
0, 52
helping us rule out
CO
during prolonged cycling. The absence of the serious
accumulation of Li CO during long-term cycling demonstrates
the superior stability of the RuO @m-BCN upon Li and
superoxide-like species. For the observed small peak of Li CO
the serious situation of formation and accumulation of Li
2
3
Figure 6. (a) cycle performance and (b) selected discharge-
charge profiles of the RuO @m-BCN cathode with a current
2
-2
2
3
density of 0.5 mA cm . The inset of (a) shows the photograph
of the lithium anode after 141 cycles.
2
2 2
O
2
3
,
To further evidence the enhanced stability of this carbon-free
cathode, the cell that has shown significant capacity decay after
141 cycles was reopened in the glove-box (Figure 6a). No
obvious changes on cathode and electrolyte (e.g., colour) can
be observed by naked eye, in some extent indicating good
stability. While conspicuous changes were observed in lithium
anode. Specifically, the colour changes from metallic into grey;
and the flat surface becomes uneven and is covered by a layer
of moss-like structures (inset of Figure 6a). The same
observations of the failure of Li anode were reported by
it may result from the electrode surface contamination by
limited exposure to the ambient air during XPS tests, as the peak
2
−
with feature of CO also appeared in the commercialized Li
O
2 2
3
and LiOH powders (Figure 5b).
51
56
Schroeder et al. and Lu et al. . Ex situ XRD analysis of another
anode failed after 107 cycles indicates that the moss-like
structures consist of the mixture of LiOH and Li
The formation of such insulators significantly increase Li and e
transfer barriers, and may be the reason answering for the
death of RuO @m-BCN based battery. To verify this, the cycled
anode was replaced with a new one, and the cell was
reassembled with other components unchanged. It is surprising
that the reassembled cell works healthily as it behaved before
the cell failure (Figure 6a-b). The cathode shows a full recovery
2
O (Figure S13).
+
-
Figure 5. Ex-situ measurements on oxygen electrodes. Ex-situ
XPS obtained from 5th discharged, 5th charged, 20th
discharged, and 40th discharged electrodes collected in the (a)
Li 1s, (b) C 1s/Ru 3d; (c) Ex-situ FTIR spectra obtained from
2
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Jo uP rl ne aa sl eo fd oM na ot et r ai ad l jsu sCt hme am r gi si nt rs y A
Page 6 of 7
ARTICLE
Journal Name
7
Y. Lu, B. M. Gallant, D. G. Kwabi, J. R. Harding, RV.iewR.AMrticiltecOhnelinlle,
of the cell voltage profiles and even slightly larger capacities,
which last for another 87 cycles (Figure 6). The consistency in
cell performance (i.e., voltage profile and capacity) before and
after a cycled and “dead” cell reassembled with a new lithium
M. S. Whittingham and Y. Shao-Horn, Energy Environ. Sci.,
DOI: 10.1039/C7TA05009G
2
013, 6, 750-768.
8
9
1
1
X. Yao, Q. Dong, Q. Cheng and D. Wang, Angew. Chem. Int. Ed.,
016, 55, 11344-11353.
2
anode strongly suggests that the as-prepared RuO
2
@m-BCN
M. Balaish, A. Kraytsberg and Y. Ein-Eli, Phys. Chem. Chem.
Phys., 2014, 16, 2801-2822.
cathodes possess excellent stability and activity upon ORR/OER,
guaranteeing the impressive cell performance. This also
suggests that effective strategies capable of enhancing the
reversibility and stability of lithium anode are urgently needed
0 Z. Ma, X. Yuan, L. Li, Z. Ma, D. P. Wilkinson, L. Zhang and J.
Zhang, Energy Environ. Sci., 2015, 8, 2144-2198.
1 Y. Shao, F. Ding, J. Xiao, J. Zhang, W. Xu, S. Park, J. Zhang, Y.
Wang and J. Liu, Adv. Funct. Mater., 2013, 23, 987-1004.
for stable operation of Li-O
2
batteries.
12 Z. Peng, S. A. Freunberger, Y. Chen and P. G. Bruce, Science,
2
012, 337, 563-566.
3 S. T. Kim, N.-S. Choi, S. Park and J. Cho, Adv. Energy Mater.,
015, , 1401030.
4 K. Liao, T. Zhang, Y. Wang, F. Li, Z. Jian, H. Yu and H. Zhou,
ChemSusChem, 2015, , 1429-1434.
5 K.-H. Kwak, D. W. Kim, Y. Kang and J. Suk, J. Mater. Chem. A,
016, , 16356-16367.
6 K. Liao, X. Wang, Y. Sun, D. Tang, M. Han, P. He, X. Jiang, T.
Zhang and H. Zhou, Energy Environ. Sci., 2015, , 1992-1997.
7 X. Han, F. Cheng, C. Chen, F. Li and J. Chen, Inorg. Chem. Front.,
2016, , 866-871.
