- Email: [email protected]
graphene oxide batteries
Accepted Manuscript Metal/graphene oxide batteries Minghui Ye, Jian Gao, Yukun Xiao, Tong Xu, Yang Zhao, Liangti Qu PII:
S0008-6223(17)30946-6
DOI:
10.1016/j.carbon.2017.09.070
Reference:
CARBON 12400
To appear in:
Carbon
Received Date: 26 July 2017 Revised Date:
4 September 2017
Accepted Date: 16 September 2017
Please cite this article as: M. Ye, J. Gao, Y. Xiao, T. Xu, Y. Zhao, L. Qu, Metal/graphene oxide batteries, Carbon (2017), doi: 10.1016/j.carbon.2017.09.070. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Metal/graphene oxide batteries Minghui Ye, Jian Gao, Yukun Xiao, Tong Xu, Yang Zhao, Liangti Qu*
M AN U
SC
RI PT
Beijing Key Laboratory of Photoelectronic/Electrophotonic Conversion Materials, Key Laboratory of Cluster Science, Ministry of Education of China, Department of Chemistry, Beijing Institute of Technology, Beijing 100081, P. R. China. Email: [email protected] Fax: 86-10-68918608
TE D
Metal/graphene oxide (M/GO) batteries are demonstrated by the simple assembly of metals and GO films, in which the spontaneous redox reaction enables the conversion of chemical energy into electricity. Beyond the 2D planar structure, semi-solid
EP
3D-Gu/GO redox flow battery and smart pressure-responsive 3D-GO/Zn battery with
AC C
controllable energy release are developed.
ACCEPTED MANUSCRIPT
Metal/graphene oxide batteries Minghui Ye, Jian Gao, Yukun Xiao, Tong Xu, Yang Zhao, Liangti Qu*
RI PT
Beijing Key Laboratory of Photoelectronic/Electrophotonic Conversion Materials, Key Laboratory of Cluster Science, Ministry of Education of China, Department of Chemistry, Beijing Institute of Technology, Beijing 100081, P. R. China. Email: [email protected] Fax: 86-10-68918608
Abstract: Based on the spontaneous redox reaction between metal and graphene
SC
oxide, a novel type of metal/graphene oxide (M/GO) batteries is developed to convert chemical energy into electricity, including Li/GO, Na/GO, Zn/GO, Fe/GO, and
M AN U
Cu/GO batteries. They are fabricated via the simple assembly of metal foils with GO films, in which M plays the role of anode, and GO acts as both cathode and separator. Among them, Li/GO battery generates the highest specific capacity of 1,572 mAh cm–3 (about 1,604 mAh g–1). The energy density of M/GO battery is determined by
TE D
the contact area of M with GO. Therefore, three-dimensional (3D) M/GO battery will deliver higher energy in comparison to 2D planar M/GO battery. As expected, a
EP
semi-solid 3D-Cu/GO redox flow battery (RFB) is assembled by 3D Cu foam with flowing GO/ionic-liquid catholyte. Its specific capacity is ca. 97 times that of Cu
AC C
foil/GO RFB. Besides, a compressible, all-solid-state, and pressure-responsive 3D-GO/Zn battery is also fabricated. It can accurately control the energy output in response to pressure stimulations without the aid of conventional battery management system. Beyond those demonstrated in this work, the concept of M/GO battery will shed light on the design of similar electrochemical power sources.
1. Introduction
ACCEPTED MANUSCRIPT Graphene is a two-dimensional (2D) honeycomb lattice of sp2-bonded carbon atoms [1]. Graphene oxide (GO) is a graphene derivative with abundant oxygen-related functional groups attaching to its basal plane and sheet edge, such as
RI PT
hydroxyl, epoxy, and carboxyl groups [2]. It can be cheaply obtained in large quantities via the chemical exfoliation of graphite [3–7]. GO could be reduced by active metal, such as zinc (Zn) [8,9], ferrum (Fe) [10], and copper (Cu) [11], upon
SC
inserting metal foil into aqueous GO dispersion at ambient condition. As illustrated in
M AN U
our previous work [8–11], the contact of GO with metal induced the redox reaction: M + GO → rGO + MxOy, where rGO and MxOy were reduced graphene oxide and the resultant metal oxide, respectively. Electrons could spontaneously transfer from metal to GO, leading to the reduction of GO and the oxidation of metal. Based on the redox
TE D
reaction between active metal and GO, a series of metal/GO galvanic batteries are developed by the simple assembly of active metal foils and solid-state GO(s) films. They can directly output available electricity resulted from the conversion of chemical
EP
energy. The active metals can be further extended to lithium (Li) and sodium (Na).
AC C
There are no separation membranes existing in those M/GO batteries. This is because GO(s) acts as both cathode and separator to avoid the short circuit of the battery due to its insulating nature. Meanwhile, M plays the role of anode in offering electrons, so that electrical current is allowed to flow in external circuit. As expected, Li/GO battery generates a specific capacity of ~1,572 mAh cm–3 (corresponding to 1,604 mAh g–1). Beyond the 2D planar structure of M/GO battery, constructing 3D M/GO battery with higher energy density and controlled energy release is very important and
ACCEPTED MANUSCRIPT significative. Accordingly, a semi-solid 3D-Cu/GO redox flow battery (3D-Cu/GO RFB) is achieved by using Cu foam as anode and flowing GO/ionic-liquid as cathode, whose specific capacity is ca. 97 times that of Cu foil/GO RFB. Besides, a
RI PT
compressible, all-solid-state, and pressure-responsive 3D-GO/Zn battery is also developed, in which GO-coated sponge and Zn foil serve as cathode and anode, respectively. The 3D-GO/Zn battery can be repeatedly compressed and released to
M AN U
of conventional battery management system.
SC
accurately control its energy output in response to pressure stimulations without need
2. Experimental
2.1 The fabrication of GO0.3 and GO0.6
TE D
The aqueous GO dispersion was prepared through the oxidation exfoliation of graphite according to the modified Hummer’s method [22]. In this context, two kinds of GO with O/C atomic ratios of 0.3 and 0.6 were denoted as GO0.3 and GO0.6,
EP
respectively. The preparation of GO0.3 was as follows: First, 80 mL concentrated
AC C
H2SO4 was added to 250 mL flask filled with 3 g graphite in an ice bath with mechanical agitation, and then 3 g NaNO3 and 9 g KMnO4 were slowly added, respectively. Second, the flask equipped with condensing system was transferred to 36 o
C oil bath and vigorously stirred for 1 h, followed by the addition of 150 mL
deionized water. Successively, the reaction system was heated up to 90 oC and maintained for 20 min, followed by the dropwise addition of 5 mL H2O2 (30%). Third, the suspension was cooled to room temperature, which was then filtered and washed
ACCEPTED MANUSCRIPT with about 1.2 M HCl aqueous solution to remove metal impurity. The obtained solid was further filtered by adding deionized water and then dispersed in 400 mL water, followed by stirring overnight. Finally, the as-produced GO dispersion was
RI PT
centrifuged at 10,000 r.p.m. for 20 min to remove the unexfoliated particles, followed by the dialysis in deionized water for one week using dialysis bags.
The preparation of GO0.6 was identical to that of GO0.3 except changing the dose
SC
of KMnO4 into 18 g and prolonging the oxidized time to 16 h at a higher temperature
M AN U
of 50 oC.
2.2 The fabrication of M/GO batteries
GO film with a density of ~0.98 g cm–3 was obtained by the freeze-drying of
TE D
aqueous GO dispersion and subsequent tableting with a pressure of ca. 20 MPa. The thickness of GO film was about 30 µm. The M foil (i.e., Li, Na, Zn, Fe, Cu) and GO film were sandwiched between the lower battery shell and stainless steel (SS) foil
EP
plated with Au, then encapsulated into CR2032-type coin cells in Ar-filled glove box
AC C
(Fig. S1). The electrolytes for Li/GO battery and Na/GO battery were 1 M LiPF6 dissolved in ethylene carbonate (EC)/dimethyl carbonate (DMC)/diethyl carbonate (DEC) (1:1:1 in volume), and 1 M NaPF6 in a 1:2 (v/v) mixture of EC and DEC, respectively.
Zn/GO
battery,
Fe/GO
battery,
and
Cu/GO
battery
used
1-butyl-3-mthylimidazolium tetrafluoroborate (BMIMBF4) as electrolyte. The voltage and current outputs were recorded by a Keithley 2400 sourcemeter. The circuit parameters of the open circuit voltage test were current = 0 mA and step index = 5
ACCEPTED MANUSCRIPT points per s. The circuit parameters of the short circuit current test were voltage = 0 mV and step index = 5 points per s. The specific capacity of M/GO battery was calculated from I-T curve (Fig. 2d and Fig. S5a, b, d, f) by using the following
C =
∫ idt
RI PT
formula:
where C was the volume capacity (mAh cm–3), i was the discharging current density
M AN U
SC
(mA cm–3) based on the volume of GO film, and dt was the discharging time (h) [13].
2.3 The fabrication of 3D-Cu/GO RFB
The catholyte was obtained via adding freeze-drying GO monoliths into BMIMBF4 to form 2 mg mL–1 dispersion, and then poured into a beaker. Afterward, a
TE D
piece of fresh Cu foam was inserted into catholyte with Au foil as the current collector to construct a two-electrode system (Fig. S17b). A magnetic stirring bar was placed into the beaker to enable GO catholyte to flow when the RFB was in operation. The
AC C
EP
electrical signals were recorded by Keithley 2400 sourcemeter.
2.4 The fabrication of 3D-GO/Zn battery Commercial polyurethane (PU) sponge was cut into cuboids and then immersed
into 2 mg mL–1 GO solution by a squeezing procedure. Subsequently, the lyophilization was carried out to obtain GO-coated PU sponge using liquid nitrogen, whose density was about 16.1 mg cm−3. GO accounted for ~16% of the whole sponge weight. The 3D-GO sponge was mounted on the fresh Zn foil to form 3D-GO/Zn
ACCEPTED MANUSCRIPT battery in the absence of electrolyte (Fig. S18). To investigate the controllable power output behavior, the 3D-GO/Zn battery was sandwiched in a CR2032-type coin cell without encapsulation. The lower battery shell laminated to a nonconductive
RI PT
polyethylene terephthalate (PET) film was fixed to the pressure testing system, while
compression/decompression cycles.
M AN U
2.5 The fabrication of lab-made coin battery
SC
the upper shell with PET film was able to move up and down during the
The home-built coin battery was composed of Li2TiO3 (LTO), Li foil, and commercial separator membrane (celgard 2400 microporous polypropylene membrane). The LTO electrode was prepared by mixing Li2TiO3, acetylene black, and
TE D
polyvinylidene fluoride (PVDF) binder with weight ratios of 8:1:1 into N-methylpyrrolidone (NMP) to form monodispersed slurry. Then the as-produced slurry was pasted onto a Cu foil and dried at 120 oC for 12 h under vacuum. For one
EP
electrode, it typically had a loading of ~0.65 mg cm–2. The electrolyte was 1.0 M
AC C
LiPF6 in ethylene carbonate/dimethyl carbonate/diethyl carbonate (EC: DMC: DEC = 1:1:1 in volume). The assembly of coin cell was performed in an argon-filled glove box, where moisture and oxygen concentrations were strictly limited below 1.0 ppm. The cells were then aged for 12 h before measurement. The galvanostatic discharge/charge test was carried out over a voltage range of 0.01–2.5 V versus Li+/Li on the LAND CT2001A battery system at room temperature. Electrochemical impedance spectral measurements were carried out in the frequency range from 100
ACCEPTED MANUSCRIPT kHz to 0.01 Hz on CHI760D electrochemical workstation.
2.6 Characterization
RI PT
Field-emission scanning electron microscopy (FE-SEM) and X-ray energy dispersive spectroscopy (EDS) of the samples were carried out on a JSM-7001F SEM unit (Japan Electron Optics Laboratory Co., Ltd, Japan). Transmission electron
SC
microscopy (TEM) and high-resolution TEM (HR-TEM) were conducted using a
High-angle annular dark
field
M AN U
TecnaiG2 20ST (T20) at an acceleration voltage of 120 kV (FEI corporation, USA). scanning transmission
electron
microscopy
(HAADF-STEM) was operated at 200 KV (FEI, Technai G2 F30). X-ray photograph electron spectroscopy (XPS) data were carried on an ESCALab220i-XL electron
TE D
spectrometer using 300 W AlK irradiation (VG Scientific Co., Ltd, UK). The base pressure was about 3 × 10–9 mbar. Powder X-ray diffraction (XRD) patterns were performed on a Bruker D8 advance X-ray diffractometer (Bruker Corporation,
EP
Germany) using a Netherlands 1,710 diffractometer with a Cu Kα irradiation source
AC C
(λ = l.54 Å) and a self-calibration process was performed with a SiO2 internal standard sample prior to target measurement. Raman spectra were recorded using a RM 2000 Microscopic Confocal Raman Spectrometer (Renishaw PLC, England) with a 514 nm laser. The repeated compression/decompression tests of 3D-GO/Zn battery were conducted with an Instron material testing system (Instron 3342), and the current variation was recorded in real time using a Keithley 2612A sourcemeter, which was controlled by a LabView-based data acquisition system.
ACCEPTED MANUSCRIPT
3. Results and discussion Fig. 1a schematically illustrates the M/GO batteries formed by the simple
RI PT
assembly of active metal (i.e., Li, Na, Zn, Fe, and Cu) foils and GO films into button-type cells without any separators (Fig. S1). GO film with a thickness of ~30 µm (Fig. S2a–d) can be obtained by the freeze-drying of aqueous GO dispersion and
SC
subsequent tableting. Both shells of the battery are plated with aurum (Au) to act as
M AN U
the current collector and its encapsulation is performed at argon-filled glove box. In the designed M/GO battery, active metal and GO(s) act as the anode and cathode, respectively. Electrical current is allowed to flow in external circuit along with the redox reaction of two electrodes, thus converting chemical energy into electricity. It is
TE D
worth noting that GO film in touch with Li, Na, Zn, Fe, and Cu is reduced into rGO and the yielded MxOy (i.e., Li2O/Li2CO3, Na2O/Na2CO3, ZnO, Fe3O4, and CuO, respectively) are anchored onto rGO nanosheets (Fig. 1b). According to the
EP
relationship between the thermodynamic function and the cell potential as follows:
AC C
∆G = –nFE = –nF[φrGO/GO –φM/Mn+]
Where, ∆G, F, E, and Mn+ are the Gibbs free energy variation, Faraday constant, cell potential, and metal ions with n charges (n = 1, 2, or 3), respectively [12]. It can be noted that Li, Na, Zn, Fe, and Cu (φLi/Li+ = −3.04 V, φNa/Na+ = −2.71 V, φZn/Zn2+ = −0.76 V, φFe/Fe3+ = −0.04 V and φCu/Cu2+ = 0.34 V vs. standard hydrogen electrode (SHE), respectively) can spontaneously reduce GO (φrGO/GO = ca. 0.6 V vs. SHE) as ∆G < 0 (Fig. 1c) [8,13]. However, Au
ACCEPTED MANUSCRIPT possesses a higher potential (φAu/Au+ = ~1.69 V vs. SHE) than that of GO, indicating ∆G > 0, thus Au cannot spontaneously reduce GO. That is the reason why Au is chosen as the current collector for M/GO battery. Besides, GO has a
RI PT
work function of –4.70 eV [14] vs. standard vacuum energy (SVE) to match well with that of Li, Na, Zn, Fe, and Cu. As a result, the outermost electrons of M can readily transfer to GO, leading to the reduction of GO and the oxidation
SC
of M (Fig. 1d) [15]. Both the thermodynamic calculation and the work function
M AN U
offer the theoretical base for that M/GO battery can spontaneously convert its
AC C
EP
TE D
inherent chemical energy into electric power.
Fig. 1. Assembly and reaction mechanism of M/GO batteries. (a) Schematic of M/GO battery formed by the simple assembly of metal foil and GO film, in which active metal and GO act as the anode and cathode, respectively. Electrons can spontaneously transfer from M to GO, and electrical current is allowed to flow in external circuit so as to convert chemical energy into electricity. (b) GO film in touch with Li, Na, Zn, Fe, or Cu can be reduced into rGO incorporated by the yielding MxOy on the basis of the redox reaction: M + GO → rGO + MxOy. (c) The standard electrode potential of M/Mn+ vs. SHE. (d) The
ACCEPTED MANUSCRIPT work function of M vs. SVE. The outermost electrons of M can spontaneously transfer to GO.
The voltage output and specific energy density are two key criteria for evaluating the performance of M/GO battery, which is related to metal activity and the oxygen
RI PT
content of GO. To investigate the relationship, five metals (i.e., Li, Na, Zn, Fe, and Cu) and two kinds of GO with different oxygen content are assembled, respectively. The
SC
O/C atomic ratios of GO are about 0.6 and 0.3; accordingly they are denoted as GO0.6 and GO0.3, respectively (Fig. S2e–h). Fig. 2a−b schematically depicts that GO0.6 has
M AN U
more oxygen-related functional groups but smaller sheet size than that of GO0.3. The atomic force microscopy (AFM) image shows the lateral size distributions of GO0.6 and GO0.3 are mainly focused on about 0.3–1.8 µm and 1.5–2.5 µm, respectively (Fig. S3). Besides, GO0.3 has less defects and better crystalline structure compared with
TE D
GO0.6, which can be certified by the corresponding X-ray diffraction (XRD) pattern (Fig. S4a) and Raman spectrum (Fig. S4b). The X-ray photoelectron spectroscopy
EP
(XPS) shows the chemical composition of GO0.3 and GO0.6, both of which are composed of C and O elements. Besides, the fitted C1s spectrum reflects the presence
AC C
of four types of carbon bonds: C–C/C=C, C–O, C=O, and O–C=O (Fig. S4c−d) [16]. Fig. 2c displays that the voltage output of Li/GO0.3 battery gradually decreases from initial ~1.8 V to 1 V within 0.7 h, which is associated with the wetting of electrolyte and the initial activation process [13]. Subsequently, the voltage output can remain stable and show extremely long platform, which is determined by the intrinsic Gibbs free energy of Li and GO0.3, and independent of other factors. As the redox reaction continues, more and more reaction products (i.e., Li2O/Li2CO3) form, resulting in the
ACCEPTED MANUSCRIPT increase of internal resistance and the decrease of current output. As expected, the current output slowly decays from initial ~970 mA cm–3 until eventually a relatively steady value of ca. 108 mA cm–3 after 5 h (Fig. 2d). By contrast, the initial voltage
RI PT
and current output of Li/GO0.6 battery can reach up to 2.6 V and ~895 mA cm–3, respectively; After 5 h, Li/GO0.6 battery can generate the voltage and current as high as ~1.8 V and ~98 mA cm–3, respectively. Different from Li/GO battery and Na/GO
SC
battery (Fig. 2e), the V-T curves of Zn/GO battery (Fig. 2f), Fe/GO battery (Fig. S5c),
M AN U
and Cu/GO battery (Fig. S5e) first increase rapidly then keep steady. The summary data about the performance of M/GO batteries are listed in Table S1. On the basis of these results, it can be concluded that: (1) as the oxygen content of GO increases, the voltage of M/GO battery increases but its current output decreases. (2) The voltage
TE D
output and specific energy density of M/GO battery follow this order: Li > Na > Zn >
AC C
EP
Fe > Cu, which is directly associated with the metal activity.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
TE D
Fig. 2. Schematic representation of GO and electrochemical performance of M/GO batteries. (a,b) GO0.6 has more oxygen-related functional groups but smaller sheet size than that of GO0.3. (c,d) Voltage and current outputs vs. time profiles of Li/GO battery. (e,f) Voltage output vs. time profiles of Na/GO battery and Zn/GO battery.
EP
After a series of electrochemical tests, M/GO batteries are disassembled to have
AC C
insight into the reaction of M and GO inside the cells. The pristine GO film within Li/GO battery can be reduced into an integrated rGO film incorporated with Li2O/Li2CO3 after discharging, as schematically depicted in Fig. 3a. The side-view SEM images show the rGO film with a thickness of ~40 µm has a layer-stacked structure (Fig. 3b–c). The top-view SEM image of rGO displays its rough surface (Fig. 3d) and lots of agglomerated nanoparticles (i.e., Li2O/Li2CO3) are coated onto rGO nanosheets (Fig. 3e–f), which can be confirmed by XPS and XRD. The corresponding
ACCEPTED MANUSCRIPT energy dispersive X-ray spectroscopy (EDS) confirms the existence of C and O elements with O/C atomic ratio of ca. 1.6 (Fig. 3g), while Li signal is undetectable. The high-resolution C1s spectrum can be deconvoluted into five peaks (Fig. 3h) and
RI PT
the newly appeared peak located at ~289.7 eV can be assigned to Li2CO3 and/or lithium alkyl carbonates (R-OCO2Li), which might be related to the decomposition of organic carbonate-based electrolyte [17]. The XRD pattern shows eight main peaks
SC
with 2θ values of ca. 21.3o, 23.4o, 30.6o, 31.8o, 37.0o, 40.0o, 48.9o and 58.4o that can
M AN U
be indexed to Li2CO3 (110), (200), (–202), (002), (–311), (310), (022) and (113) diffractions, respectively (Fig. 3i, JCPDS card No. 87-729). Another peak located at ~33.6o marked by blue circle can be indexed to Li2O (111) surface. The typical GO peak (2θ = ca. 11o) is disappeared, indicating GO is completely reduced into rGO by
TE D
Li. As expected, a pure rGO film with the O/C atomic ratio of ~0.12 is obtained after removing the Li-based components (Fig. S6). In addition, the thickness of GO film is also optimized to prevent Li/GO battery from short circuit. The experimental results
EP
show that when GO thickness reaches up to ~540 µm, Li cannot completely reduce
AC C
GO, thus enabling Li/GO battery to safely output electricity (Fig. S7).
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
TE D
Fig. 3. Characterization of reaction products of Li/GO battery. (a) Li can completely reduce GO film into an integrated rGO film decorated with Li2O/Li2CO3. (b,c) side-view and (d,e) top-view SEM images of Li2O-Li2CO3/rGO film. (g) EDS is originated from (e). (f) TEM image, (h) C 1s spectrum and (i) XRD pattern of Li2O-Li2CO3/rGO.
Different from Li/GO battery, Zn can only partially reduce GO within Zn/GO battery, thereby forming an asymmetric GO/rGO hybrid film (Fig. 4a). SEM and XPS
EP
studies of GO/rGO hybrid show its top is the typically layered (side-view) and smooth
AC C
(top-view) GO (Fig. 4b, c, h and Fig. S8b), but the bottom is compact (side-view) and rough (bottom-view) rGO (Fig. 4d, i and Fig. S8d). The cross-sectional element mapping of GO/rGO film demonstrates the gradient distribution of Zn component (Fig. 4e–g), i.e., the Zn content gradually decreases from the bottom side to the top side. That is because the bottom side of GO in touch with Zn foil can be reduced into rGO with incorporated ZnO nanoparticles having a size of ca. 10–35 nm (Fig. 4j). However, the top side, out of touch with Zn, is still fresh GO, which is not reduced by
ACCEPTED MANUSCRIPT Zn. As a result, the bottom has more Zn component than that of the top. The XRD pattern displays two main peaks with 2θ values of ~9.9o and ~26.3o (corresponding to the interlayer spacing of ~0.9 nm and ~0.34 nm) that can be assigned to GO (001) and
RI PT
rGO (002) diffractions, respectively (Fig. 4k). Although the characteristic peaks of ZnO are not found in XRD pattern, the EDS (Fig. 4j, inset) and XPS spectrum (Fig. S8c) exactly confirm the existence of ZnO, implying its amorphous state. In addition,
SC
the resultant Zn foil originating from the unpacked Zn/GO battery possesses a
M AN U
relatively smooth surface compared with the fresh Zn foil with the “bedded rock”-like morphology (Fig. S9), which can be attributed to the consumption of Zn. Interestingly, the other three batteries after discharge also exhibit similar GO/rGO hybrid structure, as shown in Fig. S10 (Na/GO battery), Fig. S11–S12 (Fe/GO battery), and Fig. S13
AC C
EP
TE D
(Cu/GO battery).
ACCEPTED MANUSCRIPT
RI PT
Fig. 4. Characterization of reaction products of Zn/GO battery. (a) The bottom of GO film in touch with Zn can be reduced into rGO incorporated by ZnO nanoparticles. However, its top out of touch with Zn cannot be reduced and still remain the status of fresh GO, thus forming an asymmetric GO/rGO hybrid film. (b–d) side-view, (h) top-view, and (i) bottom-view SEM images of GO/rGO hybrid film. (e–g) Zn, C, and O maps originated from (b). (j) TEM of ZnO/rGO and the inset is EDS acquired from (i). (k) XRD pattern of GO/rGO hybrid.
As described above, M/GO batteries enable the conversion of chemical energy into directly usable electricity by virtue of the redox reaction between M and GO. As
SC
expected, one Li/GO battery can light up a red light-emitting diode (LED) (Fig. 5a
M AN U
and Fig. S14), and three Li/GO batteries assembled in series can repeatedly charge a home-made coin cell (Fig. S15) from 0.01 V to 2.5 V (Fig. 5b–d). The photograph of the charging circuit and the corresponding equivalent circuit are shown in Fig. S16. Prior to charging, the lab-made battery is initially activated for three cycles in the
TE D
voltage range of 0.01–2.5 V and finally discharged to 0.01 V (Fig. S15g). When recharged by our Li/GO battery pack, the real-time current and voltage of the button-type battery are recorded. Fig. 5c shows three Li/GO batteries connected in
EP
series can charge the button-type battery from 0.01 V to 2.5 V within ~38 min for the
AC C
first time. Then the latter is discharged at 0.2C (1C = 175 mA g–1) and attains a specific capacity of ~262.2 mAh g–1 (Fig. 5d). Subsequently, the 2nd, 3th, 4th, 5th charging time is kept at 0.94 h, 1.08 h, 1.48 h, 1.58 h, respectively (Fig. 5c), and the corresponding discharging capacities can reach up to 245.6 mAh g–1, 239.7 mAh g–1, 227.5 mAh g–1, 210.5 mAh g–1 at 0.5C, 1C, 2C, 5C, respectively (Fig. 5d), indicating the excellent rate performance. The increase of charging time from 1st to 5th can be attributed to the successive consumption of power within the Li/GO battery pack.
ACCEPTED MANUSCRIPT Although Li/GO primary battery cannot be recharged for reuse after electricity exhaustion, it is the first time to utilize M/GO battery pack as the reliable
M AN U
SC
RI PT
electrochemical energy source to directly provide electricity.
AC C
EP
TE D
Fig. 5. Li/GO batteries pack for energy storage and 3D-Cu/GO RFB. (a) One Li/GO battery can light up a red LED. (b) Three Li/GO batteries assembled in series can repeatedly charge a lab-made coin cell without the aid of external power source and (c) the corresponding charging current and voltage curves. (d) Discharge profiles of the button-type battery at different rate. (e) Schematic illustration of the concept of semi-solid 3D-Cu/GO RFB formed by using Cu foam, GO/ionic-liquid, and Au foil as anode, cathode, and current collector, respectively. GO catholyte flows through the system during operation and can be stored in the tank at downtime. (f) The voltage and current outputs of 3D-Cu/GO RFB and 2D-Cu/GO RFB.
From the viewpoint of spatial dimension, the energy output of 2D planar M/GO
battery is significantly restricted to the interface reaction of metal and GO film. For example, Cu/GO(s) battery delivers a low specific capacity of 0.93 µAh cm–2 (about 0.31 mAh cm–3, Table S1). To address the issue, a new type of 3D-Cu/GO redox flow battery (RFB) is proposed (Fig. 5e), among which 3D Cu foam is used as anode while GO/ionic-liquid is used as cathode with Au foil as current collector. The experimental
ACCEPTED MANUSCRIPT demonstration of 3D-Cu/GO RFB is shown in Fig. S17a–b. Such a design enables GO catholyte to flow through the Cu foam, thus increasing their contact area and accelerating the reaction rate to achieve higher specific capacity. As expected,
RI PT
3D-Cu/GO RFB can deliver the voltage output of 0.22 V and specific capacity of 19.4 µAh cm–2, which is about 97 times that of 2D-Cu foil/GO RFB (Fig. 5f). In addition, it is noted that Cu foam after discharging is decorated with sporadic rGO nanosheets
SC
(Fig. S17c–d). The corresponding EDS shows the C and O elements can be attributed
Cu/GO(s) battery (Fig. S13).
M AN U
to rGO/CuO hybrid (Fig. S17e), which is in consistent with the aforementioned
Smart and tactile sensing battery with energy management function has attracted significant attention and opened up the possibility to develop next-generation
TE D
intelligent energy storage devices [18]. In addition to 3D-Cu/GO RFB, a compressible, all-solid-state, and pressure-responsive 3D-GO/Zn battery is also fabricated. It can accurately control its energy output in response to pressure stimulations without the
EP
aid of conventional battery management system [19]. The fabrication procedure of
AC C
3D-GO/Zn battery is shown in Fig. 6a–g. First, GO is coated on the backbone of a commercial polyurethane (PU) sponge through a solution dip-coating method and subsequent freeze-drying, accompanied with the color change from white to yellow (Fig. 6a, d, e). The effective attachment of GO nanosheets along the framework of PU endows the whole cathode with excellent compressibility. Second, the compressible 3D-GO sponge is mounted on Zn foil to construct all-solid-state 3D-GO/Zn battery (Fig. 6b). In the 3D-GO/Zn battery, Zn acts as anode and 3D-GO serves as both
ACCEPTED MANUSCRIPT cathode and solid-state electrolyte. Compared with liquid electrolyte, solid-state electrolyte has no leakage and volatilization issues. In terms of the redox reaction nature of M/GO batteries, electrons transfer from M to oxygen-related groups of GO
RI PT
(i.e., –OH, –C=O, and –COOH). Therefore, Zn can react with solid-state GO at the contact interface despite in the absence of liquid electrolyte. Attributing to the redox
conversion of chemical energy as described above.
SC
reaction, 3D-GO/Zn battery is able to directly output electricity originated from the
M AN U
To investigate the controllable power output behavior, 3D-GO/Zn battery is sandwiched between CR2032-type battery shells (Fig. 6f and Fig. S18a–d). The battery is not sealed, in which its lower shell is fixed to a nonconductive polyethylene terephthalate (PET) film, and the upper shell is connected to a movable PET/Au-load
TE D
(Fig. S18e). By moving up-and-down the load, the 3D-GO/Zn battery is repeatedly compressed and released to generate the desirable current output (Fig. 6b, c, g and Movie S1). Fig. 6h–j show the dynamic responses of 3D-GO/Zn battery based on
EP
different compression frequency. The 3D-GO/Zn battery is compressed to a strain (ε)
AC C
up to 10% then decompressed at the same peed, which is denoted as one cycle. The current output firstly increases from 0.14 µA g–1 to a maximum value of ~0.29 µA g–1, then returns to the initial value. During the same time of 400 s, the responsive cycles at different frequencies of 0.25 Hz, 0.5 Hz, and 1.0 Hz are ca. 118, 211, and 308, respectively; and each cycle accounts for ca. 3.4 s, 1.9 s, and 1.3 s, respectively. As a result, it can be concluded that as the compression frequency increases, the responsive time of each cycle decreases but the accomplished cycles increase. Besides,
ACCEPTED MANUSCRIPT 3D-GO/Zn battery can not only respond fast and synchronously to the applied dynamic pressure, but also keep a steady working state without any delay. Fig. 6k shows the current output of 3D-GO/Zn battery under cyclic compression/
RI PT
decompression of different strains with the same frequency of 0.5 Hz. The output current progressively increases from ca. 0.19 µA g−1 to 1.29 µA g−1 with the increased strain from 5% to 40%, respectively. The corresponding stress-strain curves of
SC
3D-GO/Zn battery reveal that the elastic deformation of 3D-GO sponge without
M AN U
mechanical failure (Fig. S19). The cyclic stability of 3D-GO/Zn battery at different strains is also tested (Fig. S20), and the consistent current output indicates the
AC C
EP
TE D
long-time stability, little hysteresis, and high reliability of 3D-GO/Zn battery.
Fig. 6. 3D-GO/Zn battery with controlled energy release. (a,d) The commercial PU sponge with 3D interwoven framework. (e,f) After coating GO, the 3D-GO/PU sponge is obtained, which is
ACCEPTED MANUSCRIPT
RI PT
subsequently mounted on Zn foil to construct all-solid-state 3D-GO/Zn battery. (b,c,g) By moving up-and-down the load, the 3D-GO/Zn battery is repeatedly compressed and released to accurately control its energy output in response to pressure stimulations. (h–j) Current output under cyclic pressing–releasing with a strain of 10% at frequencies of 0.25 Hz, 0.5 Hz, and 1.0 Hz. (k) Current output under various cyclic strains at a frequency of 0.5 Hz. (l) Current output of 3D-GO/Zn battery at a given compression condition (decompressing time T1 and pressing time T2 with a strain of 10%).
In addition to dynamic responses, the periodically static responses of 3D-GO/Zn
SC
battery are also investigated. When 3D-GO/Zn battery is kept at the decompressed state for a certain period of time (T1) and then compressed (ε = 10%) and held for T2,
M AN U
the current output shows a synchronous response to the decompression–compression without distinct drifts (Fig. 6l). It is worth noting that the current increases but the resistance decreases under compression period (Fig. S21). This can be attributed to the following two aspects: (1) the interface contact area of GO and Zn increases under
TE D
compression, thus delivering higher electricity originated from the conversion of chemical energy as described above (Fig. S22). (2) When the pressure is imposed to
EP
the battery, the GO and PU microfibers inside the 3D GO-PU sponge would begin to make contact with each other (Fig. S23), resulting in the resistence decrease of the
AC C
overall framework [20,21]. After releasing the pressure, the deformation of GO-PU sponge would recover to the initial condition.
4. Conclusions A novel type of metal/graphene oxide (M/GO) batteries is developed via the simple assembly of metal foils with GO films, including Li/GO, Na/GO, Zn/GO, Fe/GO, and Cu/GO batteries. Taking advantage of the spontaneous redox reaction
ACCEPTED MANUSCRIPT between M and GO, the chemical energy can be readily converted into the directly available electricity. Li/GO battery generates a specific capacity of ~1,572 mAh cm–3 (about 1,604 mAh g–1). In terms of 2D planar structure of M/GO batteries, their
RI PT
energy outputs are constricted to the interface reaction of metals and GO films. To address this issue, construction of 3D M/GO batteries with more active sites is needed to achieve higher energy density. Therefore, 3D-Cu foam anode and flowing
SC
GO/ionic-liquid catholyte are assembled to form a semi-solid 3D-Cu/GO redox flow
M AN U
battery (3D-Cu/GO RFB). Such a design enables GO catholyte to flow through the Cu foam, thus increasing their contact area and accelerating the reaction rate to achieve higher specific capacity. As expected, the specific capacity of 3D-Cu/GO RFB is ca. 97 times that of Cu foil/GO RFB. More importantly, a compressible, all-solid-state,
TE D
and pressure-responsive 3D-GO/Zn battery is fabricated, which can accurately control the energy output in response to pressure stimulations without the aid of conventional battery management system. It can not only respond fast and synchronously to the
EP
periodically dynamic and static pressure, but also exhibit excellent cyclic stability and
AC C
high reliability. Overall, 3D-GO/Zn battery with energy management function makes an important step towards next-generation smart energy storage devices. Beyond those demonstrated in this work, the concept of M/GO battery will be a new starting point for the design of similar electrochemical power sources and other functions.
Acknowledgements
ACCEPTED MANUSCRIPT This work was supported by NSFC (No. 21325415, 51673026), Beijing Natural Science Foundation (2152028), and Beijing Municipal Science and Technology
Appendix A. Supplementary data
RI PT
Commission (Z161100002116022), and 111 Project 807012.
Supplementary data associated with this article can be found in the online version at
SC
*********.
References
[5] [6] [7]
AC C
[8] [9]
M AN U
[4]
TE D
[3]
A. K. Geim, Science 324 (2009) 1530−1534. D. R. Dreyer, S. Park, C. W. Bielawski, R. S. Ruoff, Chem. Soc. Rev. 39 (2010) 228−240. F. Zhao, H. H. Cheng, Z. P. Zhang, L. Jiang, L. T. Qu, Adv. Mater. 27 (2015) 4351−4357. F. Zhao, Y. Liang, H. H. Cheng, L. Jiang, L. T. Qu, Energ. Environ. Sci. 9 (2016) 912−916. Y. Jiang, H. Shao, C. Li, T. Xu, Y. Zhao, G. Q. Shi, L. Jiang, L. T. Qu, Adv. Mater. 28 (2016) 10287−10292. Y. Zhao, C. G. Hu, Y. Hu, H. H. Cheng, G. Q. Shi, L. T. Qu, Angew. Chem. Int. Ed. 51 (2012) 11371−11375. H. H. Cheng, M. H. Ye, F. Zhao, C. G. Hu, Y. Zhao, Y. Liang, N. Chen, S. L. Chen, L. Jiang, L. T. Qu, Adv. Mater. 28 (2016) 3305−3312. M. H. Ye, C. X. Li, Y. Zhao, L. T. Qu, Carbon 106 (2016) 9−19. C. G. Hu, G. P. Zheng, F. Zhao, H. B. Shao, Z. P. Zhang, N. Chen, L. Jiang, L. T. Qu, Energ. Environ. Sci. 7 (2014) 3699−3708. C. G. Hu, X. Q. Zhai, L. L. Liu, Y. Zhao, L. Jiang, L. T. Qu, Sci. Rep. 3 (2013) 2065. M. H. Ye, Z. L. Dong, C. G. Hu, H. H. Cheng, H. B. Shao, N. Chen, L. T. Qu, Small 10 (2014) 5035−5041. X. B. Cao, D. P. Qi, S. Y. Yin, J. Bu, F. J. Li, C. F. Goh, S. Zhang, X. D. Chen, Adv. Mater. 25 (2013) 2957−2962. M. H. Ye, H. H. Cheng, J. Gao, C. X. Li, L. T. Qu, J. Mater. Chem. A 4 (2016) 19154−19159. J. Liu, M. Durstock, L. M. Dai, Energ. Environ. Sci. 7 (2016) 1297−1306. L. X. Wang, F. Zhao, Q. Han, C. G. Hu, L. X. Lv, N. Chen, L. T. Qu, Nanoscale 7 (2015) 9694−9702.
EP
[1] [2]
[10] [11] [12] [13] [14] [15]
ACCEPTED MANUSCRIPT
[17] [18] [19] [20] [21]
AC C
EP
TE D
M AN U
SC
[22]
Y. N. Meng, Y. Zhao, C. G. Hu, H. H. Cheng, Y. Hu, Z. P. Zhang, G. Q. Shi, L. T. Qu, Adv. Mater. 25 (2013) 2326−2331. R. Dedryvere, S. Laruelle, S. Grugeon, L. Gireaud, J. M. Tarascon, D. Gonbeau, J. Electrochem. Soc. 152 (2005) A689−A696. X. P. Wang, J. Gao, Z. H. Cheng, N. Chen, L. T. Qu, Angew. Chem. Int. Ed. 55 (2016) 14643−14647. D. Lin, Y. Liu, Y. Cui, Nat. Nanotechnol. 12 (2017) 194−206. P. P. Zhang, L. X. Lv, Y. Liang, J. Li, H. Cheng, Y. Zhao, L. T. Qu, J. Mater. Chem. A 4 (2016) 10118−10123. L. Z. Sheng, Y. Liang, L. L. Jiang, Q. Wang, T. Wei, L. T. Qu, Z. J. Fan, Adv. Funct. Mater. 25 (2015) 6545−6551. J. Chen, Y. Li, L. Huang, N. Jia, C. Li, G. Q. Shi, Adv. Mater. 27 (2015) 3654−3660.
RI PT
[16]
S0008-6223(17)30946-6
DOI:
10.1016/j.carbon.2017.09.070
Reference:
CARBON 12400
To appear in:
Carbon
Received Date: 26 July 2017 Revised Date:
4 September 2017
Accepted Date: 16 September 2017
Please cite this article as: M. Ye, J. Gao, Y. Xiao, T. Xu, Y. Zhao, L. Qu, Metal/graphene oxide batteries, Carbon (2017), doi: 10.1016/j.carbon.2017.09.070. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Metal/graphene oxide batteries Minghui Ye, Jian Gao, Yukun Xiao, Tong Xu, Yang Zhao, Liangti Qu*
M AN U
SC
RI PT
Beijing Key Laboratory of Photoelectronic/Electrophotonic Conversion Materials, Key Laboratory of Cluster Science, Ministry of Education of China, Department of Chemistry, Beijing Institute of Technology, Beijing 100081, P. R. China. Email: [email protected] Fax: 86-10-68918608
TE D
Metal/graphene oxide (M/GO) batteries are demonstrated by the simple assembly of metals and GO films, in which the spontaneous redox reaction enables the conversion of chemical energy into electricity. Beyond the 2D planar structure, semi-solid
EP
3D-Gu/GO redox flow battery and smart pressure-responsive 3D-GO/Zn battery with
AC C
controllable energy release are developed.
ACCEPTED MANUSCRIPT
Metal/graphene oxide batteries Minghui Ye, Jian Gao, Yukun Xiao, Tong Xu, Yang Zhao, Liangti Qu*
RI PT
Beijing Key Laboratory of Photoelectronic/Electrophotonic Conversion Materials, Key Laboratory of Cluster Science, Ministry of Education of China, Department of Chemistry, Beijing Institute of Technology, Beijing 100081, P. R. China. Email: [email protected] Fax: 86-10-68918608
Abstract: Based on the spontaneous redox reaction between metal and graphene
SC
oxide, a novel type of metal/graphene oxide (M/GO) batteries is developed to convert chemical energy into electricity, including Li/GO, Na/GO, Zn/GO, Fe/GO, and
M AN U
Cu/GO batteries. They are fabricated via the simple assembly of metal foils with GO films, in which M plays the role of anode, and GO acts as both cathode and separator. Among them, Li/GO battery generates the highest specific capacity of 1,572 mAh cm–3 (about 1,604 mAh g–1). The energy density of M/GO battery is determined by
TE D
the contact area of M with GO. Therefore, three-dimensional (3D) M/GO battery will deliver higher energy in comparison to 2D planar M/GO battery. As expected, a
EP
semi-solid 3D-Cu/GO redox flow battery (RFB) is assembled by 3D Cu foam with flowing GO/ionic-liquid catholyte. Its specific capacity is ca. 97 times that of Cu
AC C
foil/GO RFB. Besides, a compressible, all-solid-state, and pressure-responsive 3D-GO/Zn battery is also fabricated. It can accurately control the energy output in response to pressure stimulations without the aid of conventional battery management system. Beyond those demonstrated in this work, the concept of M/GO battery will shed light on the design of similar electrochemical power sources.
1. Introduction
ACCEPTED MANUSCRIPT Graphene is a two-dimensional (2D) honeycomb lattice of sp2-bonded carbon atoms [1]. Graphene oxide (GO) is a graphene derivative with abundant oxygen-related functional groups attaching to its basal plane and sheet edge, such as
RI PT
hydroxyl, epoxy, and carboxyl groups [2]. It can be cheaply obtained in large quantities via the chemical exfoliation of graphite [3–7]. GO could be reduced by active metal, such as zinc (Zn) [8,9], ferrum (Fe) [10], and copper (Cu) [11], upon
SC
inserting metal foil into aqueous GO dispersion at ambient condition. As illustrated in
M AN U
our previous work [8–11], the contact of GO with metal induced the redox reaction: M + GO → rGO + MxOy, where rGO and MxOy were reduced graphene oxide and the resultant metal oxide, respectively. Electrons could spontaneously transfer from metal to GO, leading to the reduction of GO and the oxidation of metal. Based on the redox
TE D
reaction between active metal and GO, a series of metal/GO galvanic batteries are developed by the simple assembly of active metal foils and solid-state GO(s) films. They can directly output available electricity resulted from the conversion of chemical
EP
energy. The active metals can be further extended to lithium (Li) and sodium (Na).
AC C
There are no separation membranes existing in those M/GO batteries. This is because GO(s) acts as both cathode and separator to avoid the short circuit of the battery due to its insulating nature. Meanwhile, M plays the role of anode in offering electrons, so that electrical current is allowed to flow in external circuit. As expected, Li/GO battery generates a specific capacity of ~1,572 mAh cm–3 (corresponding to 1,604 mAh g–1). Beyond the 2D planar structure of M/GO battery, constructing 3D M/GO battery with higher energy density and controlled energy release is very important and
ACCEPTED MANUSCRIPT significative. Accordingly, a semi-solid 3D-Cu/GO redox flow battery (3D-Cu/GO RFB) is achieved by using Cu foam as anode and flowing GO/ionic-liquid as cathode, whose specific capacity is ca. 97 times that of Cu foil/GO RFB. Besides, a
RI PT
compressible, all-solid-state, and pressure-responsive 3D-GO/Zn battery is also developed, in which GO-coated sponge and Zn foil serve as cathode and anode, respectively. The 3D-GO/Zn battery can be repeatedly compressed and released to
M AN U
of conventional battery management system.
SC
accurately control its energy output in response to pressure stimulations without need
2. Experimental
2.1 The fabrication of GO0.3 and GO0.6
TE D
The aqueous GO dispersion was prepared through the oxidation exfoliation of graphite according to the modified Hummer’s method [22]. In this context, two kinds of GO with O/C atomic ratios of 0.3 and 0.6 were denoted as GO0.3 and GO0.6,
EP
respectively. The preparation of GO0.3 was as follows: First, 80 mL concentrated
AC C
H2SO4 was added to 250 mL flask filled with 3 g graphite in an ice bath with mechanical agitation, and then 3 g NaNO3 and 9 g KMnO4 were slowly added, respectively. Second, the flask equipped with condensing system was transferred to 36 o
C oil bath and vigorously stirred for 1 h, followed by the addition of 150 mL
deionized water. Successively, the reaction system was heated up to 90 oC and maintained for 20 min, followed by the dropwise addition of 5 mL H2O2 (30%). Third, the suspension was cooled to room temperature, which was then filtered and washed
ACCEPTED MANUSCRIPT with about 1.2 M HCl aqueous solution to remove metal impurity. The obtained solid was further filtered by adding deionized water and then dispersed in 400 mL water, followed by stirring overnight. Finally, the as-produced GO dispersion was
RI PT
centrifuged at 10,000 r.p.m. for 20 min to remove the unexfoliated particles, followed by the dialysis in deionized water for one week using dialysis bags.
The preparation of GO0.6 was identical to that of GO0.3 except changing the dose
SC
of KMnO4 into 18 g and prolonging the oxidized time to 16 h at a higher temperature
M AN U
of 50 oC.
2.2 The fabrication of M/GO batteries
GO film with a density of ~0.98 g cm–3 was obtained by the freeze-drying of
TE D
aqueous GO dispersion and subsequent tableting with a pressure of ca. 20 MPa. The thickness of GO film was about 30 µm. The M foil (i.e., Li, Na, Zn, Fe, Cu) and GO film were sandwiched between the lower battery shell and stainless steel (SS) foil
EP
plated with Au, then encapsulated into CR2032-type coin cells in Ar-filled glove box
AC C
(Fig. S1). The electrolytes for Li/GO battery and Na/GO battery were 1 M LiPF6 dissolved in ethylene carbonate (EC)/dimethyl carbonate (DMC)/diethyl carbonate (DEC) (1:1:1 in volume), and 1 M NaPF6 in a 1:2 (v/v) mixture of EC and DEC, respectively.
Zn/GO
battery,
Fe/GO
battery,
and
Cu/GO
battery
used
1-butyl-3-mthylimidazolium tetrafluoroborate (BMIMBF4) as electrolyte. The voltage and current outputs were recorded by a Keithley 2400 sourcemeter. The circuit parameters of the open circuit voltage test were current = 0 mA and step index = 5
ACCEPTED MANUSCRIPT points per s. The circuit parameters of the short circuit current test were voltage = 0 mV and step index = 5 points per s. The specific capacity of M/GO battery was calculated from I-T curve (Fig. 2d and Fig. S5a, b, d, f) by using the following
C =
∫ idt
RI PT
formula:
where C was the volume capacity (mAh cm–3), i was the discharging current density
M AN U
SC
(mA cm–3) based on the volume of GO film, and dt was the discharging time (h) [13].
2.3 The fabrication of 3D-Cu/GO RFB
The catholyte was obtained via adding freeze-drying GO monoliths into BMIMBF4 to form 2 mg mL–1 dispersion, and then poured into a beaker. Afterward, a
TE D
piece of fresh Cu foam was inserted into catholyte with Au foil as the current collector to construct a two-electrode system (Fig. S17b). A magnetic stirring bar was placed into the beaker to enable GO catholyte to flow when the RFB was in operation. The
AC C
EP
electrical signals were recorded by Keithley 2400 sourcemeter.
2.4 The fabrication of 3D-GO/Zn battery Commercial polyurethane (PU) sponge was cut into cuboids and then immersed
into 2 mg mL–1 GO solution by a squeezing procedure. Subsequently, the lyophilization was carried out to obtain GO-coated PU sponge using liquid nitrogen, whose density was about 16.1 mg cm−3. GO accounted for ~16% of the whole sponge weight. The 3D-GO sponge was mounted on the fresh Zn foil to form 3D-GO/Zn
ACCEPTED MANUSCRIPT battery in the absence of electrolyte (Fig. S18). To investigate the controllable power output behavior, the 3D-GO/Zn battery was sandwiched in a CR2032-type coin cell without encapsulation. The lower battery shell laminated to a nonconductive
RI PT
polyethylene terephthalate (PET) film was fixed to the pressure testing system, while
compression/decompression cycles.
M AN U
2.5 The fabrication of lab-made coin battery
SC
the upper shell with PET film was able to move up and down during the
The home-built coin battery was composed of Li2TiO3 (LTO), Li foil, and commercial separator membrane (celgard 2400 microporous polypropylene membrane). The LTO electrode was prepared by mixing Li2TiO3, acetylene black, and
TE D
polyvinylidene fluoride (PVDF) binder with weight ratios of 8:1:1 into N-methylpyrrolidone (NMP) to form monodispersed slurry. Then the as-produced slurry was pasted onto a Cu foil and dried at 120 oC for 12 h under vacuum. For one
EP
electrode, it typically had a loading of ~0.65 mg cm–2. The electrolyte was 1.0 M
AC C
LiPF6 in ethylene carbonate/dimethyl carbonate/diethyl carbonate (EC: DMC: DEC = 1:1:1 in volume). The assembly of coin cell was performed in an argon-filled glove box, where moisture and oxygen concentrations were strictly limited below 1.0 ppm. The cells were then aged for 12 h before measurement. The galvanostatic discharge/charge test was carried out over a voltage range of 0.01–2.5 V versus Li+/Li on the LAND CT2001A battery system at room temperature. Electrochemical impedance spectral measurements were carried out in the frequency range from 100
ACCEPTED MANUSCRIPT kHz to 0.01 Hz on CHI760D electrochemical workstation.
2.6 Characterization
RI PT
Field-emission scanning electron microscopy (FE-SEM) and X-ray energy dispersive spectroscopy (EDS) of the samples were carried out on a JSM-7001F SEM unit (Japan Electron Optics Laboratory Co., Ltd, Japan). Transmission electron
SC
microscopy (TEM) and high-resolution TEM (HR-TEM) were conducted using a
High-angle annular dark
field
M AN U
TecnaiG2 20ST (T20) at an acceleration voltage of 120 kV (FEI corporation, USA). scanning transmission
electron
microscopy
(HAADF-STEM) was operated at 200 KV (FEI, Technai G2 F30). X-ray photograph electron spectroscopy (XPS) data were carried on an ESCALab220i-XL electron
TE D
spectrometer using 300 W AlK irradiation (VG Scientific Co., Ltd, UK). The base pressure was about 3 × 10–9 mbar. Powder X-ray diffraction (XRD) patterns were performed on a Bruker D8 advance X-ray diffractometer (Bruker Corporation,
EP
Germany) using a Netherlands 1,710 diffractometer with a Cu Kα irradiation source
AC C
(λ = l.54 Å) and a self-calibration process was performed with a SiO2 internal standard sample prior to target measurement. Raman spectra were recorded using a RM 2000 Microscopic Confocal Raman Spectrometer (Renishaw PLC, England) with a 514 nm laser. The repeated compression/decompression tests of 3D-GO/Zn battery were conducted with an Instron material testing system (Instron 3342), and the current variation was recorded in real time using a Keithley 2612A sourcemeter, which was controlled by a LabView-based data acquisition system.
ACCEPTED MANUSCRIPT
3. Results and discussion Fig. 1a schematically illustrates the M/GO batteries formed by the simple
RI PT
assembly of active metal (i.e., Li, Na, Zn, Fe, and Cu) foils and GO films into button-type cells without any separators (Fig. S1). GO film with a thickness of ~30 µm (Fig. S2a–d) can be obtained by the freeze-drying of aqueous GO dispersion and
SC
subsequent tableting. Both shells of the battery are plated with aurum (Au) to act as
M AN U
the current collector and its encapsulation is performed at argon-filled glove box. In the designed M/GO battery, active metal and GO(s) act as the anode and cathode, respectively. Electrical current is allowed to flow in external circuit along with the redox reaction of two electrodes, thus converting chemical energy into electricity. It is
TE D
worth noting that GO film in touch with Li, Na, Zn, Fe, and Cu is reduced into rGO and the yielded MxOy (i.e., Li2O/Li2CO3, Na2O/Na2CO3, ZnO, Fe3O4, and CuO, respectively) are anchored onto rGO nanosheets (Fig. 1b). According to the
EP
relationship between the thermodynamic function and the cell potential as follows:
AC C
∆G = –nFE = –nF[φrGO/GO –φM/Mn+]
Where, ∆G, F, E, and Mn+ are the Gibbs free energy variation, Faraday constant, cell potential, and metal ions with n charges (n = 1, 2, or 3), respectively [12]. It can be noted that Li, Na, Zn, Fe, and Cu (φLi/Li+ = −3.04 V, φNa/Na+ = −2.71 V, φZn/Zn2+ = −0.76 V, φFe/Fe3+ = −0.04 V and φCu/Cu2+ = 0.34 V vs. standard hydrogen electrode (SHE), respectively) can spontaneously reduce GO (φrGO/GO = ca. 0.6 V vs. SHE) as ∆G < 0 (Fig. 1c) [8,13]. However, Au
ACCEPTED MANUSCRIPT possesses a higher potential (φAu/Au+ = ~1.69 V vs. SHE) than that of GO, indicating ∆G > 0, thus Au cannot spontaneously reduce GO. That is the reason why Au is chosen as the current collector for M/GO battery. Besides, GO has a
RI PT
work function of –4.70 eV [14] vs. standard vacuum energy (SVE) to match well with that of Li, Na, Zn, Fe, and Cu. As a result, the outermost electrons of M can readily transfer to GO, leading to the reduction of GO and the oxidation
SC
of M (Fig. 1d) [15]. Both the thermodynamic calculation and the work function
M AN U
offer the theoretical base for that M/GO battery can spontaneously convert its
AC C
EP
TE D
inherent chemical energy into electric power.
Fig. 1. Assembly and reaction mechanism of M/GO batteries. (a) Schematic of M/GO battery formed by the simple assembly of metal foil and GO film, in which active metal and GO act as the anode and cathode, respectively. Electrons can spontaneously transfer from M to GO, and electrical current is allowed to flow in external circuit so as to convert chemical energy into electricity. (b) GO film in touch with Li, Na, Zn, Fe, or Cu can be reduced into rGO incorporated by the yielding MxOy on the basis of the redox reaction: M + GO → rGO + MxOy. (c) The standard electrode potential of M/Mn+ vs. SHE. (d) The
ACCEPTED MANUSCRIPT work function of M vs. SVE. The outermost electrons of M can spontaneously transfer to GO.
The voltage output and specific energy density are two key criteria for evaluating the performance of M/GO battery, which is related to metal activity and the oxygen
RI PT
content of GO. To investigate the relationship, five metals (i.e., Li, Na, Zn, Fe, and Cu) and two kinds of GO with different oxygen content are assembled, respectively. The
SC
O/C atomic ratios of GO are about 0.6 and 0.3; accordingly they are denoted as GO0.6 and GO0.3, respectively (Fig. S2e–h). Fig. 2a−b schematically depicts that GO0.6 has
M AN U
more oxygen-related functional groups but smaller sheet size than that of GO0.3. The atomic force microscopy (AFM) image shows the lateral size distributions of GO0.6 and GO0.3 are mainly focused on about 0.3–1.8 µm and 1.5–2.5 µm, respectively (Fig. S3). Besides, GO0.3 has less defects and better crystalline structure compared with
TE D
GO0.6, which can be certified by the corresponding X-ray diffraction (XRD) pattern (Fig. S4a) and Raman spectrum (Fig. S4b). The X-ray photoelectron spectroscopy
EP
(XPS) shows the chemical composition of GO0.3 and GO0.6, both of which are composed of C and O elements. Besides, the fitted C1s spectrum reflects the presence
AC C
of four types of carbon bonds: C–C/C=C, C–O, C=O, and O–C=O (Fig. S4c−d) [16]. Fig. 2c displays that the voltage output of Li/GO0.3 battery gradually decreases from initial ~1.8 V to 1 V within 0.7 h, which is associated with the wetting of electrolyte and the initial activation process [13]. Subsequently, the voltage output can remain stable and show extremely long platform, which is determined by the intrinsic Gibbs free energy of Li and GO0.3, and independent of other factors. As the redox reaction continues, more and more reaction products (i.e., Li2O/Li2CO3) form, resulting in the
ACCEPTED MANUSCRIPT increase of internal resistance and the decrease of current output. As expected, the current output slowly decays from initial ~970 mA cm–3 until eventually a relatively steady value of ca. 108 mA cm–3 after 5 h (Fig. 2d). By contrast, the initial voltage
RI PT
and current output of Li/GO0.6 battery can reach up to 2.6 V and ~895 mA cm–3, respectively; After 5 h, Li/GO0.6 battery can generate the voltage and current as high as ~1.8 V and ~98 mA cm–3, respectively. Different from Li/GO battery and Na/GO
SC
battery (Fig. 2e), the V-T curves of Zn/GO battery (Fig. 2f), Fe/GO battery (Fig. S5c),
M AN U
and Cu/GO battery (Fig. S5e) first increase rapidly then keep steady. The summary data about the performance of M/GO batteries are listed in Table S1. On the basis of these results, it can be concluded that: (1) as the oxygen content of GO increases, the voltage of M/GO battery increases but its current output decreases. (2) The voltage
TE D
output and specific energy density of M/GO battery follow this order: Li > Na > Zn >
AC C
EP
Fe > Cu, which is directly associated with the metal activity.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
TE D
Fig. 2. Schematic representation of GO and electrochemical performance of M/GO batteries. (a,b) GO0.6 has more oxygen-related functional groups but smaller sheet size than that of GO0.3. (c,d) Voltage and current outputs vs. time profiles of Li/GO battery. (e,f) Voltage output vs. time profiles of Na/GO battery and Zn/GO battery.
EP
After a series of electrochemical tests, M/GO batteries are disassembled to have
AC C
insight into the reaction of M and GO inside the cells. The pristine GO film within Li/GO battery can be reduced into an integrated rGO film incorporated with Li2O/Li2CO3 after discharging, as schematically depicted in Fig. 3a. The side-view SEM images show the rGO film with a thickness of ~40 µm has a layer-stacked structure (Fig. 3b–c). The top-view SEM image of rGO displays its rough surface (Fig. 3d) and lots of agglomerated nanoparticles (i.e., Li2O/Li2CO3) are coated onto rGO nanosheets (Fig. 3e–f), which can be confirmed by XPS and XRD. The corresponding
ACCEPTED MANUSCRIPT energy dispersive X-ray spectroscopy (EDS) confirms the existence of C and O elements with O/C atomic ratio of ca. 1.6 (Fig. 3g), while Li signal is undetectable. The high-resolution C1s spectrum can be deconvoluted into five peaks (Fig. 3h) and
RI PT
the newly appeared peak located at ~289.7 eV can be assigned to Li2CO3 and/or lithium alkyl carbonates (R-OCO2Li), which might be related to the decomposition of organic carbonate-based electrolyte [17]. The XRD pattern shows eight main peaks
SC
with 2θ values of ca. 21.3o, 23.4o, 30.6o, 31.8o, 37.0o, 40.0o, 48.9o and 58.4o that can
M AN U
be indexed to Li2CO3 (110), (200), (–202), (002), (–311), (310), (022) and (113) diffractions, respectively (Fig. 3i, JCPDS card No. 87-729). Another peak located at ~33.6o marked by blue circle can be indexed to Li2O (111) surface. The typical GO peak (2θ = ca. 11o) is disappeared, indicating GO is completely reduced into rGO by
TE D
Li. As expected, a pure rGO film with the O/C atomic ratio of ~0.12 is obtained after removing the Li-based components (Fig. S6). In addition, the thickness of GO film is also optimized to prevent Li/GO battery from short circuit. The experimental results
EP
show that when GO thickness reaches up to ~540 µm, Li cannot completely reduce
AC C
GO, thus enabling Li/GO battery to safely output electricity (Fig. S7).
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
TE D
Fig. 3. Characterization of reaction products of Li/GO battery. (a) Li can completely reduce GO film into an integrated rGO film decorated with Li2O/Li2CO3. (b,c) side-view and (d,e) top-view SEM images of Li2O-Li2CO3/rGO film. (g) EDS is originated from (e). (f) TEM image, (h) C 1s spectrum and (i) XRD pattern of Li2O-Li2CO3/rGO.
Different from Li/GO battery, Zn can only partially reduce GO within Zn/GO battery, thereby forming an asymmetric GO/rGO hybrid film (Fig. 4a). SEM and XPS
EP
studies of GO/rGO hybrid show its top is the typically layered (side-view) and smooth
AC C
(top-view) GO (Fig. 4b, c, h and Fig. S8b), but the bottom is compact (side-view) and rough (bottom-view) rGO (Fig. 4d, i and Fig. S8d). The cross-sectional element mapping of GO/rGO film demonstrates the gradient distribution of Zn component (Fig. 4e–g), i.e., the Zn content gradually decreases from the bottom side to the top side. That is because the bottom side of GO in touch with Zn foil can be reduced into rGO with incorporated ZnO nanoparticles having a size of ca. 10–35 nm (Fig. 4j). However, the top side, out of touch with Zn, is still fresh GO, which is not reduced by
ACCEPTED MANUSCRIPT Zn. As a result, the bottom has more Zn component than that of the top. The XRD pattern displays two main peaks with 2θ values of ~9.9o and ~26.3o (corresponding to the interlayer spacing of ~0.9 nm and ~0.34 nm) that can be assigned to GO (001) and
RI PT
rGO (002) diffractions, respectively (Fig. 4k). Although the characteristic peaks of ZnO are not found in XRD pattern, the EDS (Fig. 4j, inset) and XPS spectrum (Fig. S8c) exactly confirm the existence of ZnO, implying its amorphous state. In addition,
SC
the resultant Zn foil originating from the unpacked Zn/GO battery possesses a
M AN U
relatively smooth surface compared with the fresh Zn foil with the “bedded rock”-like morphology (Fig. S9), which can be attributed to the consumption of Zn. Interestingly, the other three batteries after discharge also exhibit similar GO/rGO hybrid structure, as shown in Fig. S10 (Na/GO battery), Fig. S11–S12 (Fe/GO battery), and Fig. S13
AC C
EP
TE D
(Cu/GO battery).
ACCEPTED MANUSCRIPT
RI PT
Fig. 4. Characterization of reaction products of Zn/GO battery. (a) The bottom of GO film in touch with Zn can be reduced into rGO incorporated by ZnO nanoparticles. However, its top out of touch with Zn cannot be reduced and still remain the status of fresh GO, thus forming an asymmetric GO/rGO hybrid film. (b–d) side-view, (h) top-view, and (i) bottom-view SEM images of GO/rGO hybrid film. (e–g) Zn, C, and O maps originated from (b). (j) TEM of ZnO/rGO and the inset is EDS acquired from (i). (k) XRD pattern of GO/rGO hybrid.
As described above, M/GO batteries enable the conversion of chemical energy into directly usable electricity by virtue of the redox reaction between M and GO. As
SC
expected, one Li/GO battery can light up a red light-emitting diode (LED) (Fig. 5a
M AN U
and Fig. S14), and three Li/GO batteries assembled in series can repeatedly charge a home-made coin cell (Fig. S15) from 0.01 V to 2.5 V (Fig. 5b–d). The photograph of the charging circuit and the corresponding equivalent circuit are shown in Fig. S16. Prior to charging, the lab-made battery is initially activated for three cycles in the
TE D
voltage range of 0.01–2.5 V and finally discharged to 0.01 V (Fig. S15g). When recharged by our Li/GO battery pack, the real-time current and voltage of the button-type battery are recorded. Fig. 5c shows three Li/GO batteries connected in
EP
series can charge the button-type battery from 0.01 V to 2.5 V within ~38 min for the
AC C
first time. Then the latter is discharged at 0.2C (1C = 175 mA g–1) and attains a specific capacity of ~262.2 mAh g–1 (Fig. 5d). Subsequently, the 2nd, 3th, 4th, 5th charging time is kept at 0.94 h, 1.08 h, 1.48 h, 1.58 h, respectively (Fig. 5c), and the corresponding discharging capacities can reach up to 245.6 mAh g–1, 239.7 mAh g–1, 227.5 mAh g–1, 210.5 mAh g–1 at 0.5C, 1C, 2C, 5C, respectively (Fig. 5d), indicating the excellent rate performance. The increase of charging time from 1st to 5th can be attributed to the successive consumption of power within the Li/GO battery pack.
ACCEPTED MANUSCRIPT Although Li/GO primary battery cannot be recharged for reuse after electricity exhaustion, it is the first time to utilize M/GO battery pack as the reliable
M AN U
SC
RI PT
electrochemical energy source to directly provide electricity.
AC C
EP
TE D
Fig. 5. Li/GO batteries pack for energy storage and 3D-Cu/GO RFB. (a) One Li/GO battery can light up a red LED. (b) Three Li/GO batteries assembled in series can repeatedly charge a lab-made coin cell without the aid of external power source and (c) the corresponding charging current and voltage curves. (d) Discharge profiles of the button-type battery at different rate. (e) Schematic illustration of the concept of semi-solid 3D-Cu/GO RFB formed by using Cu foam, GO/ionic-liquid, and Au foil as anode, cathode, and current collector, respectively. GO catholyte flows through the system during operation and can be stored in the tank at downtime. (f) The voltage and current outputs of 3D-Cu/GO RFB and 2D-Cu/GO RFB.
From the viewpoint of spatial dimension, the energy output of 2D planar M/GO
battery is significantly restricted to the interface reaction of metal and GO film. For example, Cu/GO(s) battery delivers a low specific capacity of 0.93 µAh cm–2 (about 0.31 mAh cm–3, Table S1). To address the issue, a new type of 3D-Cu/GO redox flow battery (RFB) is proposed (Fig. 5e), among which 3D Cu foam is used as anode while GO/ionic-liquid is used as cathode with Au foil as current collector. The experimental
ACCEPTED MANUSCRIPT demonstration of 3D-Cu/GO RFB is shown in Fig. S17a–b. Such a design enables GO catholyte to flow through the Cu foam, thus increasing their contact area and accelerating the reaction rate to achieve higher specific capacity. As expected,
RI PT
3D-Cu/GO RFB can deliver the voltage output of 0.22 V and specific capacity of 19.4 µAh cm–2, which is about 97 times that of 2D-Cu foil/GO RFB (Fig. 5f). In addition, it is noted that Cu foam after discharging is decorated with sporadic rGO nanosheets
SC
(Fig. S17c–d). The corresponding EDS shows the C and O elements can be attributed
Cu/GO(s) battery (Fig. S13).
M AN U
to rGO/CuO hybrid (Fig. S17e), which is in consistent with the aforementioned
Smart and tactile sensing battery with energy management function has attracted significant attention and opened up the possibility to develop next-generation
TE D
intelligent energy storage devices [18]. In addition to 3D-Cu/GO RFB, a compressible, all-solid-state, and pressure-responsive 3D-GO/Zn battery is also fabricated. It can accurately control its energy output in response to pressure stimulations without the
EP
aid of conventional battery management system [19]. The fabrication procedure of
AC C
3D-GO/Zn battery is shown in Fig. 6a–g. First, GO is coated on the backbone of a commercial polyurethane (PU) sponge through a solution dip-coating method and subsequent freeze-drying, accompanied with the color change from white to yellow (Fig. 6a, d, e). The effective attachment of GO nanosheets along the framework of PU endows the whole cathode with excellent compressibility. Second, the compressible 3D-GO sponge is mounted on Zn foil to construct all-solid-state 3D-GO/Zn battery (Fig. 6b). In the 3D-GO/Zn battery, Zn acts as anode and 3D-GO serves as both
ACCEPTED MANUSCRIPT cathode and solid-state electrolyte. Compared with liquid electrolyte, solid-state electrolyte has no leakage and volatilization issues. In terms of the redox reaction nature of M/GO batteries, electrons transfer from M to oxygen-related groups of GO
RI PT
(i.e., –OH, –C=O, and –COOH). Therefore, Zn can react with solid-state GO at the contact interface despite in the absence of liquid electrolyte. Attributing to the redox
conversion of chemical energy as described above.
SC
reaction, 3D-GO/Zn battery is able to directly output electricity originated from the
M AN U
To investigate the controllable power output behavior, 3D-GO/Zn battery is sandwiched between CR2032-type battery shells (Fig. 6f and Fig. S18a–d). The battery is not sealed, in which its lower shell is fixed to a nonconductive polyethylene terephthalate (PET) film, and the upper shell is connected to a movable PET/Au-load
TE D
(Fig. S18e). By moving up-and-down the load, the 3D-GO/Zn battery is repeatedly compressed and released to generate the desirable current output (Fig. 6b, c, g and Movie S1). Fig. 6h–j show the dynamic responses of 3D-GO/Zn battery based on
EP
different compression frequency. The 3D-GO/Zn battery is compressed to a strain (ε)
AC C
up to 10% then decompressed at the same peed, which is denoted as one cycle. The current output firstly increases from 0.14 µA g–1 to a maximum value of ~0.29 µA g–1, then returns to the initial value. During the same time of 400 s, the responsive cycles at different frequencies of 0.25 Hz, 0.5 Hz, and 1.0 Hz are ca. 118, 211, and 308, respectively; and each cycle accounts for ca. 3.4 s, 1.9 s, and 1.3 s, respectively. As a result, it can be concluded that as the compression frequency increases, the responsive time of each cycle decreases but the accomplished cycles increase. Besides,
ACCEPTED MANUSCRIPT 3D-GO/Zn battery can not only respond fast and synchronously to the applied dynamic pressure, but also keep a steady working state without any delay. Fig. 6k shows the current output of 3D-GO/Zn battery under cyclic compression/
RI PT
decompression of different strains with the same frequency of 0.5 Hz. The output current progressively increases from ca. 0.19 µA g−1 to 1.29 µA g−1 with the increased strain from 5% to 40%, respectively. The corresponding stress-strain curves of
SC
3D-GO/Zn battery reveal that the elastic deformation of 3D-GO sponge without
M AN U
mechanical failure (Fig. S19). The cyclic stability of 3D-GO/Zn battery at different strains is also tested (Fig. S20), and the consistent current output indicates the
AC C
EP
TE D
long-time stability, little hysteresis, and high reliability of 3D-GO/Zn battery.
Fig. 6. 3D-GO/Zn battery with controlled energy release. (a,d) The commercial PU sponge with 3D interwoven framework. (e,f) After coating GO, the 3D-GO/PU sponge is obtained, which is
ACCEPTED MANUSCRIPT
RI PT
subsequently mounted on Zn foil to construct all-solid-state 3D-GO/Zn battery. (b,c,g) By moving up-and-down the load, the 3D-GO/Zn battery is repeatedly compressed and released to accurately control its energy output in response to pressure stimulations. (h–j) Current output under cyclic pressing–releasing with a strain of 10% at frequencies of 0.25 Hz, 0.5 Hz, and 1.0 Hz. (k) Current output under various cyclic strains at a frequency of 0.5 Hz. (l) Current output of 3D-GO/Zn battery at a given compression condition (decompressing time T1 and pressing time T2 with a strain of 10%).
In addition to dynamic responses, the periodically static responses of 3D-GO/Zn
SC
battery are also investigated. When 3D-GO/Zn battery is kept at the decompressed state for a certain period of time (T1) and then compressed (ε = 10%) and held for T2,
M AN U
the current output shows a synchronous response to the decompression–compression without distinct drifts (Fig. 6l). It is worth noting that the current increases but the resistance decreases under compression period (Fig. S21). This can be attributed to the following two aspects: (1) the interface contact area of GO and Zn increases under
TE D
compression, thus delivering higher electricity originated from the conversion of chemical energy as described above (Fig. S22). (2) When the pressure is imposed to
EP
the battery, the GO and PU microfibers inside the 3D GO-PU sponge would begin to make contact with each other (Fig. S23), resulting in the resistence decrease of the
AC C
overall framework [20,21]. After releasing the pressure, the deformation of GO-PU sponge would recover to the initial condition.
4. Conclusions A novel type of metal/graphene oxide (M/GO) batteries is developed via the simple assembly of metal foils with GO films, including Li/GO, Na/GO, Zn/GO, Fe/GO, and Cu/GO batteries. Taking advantage of the spontaneous redox reaction
ACCEPTED MANUSCRIPT between M and GO, the chemical energy can be readily converted into the directly available electricity. Li/GO battery generates a specific capacity of ~1,572 mAh cm–3 (about 1,604 mAh g–1). In terms of 2D planar structure of M/GO batteries, their
RI PT
energy outputs are constricted to the interface reaction of metals and GO films. To address this issue, construction of 3D M/GO batteries with more active sites is needed to achieve higher energy density. Therefore, 3D-Cu foam anode and flowing
SC
GO/ionic-liquid catholyte are assembled to form a semi-solid 3D-Cu/GO redox flow
M AN U
battery (3D-Cu/GO RFB). Such a design enables GO catholyte to flow through the Cu foam, thus increasing their contact area and accelerating the reaction rate to achieve higher specific capacity. As expected, the specific capacity of 3D-Cu/GO RFB is ca. 97 times that of Cu foil/GO RFB. More importantly, a compressible, all-solid-state,
TE D
and pressure-responsive 3D-GO/Zn battery is fabricated, which can accurately control the energy output in response to pressure stimulations without the aid of conventional battery management system. It can not only respond fast and synchronously to the
EP
periodically dynamic and static pressure, but also exhibit excellent cyclic stability and
AC C
high reliability. Overall, 3D-GO/Zn battery with energy management function makes an important step towards next-generation smart energy storage devices. Beyond those demonstrated in this work, the concept of M/GO battery will be a new starting point for the design of similar electrochemical power sources and other functions.
Acknowledgements
ACCEPTED MANUSCRIPT This work was supported by NSFC (No. 21325415, 51673026), Beijing Natural Science Foundation (2152028), and Beijing Municipal Science and Technology
Appendix A. Supplementary data
RI PT
Commission (Z161100002116022), and 111 Project 807012.
Supplementary data associated with this article can be found in the online version at
SC
*********.
References
[5] [6] [7]
AC C
[8] [9]
M AN U
[4]
TE D
[3]
A. K. Geim, Science 324 (2009) 1530−1534. D. R. Dreyer, S. Park, C. W. Bielawski, R. S. Ruoff, Chem. Soc. Rev. 39 (2010) 228−240. F. Zhao, H. H. Cheng, Z. P. Zhang, L. Jiang, L. T. Qu, Adv. Mater. 27 (2015) 4351−4357. F. Zhao, Y. Liang, H. H. Cheng, L. Jiang, L. T. Qu, Energ. Environ. Sci. 9 (2016) 912−916. Y. Jiang, H. Shao, C. Li, T. Xu, Y. Zhao, G. Q. Shi, L. Jiang, L. T. Qu, Adv. Mater. 28 (2016) 10287−10292. Y. Zhao, C. G. Hu, Y. Hu, H. H. Cheng, G. Q. Shi, L. T. Qu, Angew. Chem. Int. Ed. 51 (2012) 11371−11375. H. H. Cheng, M. H. Ye, F. Zhao, C. G. Hu, Y. Zhao, Y. Liang, N. Chen, S. L. Chen, L. Jiang, L. T. Qu, Adv. Mater. 28 (2016) 3305−3312. M. H. Ye, C. X. Li, Y. Zhao, L. T. Qu, Carbon 106 (2016) 9−19. C. G. Hu, G. P. Zheng, F. Zhao, H. B. Shao, Z. P. Zhang, N. Chen, L. Jiang, L. T. Qu, Energ. Environ. Sci. 7 (2014) 3699−3708. C. G. Hu, X. Q. Zhai, L. L. Liu, Y. Zhao, L. Jiang, L. T. Qu, Sci. Rep. 3 (2013) 2065. M. H. Ye, Z. L. Dong, C. G. Hu, H. H. Cheng, H. B. Shao, N. Chen, L. T. Qu, Small 10 (2014) 5035−5041. X. B. Cao, D. P. Qi, S. Y. Yin, J. Bu, F. J. Li, C. F. Goh, S. Zhang, X. D. Chen, Adv. Mater. 25 (2013) 2957−2962. M. H. Ye, H. H. Cheng, J. Gao, C. X. Li, L. T. Qu, J. Mater. Chem. A 4 (2016) 19154−19159. J. Liu, M. Durstock, L. M. Dai, Energ. Environ. Sci. 7 (2016) 1297−1306. L. X. Wang, F. Zhao, Q. Han, C. G. Hu, L. X. Lv, N. Chen, L. T. Qu, Nanoscale 7 (2015) 9694−9702.
EP
[1] [2]
[10] [11] [12] [13] [14] [15]
ACCEPTED MANUSCRIPT
[17] [18] [19] [20] [21]
AC C
EP
TE D
M AN U
SC
[22]
Y. N. Meng, Y. Zhao, C. G. Hu, H. H. Cheng, Y. Hu, Z. P. Zhang, G. Q. Shi, L. T. Qu, Adv. Mater. 25 (2013) 2326−2331. R. Dedryvere, S. Laruelle, S. Grugeon, L. Gireaud, J. M. Tarascon, D. Gonbeau, J. Electrochem. Soc. 152 (2005) A689−A696. X. P. Wang, J. Gao, Z. H. Cheng, N. Chen, L. T. Qu, Angew. Chem. Int. Ed. 55 (2016) 14643−14647. D. Lin, Y. Liu, Y. Cui, Nat. Nanotechnol. 12 (2017) 194−206. P. P. Zhang, L. X. Lv, Y. Liang, J. Li, H. Cheng, Y. Zhao, L. T. Qu, J. Mater. Chem. A 4 (2016) 10118−10123. L. Z. Sheng, Y. Liang, L. L. Jiang, Q. Wang, T. Wei, L. T. Qu, Z. J. Fan, Adv. Funct. Mater. 25 (2015) 6545−6551. J. Chen, Y. Li, L. Huang, N. Jia, C. Li, G. Q. Shi, Adv. Mater. 27 (2015) 3654−3660.
RI PT
[16]