8 R. S. Kalubarme, H. S. Jadhav, D. T. Ngo, G. E. Park, J. G. Fisher,
Y. I. Choi, W. H. Ryu and C. J. Park, Sci. Rep., 2015, , 13266.
9 D. Kundu, R. Black, E. J. Berg and L. F. Nazar, Energy Environ.
Sci., 2015, , 1292-1298.
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
5
Conclusions
In summary, we developed a promising carbon-free cathode
prepared by introducing RuO nanoparticles into mesoporous
boron doped carbon nitride (RuO @m-BCN). Benefitting from
the evenly distributed RuO and robust m-BCN, the RuO @m-
BCN shows excellent stability and activity upon ORR/OER. With
such cathodes, the Li-O batteries demonstrate the reversible
formation and decomposition of Li with no accumulation of
Li CO , relatively low charge overpotential, excellent cycle
stability and rate capability. This impressive performance is
ascribed to the tailored properties of the RuO @m-BCN,
8
2
2
4
2
2
2
8
2
3
2 2
O
5
2
3
8
2
0 M. M. Ottakam Thotiyl, S. A. Freunberger, Z. Peng, Y. Chen, Z.
Liu and P. G. Bruce, Nat. Mater., 2013, 12, 1050-1056.
1 K. R. Yoon, G. Y. Lee, J. W. Jung, N. H. Kim, S. O. Kim and I. D.
Kim, Nano Lett., 2016, 16, 2076-2083.
including the inherent stability of carbon-free supports that
decrease the amount of by-products from electrode and/or
electrolyte self-decomposition, the enhanced catalytic activity
2 Z. Guo, C. Li, J. Liu, X. Su, Y. Wang and Y. Xia, J. Mater. Chem.
A, 2015, 3, 21123-21132.
upon Li/O
RuO nanoparticles, and the interconnected porous structures
that favour the storage of more discharge products and fast
mass transfer. The enhanced stability of the RuO @m-BCN is
2
chemistry stemming from the evenly distributed
2
3 F. Li, D. M. Tang, Y. Chen, D. Golberg, H. Kitaura, T. Zhang, A.
Yamada and H. Zhou, Nano Lett., 2013, 13, 4702-4707.
4 Y. Chang, S. Dong, Y. Ju, D. Xiao, X. Zhou, L. Zhang, X. Chen, C.
2
Shang, L. Gu, Z. Peng and G. Cui, Adv. Sci., 2015, 2, 1500092.
further supported by the consistency in voltage profiles and
capacity before and after a cycled and “dead” cell reassembled
with a new lithium anode. These results indicate that
constructing highly efficient and stable cathode with optimized
5 S. Liu, G. Wang, F. Tu, J. Xie, H. Y. Yang, S. Zhang, T. Zhu, G. Cao
and X. Zhao, Nanoscale, 2015, 7, 9589-9596.
6 S. Song, W. Xu, R. Cao, L. Luo, M. H. Engelhard, M. E. Bowden,
B. Liu, L. Estevez, C. Wang and J. Zhang, Nano Energy, 2017,
33, 195-204.
porous structure is essential for the stable operation of Li-O
batteries.
2
2
2
2
3
7 W. J. Ong, L. L. Tan, Y. H. Ng, S. T. Yong and S. P. Chai, Chem.
Rev., 2016, 116, 7159-7329.
8 W. B. Luo, S. L. Chou, J. Z. Wang, Y. C. Zhai and H. K. Liu, Small,
2
015, 11, 2817-2824.
Acknowledgements
The financial supports from the National Natural Science
Foundation of China (51402339), National Key Basic Research
9 J. Yi, K. Liao, C. Zhang, T. Zhang, F. Li and H. Zhou, ACS Appl.
Mater. Interfaces, 2015, , 10823-10827.
0 P. Lou, C. Li, Z. Cui and X. Guo, J. Mater. Chem. A, 2016,
49.
7
4
, 241-
2
Program of China (2014CB921004) and Science Foundation for 31 B. Sun, L. Guo, Y. Ju, P. Munroe, E. Wang, Z. Peng and G. Wang,
Youth Scholar of State Key Laboratory of High Performance
Ceramics and Superfine Microstructures (SKL201303) are
gratefully acknowledged.
Nano Energy, 2016, 28, 486-494.
3
2 C. Zhao, C. Yu, M. N. Banis, Q. Sun, M. Zhang, X. Li, Y. Liu, Y.
Zhao, H. Huang, S. Li, X. Han, B. Xiao, Z. Song, R. Li, J. Qiu and
X. Sun, Nano Energy, 2017, 34, 399-407.
3
3 Y. Wang, J. Zhang, X. Wang, M. Antonietti and H. Li, Angew.
Chem. Int. Ed., 2010, 49, 3356-3359.
Notes and references
34 Q. Guo, Y. Xie, X. Wang, S. Lv, T. Hou and X. Liu, Chem. Phys.
Lett., 2003, 380, 84-87.
1
2
M. Armand and J. M. Tarascon, Nature, 2008, 451, 652-657.
K. M. Abraham and Z. Jiang, J. Electrochem. Soc., 1996, 143, 1-
3
3
3
3
5 Y. Wang, H. Li, J. Yao, X. Wang and M. Antonietti, Chem. Sci.,
011, , 446-450.
2
2
5
.
6 S. Zhang, L. Gao, D. Fan, X. Lv, Y. Li and Z. Yan, Chem. Phys.
Lett., 2017, 672, 26-30.
3
4
5
6
P. G. Bruce, S. A. Freunberger, L. J. Hardwick and J. M.
Tarascon, Nat. Mater., 2011, 11, 19-29.
7 Y. Lee, B.-U. Ye, H. k. Yu, J.-L. Lee, M. H. Kim and J. M. Baik, J.
Phys. Chem. C, 2011, 115, 4611-4615.
8 W. H. Ryu, T. H. Yoon, S. H. Song, S. Jeon, Y. J. Park and I. D.
Kim, Nano Lett., 2013, 13, 4190-4197.
G. Girishkumar, B. McCloskey, A. C. Luntz, S. Swanson and W.
Wilcke, J. Phys. Chem. Lett., 2010, 1, 2193-2203.
D. Aurbach, B. D. McCloskey, L. F. Nazar and P. G. Bruce, Nat.
Energy, 2016, , 16128.
1
Z. Cui, X. Guo and H. Li, Energy Environ. Sci., 2015,
8, 182-187.
6
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 7 of 7
Jo uP rl ne aa sl eo fd oM na ot et r ai ad l jsu sCt hme am r gi si nt rs y A
Journal Name
ARTICLE
3
4
9 P. Balaya, H. Li, L. Kienle and J. Maier, Adv. Funct. Mater., 2003,
View Article Online
DOI: 10.1039/C7TA05009G
13, 621-625.
0 J. Xiao, D. Mei, X. Li, W. Xu, D. Wang, G. L. Graff, W. D. Bennett,
Z. Nie, L. V. Saraf, I. A. Aksay, J. Liu and J. G. Zhang, Nano Lett.,
2
011, 11, 5071-5078.
4
4
1 N. Mozhzhukhina, L. P. Méndez De Leo and E. J. Calvo, J. Phys.
Chem. C, 2013, 117, 18375-18380.
2 M. A. Schroeder, N. Kumar, A. J. Pearse, C. Liu, S. B. Lee, G. W.
Rubloff, K. Leung and M. Noked, ACS Appl. Mater. Interfaces,
2
015,
3 Z. Cui, W. Fan and X. Guo, J. Power Sources, 2013, 235, 251-
55.
4 W. Fan, Z. Cui and X. Guo, J. Phys. Chem. C, 2013, 117, 2623-
627.
5 B. D. Adams, C. Radtke, R. Black, M. L. Trudeau, K. Zaghib and
L. F. Nazar, Energy Environ. Sci., 2013, , 1772-1778.
6 R. R. Mitchell, B. M. Gallant, C. V. Thompson and Y. Shao-Horn,
Energy Environ. Sci., 2011, , 2952-2958.
7, 11402-11411.
4
4
4
4
2
2
6
4
4
4
7 Z. Cui and X. Guo, J. Power Sources, 2014, 267, 20-25.
8 T. Ogasawara, A. Débart, M. Holzapfel, P. Novák and P. G.
Bruce, J. Am. Chem. Soc., 2006, 128, 1390-1393.
4
5
9 X. Guo and N. Zhao, Adv. Energy Mater. 2013, 3, 1413-1416.
0 J. Wang, Y. Zhang, L. Guo, E. Wang and Z. Peng, Angew. Chem.
Int. Ed., 2016, 55, 5201-5205.
1 M. A. Schroeder, A. J. Pearse, A. C. Kozen, X. Chen, K.
Gregorczyk, X. Han, A. Cao, L. Hu, S. B. Lee, G. W. Rubloff and
M. Noked, Chem. Mater., 2015, 27, 5305-5313.
2 Y. Lu, A. N. Mansour, N. Yabuuchi and Y. Shao-Horn, Chem.
Mater., 2009, 21, 4408-4424.
5
5
5
5
3 Y. Qiao and S. Ye, J. Phys. Chem. C, 2016, 120, 15830-15845.
4 D. G. Kwabi, T. P. Batcho, C. V. Amanchukwu, N. Ortiz-
Vitoriano, P. Hammond, C. V. Thompson and Y. Shao-Horn, J.
Phys. Chem. Lett., 2014,
5 F. Li, S. Wu, D. Li, T. Zhang, P. He, A. Yamada and H. Zhou, Nat.
Commun., 2015, , 7843.
5, 2850-2856.
5
5
6
6 J. Lu, Y. J. Lee, X. Luo, K. C. Lau, M. Asadi, H. H. Wang, S.
Brombosz, J. Wen, D. Zhai, Z. Chen, D. J. Miller, Y. S. Jeong, J.
B. Park, Z. Z. Fang, B. Kumar, A. Salehi-Khojin, Y. K. Sun, L. A.
Curtiss and K. Amine, Nature, 2016, 529, 377-382.
TOC
:
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins