- Email: [email protected]
Effect of ion size and charge density on the electrochemical characteristics of α-MoO3 using aqueous Be2+ and Mg2+ sulfate electrolytes
Accepted Manuscript Effect of ion size and charge density on the electrochemical characteristics of α2+ 2+ MoO3 using aqueous Be and Mg sulfate electrolytes Juan C. Icaza, Richard T. Haasch, Ramesh K. Guduru PII:
S0925-8388(17)34592-9
DOI:
10.1016/j.jallcom.2017.12.377
Reference:
JALCOM 44463
To appear in:
Journal of Alloys and Compounds
Received Date: 16 October 2017 Revised Date:
26 December 2017
Accepted Date: 31 December 2017
Please cite this article as: J.C. Icaza, R.T. Haasch, R.K. Guduru, Effect of ion size and charge density 2+ 2+ on the electrochemical characteristics of α-MoO3 using aqueous Be and Mg sulfate electrolytes, Journal of Alloys and Compounds (2018), doi: 10.1016/j.jallcom.2017.12.377. 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
Effect of ion size and charge density on the electrochemical characteristics of αMoO3 using aqueous Be2+ and Mg2+ sulfate electrolytes
RI PT
Juan C. Icazaa, Richard T. Haaschb, Ramesh K. Gudurua,1 a
Department of Mechanical Engineering, Lamar University, Beaumont, TX-77710, USA
b
M AN U
Abstract
SC
Materials Research Laboratory, University of Illinois at Urbana-Champaign, Urbana,IL-61801, USA
We investigated the electrochemical characteristics of layer structured α-MoO3 using two divalent ions (Mg2+ and Be2+) of different sizes. The cyclic voltammetry studies showed a high specific capacity with Be2+ electrolyte, and galvanostatic intermittent titration (GITT)
TE D
measurements revealed a higher diffusion coefficient for Be2+ in MoO3 despite having a higher charge density than Mg2+. The intercalation of both ions resulted in poor cyclability, but a slightly better performance with Be2+ electrolyte could be attributed to favored diffusion of
AC C
EP
Be2+ in MoO3 due to its covalent nature as well as preferred coordination number.
Keywords: Diffusion; Electrochemical reactions; Electrode materials; Nanostructured materials; Energy storage materials.
1
Corresponding author Email: [email protected]
1
ACCEPTED MANUSCRIPT
1. Introduction Li-ion batteries have long been serving various energy prime applications ranging from small
RI PT
scale electronic gadgets to automotive vehicles and power grids, etc.[1]. On the other hand, ultracapacitors are used in high power applications including regenerative brakes and emergency exits, etc.[2] However, the recent surge in demand for high energy and high power
SC
densities has been driving the researchers to develop hybrid systems of batteries and supercapacitors [2]. In the meantime, costly and hazardous natures of Li-ion batteries also
M AN U
attracted several alternative solutions, among which, low-cost aqueous electrolyte systems and multivalent chemistries have gained a significant importance due to several advantages [3,4]. For example, aqueous electrolytes are inexpensive, safer, non-flammable, non-corrosive, easy to handle, and offer higher ionic conductivity than the non-aqueous electrolytes [2,5,6],
TE D
whereas multivalent chemistries offer more charge storage than Li-ion chemistries, and thereby provide opportunities to enhance the energy density as well as increased operating currents. However, there are several challenges with multivalent electrolytes, such as, unwanted proton
EP
(H+) co-intercalation, small voltage windows [7] and unstable electrode crystal structures that
AC C
hinder the mobility of multivalent ions resulting in poor cyclability [3]. While the literature on intercalation/mobility of Li ions in different types of electrodes is abundant, the data on the mobility of multivalent ions is limited [3,7]. Recently, Gerbrand Ceder’s group [3,4] has done modeling and experimental work on understanding the intercalation of multivalent ions in different electrode crystal structures and presented the effect of diffusion topology on the mobility of multivalent ions. However, the effect of charge density on the mobility of divalent ions was never undertaken. Therefore, here, we investigated the intercalation characteristics of 2
ACCEPTED MANUSCRIPT
two divalent ions with different ionic diameter - Be2+ (0.62 Å) and Mg2+ (1.44 Å) [8] in α-MoO3 for asymmetric supercapacitors. α-MoO3 is one of the best systems for layer structured
RI PT
electrodes. It has an orthorhombic crystal structure with stacked layers of MoO6 (interlayer spacing of ~6.929 Å) held by weak van der Waals forces along the [010] axis [4,9] and is well suited for intercalation of an extensive range of ions. So far, literature showed the intercalation
SC
of Li+, Na+, H+, Mg2+ and Al3+ [4,9 - 30] into different types of nanostructures [4,9–22,27–30] of MoO3, such as, nanowires [20,21], nanoflakes [27], nanorods [13,15,16], nanobelts [22] and
M AN U
nanoplatelets [28]. In contrast, present studies evaluated the intercalation behavior of two divalent ions of different sizes/diameters into nanoplatelet structured α-MoO3 and then compared. To the best of our knowledge, this is the first study to compare the effect of charge density of divalent ions for charge storage in layer structured electrodes. Furthermore, this
TE D
work provides insight to a fundamental key to understanding the charge storage mechanism of multivalent ions, which is diffusion topology, thereby presenting another step towards
AC C
2. Experimental
EP
sustainable energy beyond lithium-ion technology.
2.1. Materials: Commercial MoO3 powders (Pfaltz & Bauer, USA) and microporous activated carbon (AC) cloth (Zorflex-double weave, Charcoal House LLC, USA) were used for electrode materials. The AC cloth was cleaned as described in our previous work [31]. 2.2. Characterization: The phase, morphology, and microstructures of electrode materials were analyzed by X-ray diffraction (XRD, Thermoscientific diffractometer, ARL Equinox 100 x-ray), 3
ACCEPTED MANUSCRIPT
and transmission electron microscopy (TEM, JEM-2800). The Brunauer-Emmett-Teller (BET, TriStar II Plus 2.03, Micromeritics Instruments Corp., USA) surface area analysis and surface
RI PT
porosity measurements (ASAP 2010, Micromeritics Instruments Corp., USA) were performed on MoO3 powders and AC cloth, respectively. The particle size distribution of MoO3 powders was evaluated from the particle size measured using the TEM images. X-ray photoelectron
SC
spectroscopy (XPS) was performed in a Kratos Axis ULTRA spectrometer (Materials Research Laboratory, University of Illinois). Binding energies were referenced to the graphitic C 1s peak at
M AN U
284.5 eV.
2.3. Electrochemical studies: The MoO3 electrodes were prepared using a slurry made of 80 wt% of MoO3 powder as the active material with 15 wt% of conductive carbon, 5 wt% of polyvinylidene difluoride binder, and N-methyl-2-pyrrolidone solvent. The slurry was then
TE D
coated onto a steel foil (thickness - 50µm) and dried at ~60 oC in a vacuum oven for 12 hours. The electrolytes used were aqueous 0.5M BeSO4 (Reagent - 98%, GFS Chemicals, USA) and aqueous 0.5M MgSO4 (Anhydrous, Reagent, GFS Chemicals, USA). All the electrochemical
EP
characterizations were carried out on a Solartron 1470E workstation (Solartron Instruments,
AC C
USA). Three-electrode tests were performed using a flat cell kit (Ametek Scientific Instruments, USA) with the active material (MoO3 or AC) as a working electrode, Ag/AgCl - reference electrode, and a platinum mesh - counter electrode. Two electrode tests on MoO3 against AC were performed using a stainless steel split cell (MTI Corporation, USA), with a glass microfiber separator (VWR, USA). All the electrochemical measurements were performed at room temperature (~25 oC).
4
ACCEPTED MANUSCRIPT
3. Results and Discussion
RI PT
The XRD pattern (Fig 1(a)) confirmed the orthorhombic phase of α-MoO3 matching with the JCPDS No. 05-0508 database with lattice parameters of a=3.96 Å, b=13.86 Å, and c=3.70 Å corresponding to the pure crystalline orthorhombic α-MoO3. The TEM image, see Fig. 1(b),
SC
showed single crystal nanoplatelets of α-MoO3 with thickness along [010] (b-axis). The BET measurements revealed a surface area of 2.59 m2g-1, which is very limited to contribute an
M AN U
appreciable amount of double layer capacitance. The surface area of AC electrodes was 1034 m2g-1 with a pore size from 3.9 to 30 Å (average ~ 17.4 Å), and more details on AC can be found in [31]. Figure 1(c) shows the particle size distribution of α-MoO3 powders with most of the
TE D
particles smaller than 2 µm, and average particle size was around 2.1 μm.
Initially, the stability of aqueous BeSO4 and MgSO4 electrolytes was evaluated through cyclic
EP
voltammetry (CV) studies using MoO3 and AC, separately, with a three electrode setup. The CV
AC C
scans for the first cycle of MoO3 with aqueous 0.5M BeSO4 (pH 2.2) and 0.5M MgSO4 (pH 6.2) are shown in Fig. 2(a). For a fair comparison of these two electrolytes, the CV scans were also obtained with reduced pH (2.2) of MgSO4, which was prepared by mixing dilute sulfuric acid with aqueous 0.5M MgSO4 (pH 6.2). All the three electrolytes showed multiple redox peaks, however, to distinguish these from H+ co-intercalation, the CV scans with dilute sulfuric acid electrolytes of pH 2.2 and 6.2 (Fig. 2(b)) were also obtained. Evidently, the redox couples associated with H+ insertion are clearly distinct from the redox couples of Be2+ and Mg2+ 5
ACCEPTED MANUSCRIPT
intercalation. The intercalation of Be2+ showed three pairs of redox couples with cathodic peaks at -0.34, 0. 11 and 0.58 V Vs Ag/AgCl, and anodic peaks at -0.072, -0.4 and -0.64 V Vs Ag/AgCl.
RI PT
The redox peaks for intercalation of Mg2+ were not discernible at pH - 6.2. A magnified view of the CV scan with aqueous 0.5M MgSO4 (pH 6.2) electrolyte is provided in the inset of Fig. 2(a). However, they became distinctive when the pH was reduced to 2.2, and exhibited a three stage
SC
intercalation with redox peaks at -0.45, 0.13 and 0.44 V Vs Ag/AgCl (cathodic), and -0.016, 0.28 and -0.66 Vs Ag/AgCl (anodic). These observations agree with multi-stage intercalation of
M AN U
Mg2+ and Al3+ into MoO3 as reported in the literature[19,23–27]. These peaks represent the redox transformations of Mo (IV) ↔ Mo (V) ↔ Mo (VI). The reaction can be described by the following equation, where A is the divalent ion being inserted into MoO3:
TE D
+ + 2 →
The peak current (i) of the CV curves and the sweep rate (v) follow the relationship i=avb. Where “a” and “b” are adjustable values, and “b” was determined from the slope of log (v) Vs
EP
log (i). In this study, the b-values were close to 0.5 for both Be2+ and Mg2+ indicating a diffusion
AC C
controlled mechanism (Fig S1). The aqueous BeSO4 showed stability between -0.75 to 1 V (Vs Ag/AgCl), whereas aqueous MgSO4 showed an increased stability with reduced pH, see Fig. 2(a). A similar dependency of electrolyte stability on pH was reported for Li-ion electrolytes by Yi Cui et al. [32].
6
ACCEPTED MANUSCRIPT
On the other hand, the CV scans of AC (not shown here) exhibited stability between -1 to 1 V (Vs Ag/AgCl) with double layer capacitance for all the three electrolytes. Fig. 2(c) shows the scan rate dependent specific capacity of MoO3 for all the three electrolytes. Interestingly, Be2+
RI PT
electrolyte yielded the highest specific capacity of 158.3 mAh g-1 (~285 Fg-1), whereas MgSO4 showed a lower value (68 mAhg-1) with a slight improvement (85 mAhg-1) after reducing the pH,
SC
which could be due to co-intercalation of H+ [14,30]. This was apparent from the CV studies of dilute sulfuric acid electrolytes as well, see Figs 2(b) and 2(c), where the specific capacity of
M AN U
MoO3 increased with reducing pH of the electrolyte. It should be noted that the capacitance measurements in three electrode tests are usually limited by the surface area of the counter electrode where double layer of anions would form. On the other hand, the AC electrode had a large surface area, and therefore, it must have facilitated in obtaining a higher capacitance in
TE D
the two electrode tests in the present studies. In addition, we performed the CV scans on bare steel foil substrate (i.e., the current collector) as well in three electrode configuration in order
EP
to ensure a negligible contribution of steel on the capacitance values measured (see Fig S2).
AC C
Following these studies, the AC electrodes were tested in a symmetric configuration with 0.5M BeSO4 (pH 2.2), 0.5M MgSO4 (pH 6.2) and 0.5M MgSO4 (pH 2.2), and obtained specific capacitances around 46, 22 and 34.2 Fg-1, respectively. Based on these, a mass ratio of 30 to 1 between AC and MoO3 was used for all the asymmetric cells (i.e., two electrode tests) to avoid limiting effects of AC on the specific capacity measurements of MoO3. Figure 3 shows the CV scans for the first cycle and scan rate dependent capacity of MoO3 electrode from AC Vs MoO3
7
ACCEPTED MANUSCRIPT
tests with different electrolytes. The CV scans (Fig 3(a)) show three stage redox peaks (barely visible) agreeing with the three electrode tests, and broad nature of these scans can be attributed to a combination of intercalation and double layer formation at MoO3 and AC
RI PT
electrodes, respectively. The MoO3 electrode exhibited a specific capacity of 335 mAh g-1 (~928 Fg-1) with aqueous BeSO4 electrolyte (@2 mVs-1), and it is considerably larger when compared with the three electrode capacity due to the large surface area provided by the AC electrode for
SC
double layer formation of anions while leading to a larger amount of Be2+ intercalation into
M AN U
MoO3. On the other hand, aqueous MgSO4 electrolytes resulted in specific capacities of 72.5 and 88.4 mAhg-1 for pH 6.2 and 2.2 (Fig 3(b)), respectively, reflecting a minimal influence of H+ intercalation. However, these values are only a slight improvement over the three-electrode test data, which certainly indicates a limited intercalation of Mg2+ into MoO3 when compared
TE D
with Be2+. The impedance data in Fig. 3(c) also depicted a larger Warburg component for MgSO4 electrolytes implying more resistance to the to the bulk diffusion of Mg2+. The Nyquist plots obtained were modeled and interpreted with the help of the appropriate equivalent circuit (see
EP
figure 3(e)). Where Relectrolyte is the ohmic resistance at the electrode/electrolyte interface, Rfaradic is the charge transfer resistance of faradic processes, RL is the leakage resistance, W is
AC C
the Warburg element, CPEdl and CPEfaradic are constant phase elements of the double layer and faradic reactions, respectively. The elements of the equivalent circuit model fitted to the impedance spectra are shown in table 1 for all the BeSO4 and MgSO4 electrolytes. The results indicate negligible differences in electrolyte resistances and double layer capacitance values. However, the capacitance contribution by the Warburg element is up to three times larger with Be2+ electrolyte when compared to Mg2+. This corroborates our findings from the CV tests
8
ACCEPTED MANUSCRIPT
representing more intercalation of Be2+ into the electrode compared to Mg2+, which was also confirmed in our later studies through GITT studies (see Fig. 4). These observations clearly
EIS parameters from the equivalent circuit for α-MoO3 in BeSO4 and MgSO4
electrolytes R-electrolyte (Ω cm2)
R-faradic (Ω cm2)
2.2 6.2 2.2
1.506 1.676 3.252
3.176 8.131 4.51
R- L (Ω cm2)
CPE-dl (farads cm-2)
CPE-faradic (farads cm-2)
W (farads cm-2)
22.92 104.6 172.3
0.6305 0.0001436 0.00239
5.139 0.22158 0.21785
122.4 35.87 40.85
M AN U
BeSO₄ MgSO₄ MgSO₄
pH
SC
Table 1.
RI PT
represent more resistance to the diffusion of Mg2+ into MoO3 electrode than Be2+.
The cyclability data presented a poor retention of specific capacity of MoO3 with high
TE D
coulombic efficiency for all the electrolytes (Fig 3(d)). The literature [3,4,7] also suggests that most of the electrode materials suffer from limited and fading capacities with multivalent
EP
electrolytes due to stronger electrostatic interactions of multivalent ions with the crystal structures of the electrodes resulting in a reduced mobility and possibly collapse of the crystal
AC C
structures of the electrode materials. In addition, a drastic localized deformation of the electrodes caused by abrupt changes in the oxidation states with partial trapping of ions was reported to be one of the major reasons for poor cyclability [3,7,33–37]. Therefore, we believe that similar and strong electrostatic interactions of Be2+ and Mg2+ with MoO3 layer structures may have resulted in poor cyclability. Be2+ being a smaller ion with more charge density compared to Mg2+ was expected to have stronger electrostatic interactions with the layers of
9
ACCEPTED MANUSCRIPT
MoO3. However, unlike the alkaline earth metal ions (e.g. Mg2+, Ca2+ and Sr2+, etc.,), Be2+ is unique with both ionic and covalent nature [38], and this could have possibly resulted in
RI PT
different characteristics and thereby probably a slightly better performance.
We also performed ex-situ XRD studies on the MoO3 samples fully charged/discharged at 6.25
SC
Ag-1 using aqueous 0.5M BeSO4 (pH 2.2) and 0.5M MgSO4 (pH 6.2) electrolytes in order to understand the structural changes of MoO3 active material. After the first full charge, the
M AN U
sample tested in 0.5M BeSO4 showed a shift in the diffracted peaks to lower angles along with broadening of the peaks (see Fig. 4(a)). The shift in (020) peak can be ascribed to an increase in the interlayer spacing. Additionally the XRD pattern exhibits new peaks at 2θ = 18.5o, 23.8o, 31.3o, and 43.7o which can potentially be associated with the insertion of beryllium and the
TE D
formation of a new BexMoO3 phase, similar to those reported for Li+, Mg2+ and Al3+ [19,27,39]. After one full charge/discharge cycle, the original peaks did not fully return to their pristine
EP
condition, and the new peaks were still present with reduced intensities. Similarly, the MoO3 electrodes tested in 0.5M MgSO4 electrolyte showed extra peaks in the XRD patterns along with
AC C
a considerable shift in the peaks to lower angles and peak broadening (see Fig. 4(b)) after the full charge cycle. These new peaks appeared at 2θ =13.76o, 22.58 o, 29.87 o and 44.23 o, which can potentially be associated with the formation of a magnesiated phase (MgxMoO3) as observed in prior reports by Aurbach et al. [27]. These results are similar to the ones obtained with BeSO4. However, the remaining new peaks were much more intense for MgSO4 electrolyte indicating to the possibility of partial entrapment of ions as reported by Aurbach et at [27].
10
ACCEPTED MANUSCRIPT
To further elucidate the electrochemical activity of MoO3 electrodes, we performed XPS studies on the MoO3 samples fully charged/discharged at 6.25 Ag-1 using aqueous 0.5M BeSO4 (pH 2.2)
RI PT
and 0.5M MgSO4 (pH 6.2) electrolytes. Fig 4(c) shows the XPS spectra of MoO3 electrodes after a full charge and discharge with aqueous 0.5M BeSO4 electrolyte. The peaks at ~233.0 eV and ~236.1 eV correspond to the characteristic Mo 3d spin-orbit doublet peaks of MoO3 [11]. After one full charge in 0.5M BeSO4, new peaks appear at ~231.6 and 234.7 eV, between MoO2 and
SC
MoO3, these peaks can be associated with the insertion of Be2+ forming a MoO2-BeO complex.
M AN U
After the completion of first discharge cycle the new peaks remained but reduced quite a bit in intensity. Similarly, the MoO3 samples tested with 0.5M MgSO4 (see Fig. S3) also showed a new peak at ~231.6 eV corresponding to the insertion of Mg2+ with formation of MoO2-MgO complexes. These results are consistent with the extra peaks appeared in the XRD spectra,
TE D
which indicates that the insertion of Be2+ and Mg2+ may not be completely reversible. In fact some of this irreversibility was evident to certain extent even in the columbic efficiency data presented in Fig. 3(d), where the columbic efficiency gradually increased with increasing
AC C
EP
number of cycles and decay in the capacitance.
To gain further insights into the diffusion of both Be2+ and Mg2+ in MoO3, the diffusion coefficients of both ions were evaluated through GITT experiments (Fig S4) using the following equation
! 4 $ = ! √#
11
ACCEPTED MANUSCRIPT Here, - current, - molar volume of the electrode, - number of charges, - Faraday's
constant, - contact area between electrode and electrolyte, dE/dδ - slope of the steady state
voltages, and dE/d√t - slope of the linearized potential E(V) during the current pulse duration of
RI PT
t(s). Figure 4(d) shows the variation of diffusion coefficients of Be2+ and Mg2+ with respect to their concentration in MoO3. The diffusion coefficients of Be2+ and Mg2+ were determined to vary from 1.9 x 10-11 (at x = 0 for BexMoO3) to 6.4 x 10-12 cm2s-1, and 1.3 x 10-11 (at x = 0 for
SC
MgxMoO3) to 4.8 x 10-13 cm2s-1, respectively. The diffusion coefficients of Mg2+ and Be2+ are on
M AN U
the lower side when compared with the diffusion coefficients values reported for Li+ in MoO3 (10-9 to 10-12 cm2s-1)[40–42]. This can be attributed to the larger electrostatic interactions of divalent ions with the lattice structure of electrode material. At x=0, the diffusion coefficients of Be2+ and Mg2+ were almost equal suggesting similar levels of resistance at the liquid
TE D
electrolyte and electrode interface. However, with increasing the x value, the diffusion coefficient of Mg2+ dropped almost by two orders of magnitude, whereas Be2+ showed a minimal drop. The reason for such a contrasting behavior could be related to the diffusion
EP
mechanisms of both ions. According to Ziqin Rong et al. [3], the hopping mechanism of a multivalent ion in a given electrode follows a lower energy path depending on the crystal
AC C
structure of the host as well as preferred coordination number of the multivalent ion. In layer structured electrodes, the diffusion of multivalent ions is expected to follow octahedraltetrahedral-octahedral path as shown in Fig. 4(e). However, the rate of hopping is dictated by the preferred coordination number of the multivalent ion. For example, Li+ and Zn2+ with a coordination number of four can diffuse easily compared to the ions that have higher coordination number [3], for example, Mg2+ and Ca2+, which have preferred coordination
12
ACCEPTED MANUSCRIPT
numbers six and eight, respectively [3,43,44] diffuse slowly compared to Li+ and Zn2+. The crystallography literature [45] suggests that the preferred coordination number of Be2+ is four,
RI PT
and therefore, when compared with Mg2+ the hopping of Be2+ through octahedral-tetrahedraloctahedral path would very likely be favored in layer structured MoO3. On the other hand, the unique features of Be2+ with ionic and covalent nature, unlike Mg2+, would also likely result in weaker electrostatic interactions and thereby probably result in slightly better characteristics.
SC
This could be the same reason why Mg2+ intercalation did not show much improvement in the
M AN U
capacity in two electrode tests in comparison with the three electrode test data. In addition, higher impedance in the Warburg region along with GITT diffusion measurements also
4. Conclusions
TE D
substantiate that the diffusion of Mg2+ will be slower than Be2+.
The intercalation characteristics of α-MoO3 were investigated with Be2+ and Mg2+ of different
EP
sizes. Be2+ showed the highest capacity for the MoO3/AC asymmetric cell, and it can be ascribed to peculiar characteristics of Be2+ with covalent and ionic nature, and a preferred coordination
AC C
number of four, which may have resulted in a higher diffusion coefficient of Be2+ in MoO3. The high capacitive behavior displayed by Be2+ presents an opportunity for improved energy densities of supercapacitors when compared against Mg2+ and possibly other multivalent ions. However, understanding of new phases with insertion of Be2+ into MoO3 needs further investigation, which will be presented in the future publications.
13
ACCEPTED MANUSCRIPT
Acknowledgements Juan C. Icaza would like to thank the Dept. of Mechanical Engineering at Lamar University for
RI PT
the Doctoral Fellowship. Also, this work is supported by the start-up funds provided by Lamar University to the faculty Dr. Ramesh K. Guduru. XPS analysis was carried out in part at the
SC
Frederick Seitz Materials Research Laboratory Central Research Facilities, University of Illinois.
References
M.A. Kiani, M.F. Mousavi, M.S. Rahmanifar, Synthesis of nano- and micro-particles of
M AN U
[1]
LiMn2O4: Electrochemical investigation and assessment as a cathode in li battery, Int. J. Electrochem. Sci. 6 (2011) 2581–2595. doi:10.1039/c1ee01598b. A. González, E. Goikolea, J.A. Barrena, R. Mysyk, Review on supercapacitors:
TE D
[2]
Technologies and materials, Renew. Sustain. Energy Rev. 58 (2016) 1189–1206. doi:10.1016/j.rser.2015.12.249.
Z. Rong, R. Malik, P. Canepa, G. Sai Gautam, M. Liu, A. Jain, K. Persson, G. Ceder,
EP
[3]
AC C
Materials Design Rules for Multivalent Ion Mobility in Intercalation Structures, Chem. Mater. 27 (2015) 6016–6021. doi:10.1021/acs.chemmater.5b02342.
[4]
P. Canepa, G. Sai Gautam, D.C. Hannah, R. Malik, M. Liu, K.G. Gallagher, K.A. Persson, G.
Ceder, Odyssey of Multivalent Cathode Materials: Open Questions and Future Challenges, Chem. Rev. 117 (2017) 4287–4341. doi:10.1021/acs.chemrev.6b00614.
[5]
C. Zhao, W. Zheng, A Review for Aqueous Electrochemical Supercapacitors, Front. Energy 14
ACCEPTED MANUSCRIPT
Res. 3 (2015) 1–11. doi:10.3389/fenrg.2015.00023. [6]
G. Wang, L. Zhang, J. Zhang, A review of electrode materials for electrochemical
[7]
RI PT
supercapacitors, Chem. Soc. Rev. 41 (2012) 797–828. doi:10.1039/c1cs15060j. R. Guduru, J. Icaza, A Brief Review on Multivalent Intercalation Batteries with Aqueous
[8]
SC
Electrolytes, Nanomaterials. 6 (2016) 41. doi:10.3390/nano6030041.
A.G. Volkov, S. Paula, D.W. Deamer, Two mechanisms of permeation of small neutral
M AN U
molecules and hydrated ions across phospholipid bilayers, Bioelectrochemistry Bioenerg. 42 (1997) 153–160. doi:10.1016/S0302-4598(96)05097-0. [9]
N.A. Chernova, M. Roppolo, A.C. Dillon, M.S. Whittingham, Layered vanadium and molybdenum oxides: batteries and electrochromics, J. Mater. Chem. 19 (2009) 2526–
[10]
TE D
2552. doi:10.1039/b819629j.
T. Brezesinski, J. Wang, S.H. Tolbert, B. Dunn, Ordered mesoporous alpha-MoO3 with iso-
EP
oriented nanocrystalline walls for thin-film pseudocapacitors., Nat. Mater. 9 (2010) 146– 151. doi:10.1038/nmat2612.
J. Jiang, J. Liu, S. Peng, D. Qian, D. Luo, Q. Wang, Z. Tian, Y. Liu, Facile synthesis of alpha-
AC C
[11]
MoO3 nanobelts and their pseudocapacitive behavior in an aqueous Li2SO4 solution, J. Mater. Chem. A. 1 (2013) 2588–2594. doi:10.1039/c2ta01120d.
[12]
N. Anbananthan, K.N. Rao, V.K. Venkatesan, Electrochemical formation of oxygendeficient molybdenum oxide, Appl. Surf. Sci. 72 (1993) 189–194. doi:10.1016/01694332(93)90011-Y. 15
ACCEPTED MANUSCRIPT
[13]
I. Shakir, M. Shahid, H.W. Yang, D.J. Kang, Structural and electrochemical characterization of α-MoO3 nanorod-based electrochemical energy storage devices, Electrochim. Acta. 56
[14]
RI PT
(2010) 376–380. doi:10.1016/j.electacta.2010.09.028. B. Mendoza-Sánchez, T. Brousse, C. Ramirez-Castro, V. Nicolosi, P. S. Grant, An
investigation of nanostructured thin film α-MoO3 based supercapacitor electrodes in an
doi:10.1016/j.electacta.2012.11.127.
Z. Cui, W. Yuan, C.M. Li, Q. Zhang, T. Wei, J. Feng, M. Zhang, W. Qian, F. Wei, A.C. Dillon,
M AN U
[15]
SC
aqueous electrolyte, Electrochim. Acta. 91 (2013) 253–260.
Template-mediated growth of microsphere, microbelt and nanorod α-MoO3 structures and their high pseudo-capacitances, J. Mater. Chem. A. 1 (2013) 12926-12931.
[16]
TE D
doi:10.1039/c3ta12688a.
L. Zheng, Y. Xu, D. Jin, Y. Xie, Q. Song, J. Guo, L.Q. Chen, R. Seshadri, G.D. Stucky, Wellaligned molybdenum oxide nanorods on metal substrates: solution-based synthesis and
EP
their electrochemical capacitor application, J. Mater. Chem. 20 (2010) 7135-7143.
[17]
AC C
doi:10.1039/c0jm00744g.
X. Xia, Q. Hao, W. Lei, W. Wang, H. Wang, X. Wang, Reduced-graphene
oxide/molybdenum oxide/polyaniline ternary composite for high energy density supercapacitors: Synthesis and properties, J. Mater. Chem. 22 (2012) 8314-8320. doi:10.1039/c2jm16216d.
[18]
J. Chang, M. Jin, F. Yao, T.H. Kim, V.T. Le, H. Yue, F. Gunes, B. Li, A. Ghosh, S. Xie, Y.H. Lee,
16
ACCEPTED MANUSCRIPT
Asymmetric supercapacitors based on graphene/MnO2 nanospheres and graphene/MoO3 nanosheets with high energy density, Adv. Funct. Mater. 23 (2013)
[19]
RI PT
5074–5083. doi:10.1002/adfm201301851. Y.P. Wu, F. Wang, X. Wang, X. Yuan, Z. Liu, X. Wu, Y. Zhu, L. Fu, A conductive polymer coated MoO3 anode enables an Al-ion capacitor with high performance, J. Mater. Chem.
[20]
SC
A. 4 (2016) 5115–5123. doi:10.1039/C6TA01398H.
I. Shakir, M. Shahid, U.A. Rana, M.F. Warsi, Y. Xie, G. Yu, W. Xing, Z. Lin, W. Wu, T. Li, H.
M AN U
Jin, P. Liu, J. Zhou, C.P. Wong, Z.L. Wang, In situ hydrogenation of molybdenum oxide nanowires for enhanced supercapacitors, RSC Adv. 4 (2014) 8741-8745. doi:10.1039/c3ra44837a.
I. Shakir, M. Shahid, M. Nadeem, D.J. Kang, Tin oxide coating on molybdenum oxide
TE D
[21]
nanowires for high performance supercapacitor devices, Electrochim. Acta. 72 (2012) 134–137. doi:10.1016/j.electacta.2012.04.002. Y. Liu, B. Zhang, Y. Yang, Z. Chang, Z. Wen, Y. Wu, Polypyrrole-coated α-MoO3 nanobelts
EP
[22]
AC C
with good electrochemical performance as anode materials for aqueous supercapacitors, J. Mater. Chem. A. 1 (2013) 13582. doi:10.1039/c3ta12902k.
[23]
M.E. Spahr, P. Nov?k, O. Haas, R. Nesper, Electrochemical insertion of lithium, sodium,
and magnesium in molybdenum(VI) oxide, J. Power Sources. 54 (1995) 346–351. doi:10.1016/0378-7753(94)02099-O.
[24]
T.S. Sian, G.B. Reddy, S.M. Shivaprasad, Study of Mg ion Intercalation in Polycrystalline 17
ACCEPTED MANUSCRIPT
MoO 3 Thin Films, Jpn. J. Appl. Phys. 43 (2004) 6248–6251. doi:10.1143/JJAP.43.6248. [25]
T.S. Sian, G.B. Reddy, S.M. Shivaprasad, Effect of Microstructure and Stoichiometry on
RI PT
Absorption in Mg Intercalated MoO3 Thin Films, Electrochem. Solid-State Lett. 9 (2006) A120-A122. doi:10.1149/1.2163427. [26]
L. Campanella, G. Pistoia, MoO3: A New Electrode Material for Nonaqueous Secondary
G. Gershinsky, H.D. Yoo, Y. Gofer, D. Aurbach, Electrochemical and Spectroscopic Analysis
M AN U
[27]
SC
Battery Applications, J. Electrochem. Soc. 118 (1971) 1905-1908. doi:10.1149/1.2407864.
of Mg 2+ Intercalation into Thin Film Electrodes of Layered Oxides: V2O5 and MoO3, Langmuir. 29 (2013) 10964–10972. doi:10.1021/la402391f. [28]
W. Tang, L. Liu, S. Tian, L. Li, Y. Yue, Y. Wu, K. Zhu, Aqueous supercapacitors of high
TE D
energy density based on MoO3 nanoplates as anode material, Chem. Commun. 47 (2011) 10058-10060. doi:10.1039/c1cc13474d.
H. Farsi, F. Gobal, H. Raissi, S. Moghiminia, On the pseudocapacitive behavior of
EP
[29]
nanostructured molybdenum oxide, J. Solid State Electrochem. 14 (2010) 643–650.
[30]
AC C
doi:10.1007/s10008-009-0830-5. H. Farsi, F. Gobal, H. Raissi, S. Moghiminia, The pH effects on the capacitive behavior of
nanostructured molybdenum oxide, J. Solid State Electrochem. 14 (2010) 681–686. doi:10.1007/s10008-009-0828-z.
[31]
J.C. Icaza, R.K. Guduru, Effect of ion charges on the electric double layer capacitance of activated carbon in aqueous electrolyte systems, 336 (2016) 360-366. 18
ACCEPTED MANUSCRIPT
doi:10.1016/j.jpowsour.2016.11.017. [32]
C. Wessells, R. Ruffο, R.A. Huggins, Y. Cui, Investigations of the Electrochemical Stability
RI PT
of Aqueous Electrolytes for Lithium Battery Applications, Electrochem. Solid-State Lett. 13 (2010) A59-A61. doi:10.1149/1.3329652. [33]
M. Liu, Z. Rong, R. Malik, P. Canepa, A. Jain, G. Ceder, K.A. Persson, Spinel compounds as
SC
multivalent battery cathodes: a systematic evaluation based on ab initio calculations,
[34]
M AN U
Energy Environ. Sci. 8 (2015) 964–974. doi:10.1039/C4EE03389B.
R.Y. Wang, B. Shyam, K.H. Stone, J.N. Weker, M. Pasta, H.-W. Lee, M.F. Toney, Y. Cui, Reversible Multivalent (Monovalent, Divalent, Trivalent) Ion Insertion in Open Framework Materials, Adv. Energy Mater. 5 (2015) 1401869.
[35]
TE D
doi:10.1002/aenm.201401869.
B. Lee, C.S. Yoon, H.R. Lee, K.Y. Chung, B.W. Cho, S.H. Oh, Electrochemically-induced reversible transition from the tunneled to layered polymorphs of manganese dioxide,
E. Levi, M.D. Levi, O. Chasid, D. Aurbach, A review on the problems of the solid state ions
AC C
[36]
EP
Scientific Reports. 4 (2014) 6066. doi:10.1038/srep06066.
diffusion in cathodes for rechargeable Mg batteries, J. Electroceramics. 22 (2009) 13–19. doi:10.1007/s10832-007-9370-5.
[37]
H.D. Yoo, I. Shterenberg, Y. Gofer, G. Gershinsky, N. Pour, D. Aurbach, Mg rechargeable batteries: an on-going challenge, Energy Environ. Sci. 6 (2013) 2265-2279. doi:10.1039/c3ee40871j. 19
ACCEPTED MANUSCRIPT
[38]
C.W. Bock, J.P. Glusker, Organization of water around a beryllium cation, Inorg. Chem. 32 (1993) 1242–1250. doi:10.1021/ic00059a036. T. Tsumura, M. Inagaki, Lithium insertion/extraction reaction on crystalline MoO3, Solid
RI PT
[39]
State Ionics. 104 (1997) 183–189. doi:10.1016/S0167-2738(97)00418-9. [40]
G. Guzman, B. Yebka, J. Livage, C. Julien, Lithium intercalation studies in hydrated
SC
molybdenum oxides, Solid State Ionics. 86–88 (1996) 407–413. doi:10.1016/0167-
[41]
M AN U
2738(96)00338-4.
C. Julien, G.A. Nazri, Transport properties of lithium-intercalated MoO3, Solid State Ionics. 68 (1994) 117–123. doi:10.1016/0167-2738(94)90245-3.
[42]
C. Julien, A. Khelfa, J.P. Guesdon, A. Gorenstein, Lithium intercalation in MoO3: A
TE D
comparison between crystalline and disordered phases, Appl. Phys. A Solids Surfaces. 59 (1994) 173–178. doi:10.1007/BF00332212. M. Wenger, T. Armbruster, Crystal chemistry of lithium: oxygen coordination and
EP
[43]
bonding, Eur. J. Mineral. 3 (1991) 387–400. doi:10.1127/ejm/3/2/0387. I.D. Brown, What factors determine cation coordination numbers?, Acta
AC C
[44]
Crystallographica. Sect. B Struct. Sci. 44 (1988) 545–553. doi:10.1107/S0108768188007712.
[45]
W.W. Emmett, Geological Survey Professional Paper, J. Hydrol. 2 (1965) 274-275. http://dx.doi.org/10.1016/0022-1694(65)90048-x..
20
ACCEPTED MANUSCRIPT
SC
RI PT
(a)
AC C
EP
TE D
M AN U
(b)
1
ACCEPTED MANUSCRIPT
Figure 1.
(a) X-ray diffraction pattern of as-received commercial MoO3 powders, (b) TEM image of MoO3 powders with selective area diffraction pattern and high-
AC C
EP
TE D
M AN U
SC
commercial MoO3 powders.
RI PT
resolution lattice image, and (c) particle size distribution of as-received
2
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure 2.
(a) CV scans of MoO3 in three electrode configuration. The inset – magnified view of CV scan with aqueous 0.5M MgSO4, (b) CV scans with dilute sulfuric acid, and (c) variation of capacity with respect to the scan rate.
1
AC C
(e)
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
1
ACCEPTED MANUSCRIPT
Figure 3.
(a) CV scans of MoO3 Vs AC asymmetric cells, (b) specific capacity of MoO3 with respect to the scan rate, and (c) impedance of BeSO4 and MgSO4 electrolytes,
RI PT
and zoom in view shown in the top-right corner, (d) cyclability of specific capacity of MoO3, and (e) the equivalent circuit model used to fit the impedance
AC C
EP
TE D
M AN U
SC
spectroscopy data.
2
ACCEPTED MANUSCRIPT
st
RI PT
1 discharging to 0V
st
1 charging to 1V
Mo (6+) (233) Mo (6+) (236.1)
(e)
AC C
EP
Mo (5+) (234.7)
TE D
Mo (5+) (231.6)
M AN U
SC
Pristine
1
ACCEPTED MANUSCRIPT
Figure 4.
XRD patterns of pristine MoO3, MoO3 after the 1st charge and MoO3 after the 1st discharge at 6.25 Ag-1 for (a) 0.5M BeSO4 and (b) 0.5M MgSO4-pH 6.2. (c) XPS spectra in the Mo 3d binding energy region of α-MoO3 after charging/discharging
RI PT
in 0.5M BeSO4 (d) Variation of diffusion coefficients of Be2+ (electrolyte -0.5M BeSO4) and Mg2+ (0.5M MgSO4: pH-6.2) as a function of concentration Be2+ in
SC
MoO3, (e) octahedral-tetrahedral-tetrahedral diffusion topology in layer
AC C
EP
TE D
M AN U
structured electrode.
2
ACCEPTED MANUSCRIPT
Highlights
Comparison of intercalation behavior of divalent ions - Be2+ and Mg2+ in α-MoO3 Effect of divalent ion charge density on the capacitance of α-MoO3 Effect of divalent ion coordination number on diffusion path topology
AC C
EP
TE D
M AN U
SC
RI PT
• • •
S0925-8388(17)34592-9
DOI:
10.1016/j.jallcom.2017.12.377
Reference:
JALCOM 44463
To appear in:
Journal of Alloys and Compounds
Received Date: 16 October 2017 Revised Date:
26 December 2017
Accepted Date: 31 December 2017
Please cite this article as: J.C. Icaza, R.T. Haasch, R.K. Guduru, Effect of ion size and charge density 2+ 2+ on the electrochemical characteristics of α-MoO3 using aqueous Be and Mg sulfate electrolytes, Journal of Alloys and Compounds (2018), doi: 10.1016/j.jallcom.2017.12.377. 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
Effect of ion size and charge density on the electrochemical characteristics of αMoO3 using aqueous Be2+ and Mg2+ sulfate electrolytes
RI PT
Juan C. Icazaa, Richard T. Haaschb, Ramesh K. Gudurua,1 a
Department of Mechanical Engineering, Lamar University, Beaumont, TX-77710, USA
b
M AN U
Abstract
SC
Materials Research Laboratory, University of Illinois at Urbana-Champaign, Urbana,IL-61801, USA
We investigated the electrochemical characteristics of layer structured α-MoO3 using two divalent ions (Mg2+ and Be2+) of different sizes. The cyclic voltammetry studies showed a high specific capacity with Be2+ electrolyte, and galvanostatic intermittent titration (GITT)
TE D
measurements revealed a higher diffusion coefficient for Be2+ in MoO3 despite having a higher charge density than Mg2+. The intercalation of both ions resulted in poor cyclability, but a slightly better performance with Be2+ electrolyte could be attributed to favored diffusion of
AC C
EP
Be2+ in MoO3 due to its covalent nature as well as preferred coordination number.
Keywords: Diffusion; Electrochemical reactions; Electrode materials; Nanostructured materials; Energy storage materials.
1
Corresponding author Email: [email protected]
1
ACCEPTED MANUSCRIPT
1. Introduction Li-ion batteries have long been serving various energy prime applications ranging from small
RI PT
scale electronic gadgets to automotive vehicles and power grids, etc.[1]. On the other hand, ultracapacitors are used in high power applications including regenerative brakes and emergency exits, etc.[2] However, the recent surge in demand for high energy and high power
SC
densities has been driving the researchers to develop hybrid systems of batteries and supercapacitors [2]. In the meantime, costly and hazardous natures of Li-ion batteries also
M AN U
attracted several alternative solutions, among which, low-cost aqueous electrolyte systems and multivalent chemistries have gained a significant importance due to several advantages [3,4]. For example, aqueous electrolytes are inexpensive, safer, non-flammable, non-corrosive, easy to handle, and offer higher ionic conductivity than the non-aqueous electrolytes [2,5,6],
TE D
whereas multivalent chemistries offer more charge storage than Li-ion chemistries, and thereby provide opportunities to enhance the energy density as well as increased operating currents. However, there are several challenges with multivalent electrolytes, such as, unwanted proton
EP
(H+) co-intercalation, small voltage windows [7] and unstable electrode crystal structures that
AC C
hinder the mobility of multivalent ions resulting in poor cyclability [3]. While the literature on intercalation/mobility of Li ions in different types of electrodes is abundant, the data on the mobility of multivalent ions is limited [3,7]. Recently, Gerbrand Ceder’s group [3,4] has done modeling and experimental work on understanding the intercalation of multivalent ions in different electrode crystal structures and presented the effect of diffusion topology on the mobility of multivalent ions. However, the effect of charge density on the mobility of divalent ions was never undertaken. Therefore, here, we investigated the intercalation characteristics of 2
ACCEPTED MANUSCRIPT
two divalent ions with different ionic diameter - Be2+ (0.62 Å) and Mg2+ (1.44 Å) [8] in α-MoO3 for asymmetric supercapacitors. α-MoO3 is one of the best systems for layer structured
RI PT
electrodes. It has an orthorhombic crystal structure with stacked layers of MoO6 (interlayer spacing of ~6.929 Å) held by weak van der Waals forces along the [010] axis [4,9] and is well suited for intercalation of an extensive range of ions. So far, literature showed the intercalation
SC
of Li+, Na+, H+, Mg2+ and Al3+ [4,9 - 30] into different types of nanostructures [4,9–22,27–30] of MoO3, such as, nanowires [20,21], nanoflakes [27], nanorods [13,15,16], nanobelts [22] and
M AN U
nanoplatelets [28]. In contrast, present studies evaluated the intercalation behavior of two divalent ions of different sizes/diameters into nanoplatelet structured α-MoO3 and then compared. To the best of our knowledge, this is the first study to compare the effect of charge density of divalent ions for charge storage in layer structured electrodes. Furthermore, this
TE D
work provides insight to a fundamental key to understanding the charge storage mechanism of multivalent ions, which is diffusion topology, thereby presenting another step towards
AC C
2. Experimental
EP
sustainable energy beyond lithium-ion technology.
2.1. Materials: Commercial MoO3 powders (Pfaltz & Bauer, USA) and microporous activated carbon (AC) cloth (Zorflex-double weave, Charcoal House LLC, USA) were used for electrode materials. The AC cloth was cleaned as described in our previous work [31]. 2.2. Characterization: The phase, morphology, and microstructures of electrode materials were analyzed by X-ray diffraction (XRD, Thermoscientific diffractometer, ARL Equinox 100 x-ray), 3
ACCEPTED MANUSCRIPT
and transmission electron microscopy (TEM, JEM-2800). The Brunauer-Emmett-Teller (BET, TriStar II Plus 2.03, Micromeritics Instruments Corp., USA) surface area analysis and surface
RI PT
porosity measurements (ASAP 2010, Micromeritics Instruments Corp., USA) were performed on MoO3 powders and AC cloth, respectively. The particle size distribution of MoO3 powders was evaluated from the particle size measured using the TEM images. X-ray photoelectron
SC
spectroscopy (XPS) was performed in a Kratos Axis ULTRA spectrometer (Materials Research Laboratory, University of Illinois). Binding energies were referenced to the graphitic C 1s peak at
M AN U
284.5 eV.
2.3. Electrochemical studies: The MoO3 electrodes were prepared using a slurry made of 80 wt% of MoO3 powder as the active material with 15 wt% of conductive carbon, 5 wt% of polyvinylidene difluoride binder, and N-methyl-2-pyrrolidone solvent. The slurry was then
TE D
coated onto a steel foil (thickness - 50µm) and dried at ~60 oC in a vacuum oven for 12 hours. The electrolytes used were aqueous 0.5M BeSO4 (Reagent - 98%, GFS Chemicals, USA) and aqueous 0.5M MgSO4 (Anhydrous, Reagent, GFS Chemicals, USA). All the electrochemical
EP
characterizations were carried out on a Solartron 1470E workstation (Solartron Instruments,
AC C
USA). Three-electrode tests were performed using a flat cell kit (Ametek Scientific Instruments, USA) with the active material (MoO3 or AC) as a working electrode, Ag/AgCl - reference electrode, and a platinum mesh - counter electrode. Two electrode tests on MoO3 against AC were performed using a stainless steel split cell (MTI Corporation, USA), with a glass microfiber separator (VWR, USA). All the electrochemical measurements were performed at room temperature (~25 oC).
4
ACCEPTED MANUSCRIPT
3. Results and Discussion
RI PT
The XRD pattern (Fig 1(a)) confirmed the orthorhombic phase of α-MoO3 matching with the JCPDS No. 05-0508 database with lattice parameters of a=3.96 Å, b=13.86 Å, and c=3.70 Å corresponding to the pure crystalline orthorhombic α-MoO3. The TEM image, see Fig. 1(b),
SC
showed single crystal nanoplatelets of α-MoO3 with thickness along [010] (b-axis). The BET measurements revealed a surface area of 2.59 m2g-1, which is very limited to contribute an
M AN U
appreciable amount of double layer capacitance. The surface area of AC electrodes was 1034 m2g-1 with a pore size from 3.9 to 30 Å (average ~ 17.4 Å), and more details on AC can be found in [31]. Figure 1(c) shows the particle size distribution of α-MoO3 powders with most of the
TE D
particles smaller than 2 µm, and average particle size was around 2.1 μm.
Initially, the stability of aqueous BeSO4 and MgSO4 electrolytes was evaluated through cyclic
EP
voltammetry (CV) studies using MoO3 and AC, separately, with a three electrode setup. The CV
AC C
scans for the first cycle of MoO3 with aqueous 0.5M BeSO4 (pH 2.2) and 0.5M MgSO4 (pH 6.2) are shown in Fig. 2(a). For a fair comparison of these two electrolytes, the CV scans were also obtained with reduced pH (2.2) of MgSO4, which was prepared by mixing dilute sulfuric acid with aqueous 0.5M MgSO4 (pH 6.2). All the three electrolytes showed multiple redox peaks, however, to distinguish these from H+ co-intercalation, the CV scans with dilute sulfuric acid electrolytes of pH 2.2 and 6.2 (Fig. 2(b)) were also obtained. Evidently, the redox couples associated with H+ insertion are clearly distinct from the redox couples of Be2+ and Mg2+ 5
ACCEPTED MANUSCRIPT
intercalation. The intercalation of Be2+ showed three pairs of redox couples with cathodic peaks at -0.34, 0. 11 and 0.58 V Vs Ag/AgCl, and anodic peaks at -0.072, -0.4 and -0.64 V Vs Ag/AgCl.
RI PT
The redox peaks for intercalation of Mg2+ were not discernible at pH - 6.2. A magnified view of the CV scan with aqueous 0.5M MgSO4 (pH 6.2) electrolyte is provided in the inset of Fig. 2(a). However, they became distinctive when the pH was reduced to 2.2, and exhibited a three stage
SC
intercalation with redox peaks at -0.45, 0.13 and 0.44 V Vs Ag/AgCl (cathodic), and -0.016, 0.28 and -0.66 Vs Ag/AgCl (anodic). These observations agree with multi-stage intercalation of
M AN U
Mg2+ and Al3+ into MoO3 as reported in the literature[19,23–27]. These peaks represent the redox transformations of Mo (IV) ↔ Mo (V) ↔ Mo (VI). The reaction can be described by the following equation, where A is the divalent ion being inserted into MoO3:
TE D
+ + 2 →
The peak current (i) of the CV curves and the sweep rate (v) follow the relationship i=avb. Where “a” and “b” are adjustable values, and “b” was determined from the slope of log (v) Vs
EP
log (i). In this study, the b-values were close to 0.5 for both Be2+ and Mg2+ indicating a diffusion
AC C
controlled mechanism (Fig S1). The aqueous BeSO4 showed stability between -0.75 to 1 V (Vs Ag/AgCl), whereas aqueous MgSO4 showed an increased stability with reduced pH, see Fig. 2(a). A similar dependency of electrolyte stability on pH was reported for Li-ion electrolytes by Yi Cui et al. [32].
6
ACCEPTED MANUSCRIPT
On the other hand, the CV scans of AC (not shown here) exhibited stability between -1 to 1 V (Vs Ag/AgCl) with double layer capacitance for all the three electrolytes. Fig. 2(c) shows the scan rate dependent specific capacity of MoO3 for all the three electrolytes. Interestingly, Be2+
RI PT
electrolyte yielded the highest specific capacity of 158.3 mAh g-1 (~285 Fg-1), whereas MgSO4 showed a lower value (68 mAhg-1) with a slight improvement (85 mAhg-1) after reducing the pH,
SC
which could be due to co-intercalation of H+ [14,30]. This was apparent from the CV studies of dilute sulfuric acid electrolytes as well, see Figs 2(b) and 2(c), where the specific capacity of
M AN U
MoO3 increased with reducing pH of the electrolyte. It should be noted that the capacitance measurements in three electrode tests are usually limited by the surface area of the counter electrode where double layer of anions would form. On the other hand, the AC electrode had a large surface area, and therefore, it must have facilitated in obtaining a higher capacitance in
TE D
the two electrode tests in the present studies. In addition, we performed the CV scans on bare steel foil substrate (i.e., the current collector) as well in three electrode configuration in order
EP
to ensure a negligible contribution of steel on the capacitance values measured (see Fig S2).
AC C
Following these studies, the AC electrodes were tested in a symmetric configuration with 0.5M BeSO4 (pH 2.2), 0.5M MgSO4 (pH 6.2) and 0.5M MgSO4 (pH 2.2), and obtained specific capacitances around 46, 22 and 34.2 Fg-1, respectively. Based on these, a mass ratio of 30 to 1 between AC and MoO3 was used for all the asymmetric cells (i.e., two electrode tests) to avoid limiting effects of AC on the specific capacity measurements of MoO3. Figure 3 shows the CV scans for the first cycle and scan rate dependent capacity of MoO3 electrode from AC Vs MoO3
7
ACCEPTED MANUSCRIPT
tests with different electrolytes. The CV scans (Fig 3(a)) show three stage redox peaks (barely visible) agreeing with the three electrode tests, and broad nature of these scans can be attributed to a combination of intercalation and double layer formation at MoO3 and AC
RI PT
electrodes, respectively. The MoO3 electrode exhibited a specific capacity of 335 mAh g-1 (~928 Fg-1) with aqueous BeSO4 electrolyte (@2 mVs-1), and it is considerably larger when compared with the three electrode capacity due to the large surface area provided by the AC electrode for
SC
double layer formation of anions while leading to a larger amount of Be2+ intercalation into
M AN U
MoO3. On the other hand, aqueous MgSO4 electrolytes resulted in specific capacities of 72.5 and 88.4 mAhg-1 for pH 6.2 and 2.2 (Fig 3(b)), respectively, reflecting a minimal influence of H+ intercalation. However, these values are only a slight improvement over the three-electrode test data, which certainly indicates a limited intercalation of Mg2+ into MoO3 when compared
TE D
with Be2+. The impedance data in Fig. 3(c) also depicted a larger Warburg component for MgSO4 electrolytes implying more resistance to the to the bulk diffusion of Mg2+. The Nyquist plots obtained were modeled and interpreted with the help of the appropriate equivalent circuit (see
EP
figure 3(e)). Where Relectrolyte is the ohmic resistance at the electrode/electrolyte interface, Rfaradic is the charge transfer resistance of faradic processes, RL is the leakage resistance, W is
AC C
the Warburg element, CPEdl and CPEfaradic are constant phase elements of the double layer and faradic reactions, respectively. The elements of the equivalent circuit model fitted to the impedance spectra are shown in table 1 for all the BeSO4 and MgSO4 electrolytes. The results indicate negligible differences in electrolyte resistances and double layer capacitance values. However, the capacitance contribution by the Warburg element is up to three times larger with Be2+ electrolyte when compared to Mg2+. This corroborates our findings from the CV tests
8
ACCEPTED MANUSCRIPT
representing more intercalation of Be2+ into the electrode compared to Mg2+, which was also confirmed in our later studies through GITT studies (see Fig. 4). These observations clearly
EIS parameters from the equivalent circuit for α-MoO3 in BeSO4 and MgSO4
electrolytes R-electrolyte (Ω cm2)
R-faradic (Ω cm2)
2.2 6.2 2.2
1.506 1.676 3.252
3.176 8.131 4.51
R- L (Ω cm2)
CPE-dl (farads cm-2)
CPE-faradic (farads cm-2)
W (farads cm-2)
22.92 104.6 172.3
0.6305 0.0001436 0.00239
5.139 0.22158 0.21785
122.4 35.87 40.85
M AN U
BeSO₄ MgSO₄ MgSO₄
pH
SC
Table 1.
RI PT
represent more resistance to the diffusion of Mg2+ into MoO3 electrode than Be2+.
The cyclability data presented a poor retention of specific capacity of MoO3 with high
TE D
coulombic efficiency for all the electrolytes (Fig 3(d)). The literature [3,4,7] also suggests that most of the electrode materials suffer from limited and fading capacities with multivalent
EP
electrolytes due to stronger electrostatic interactions of multivalent ions with the crystal structures of the electrodes resulting in a reduced mobility and possibly collapse of the crystal
AC C
structures of the electrode materials. In addition, a drastic localized deformation of the electrodes caused by abrupt changes in the oxidation states with partial trapping of ions was reported to be one of the major reasons for poor cyclability [3,7,33–37]. Therefore, we believe that similar and strong electrostatic interactions of Be2+ and Mg2+ with MoO3 layer structures may have resulted in poor cyclability. Be2+ being a smaller ion with more charge density compared to Mg2+ was expected to have stronger electrostatic interactions with the layers of
9
ACCEPTED MANUSCRIPT
MoO3. However, unlike the alkaline earth metal ions (e.g. Mg2+, Ca2+ and Sr2+, etc.,), Be2+ is unique with both ionic and covalent nature [38], and this could have possibly resulted in
RI PT
different characteristics and thereby probably a slightly better performance.
We also performed ex-situ XRD studies on the MoO3 samples fully charged/discharged at 6.25
SC
Ag-1 using aqueous 0.5M BeSO4 (pH 2.2) and 0.5M MgSO4 (pH 6.2) electrolytes in order to understand the structural changes of MoO3 active material. After the first full charge, the
M AN U
sample tested in 0.5M BeSO4 showed a shift in the diffracted peaks to lower angles along with broadening of the peaks (see Fig. 4(a)). The shift in (020) peak can be ascribed to an increase in the interlayer spacing. Additionally the XRD pattern exhibits new peaks at 2θ = 18.5o, 23.8o, 31.3o, and 43.7o which can potentially be associated with the insertion of beryllium and the
TE D
formation of a new BexMoO3 phase, similar to those reported for Li+, Mg2+ and Al3+ [19,27,39]. After one full charge/discharge cycle, the original peaks did not fully return to their pristine
EP
condition, and the new peaks were still present with reduced intensities. Similarly, the MoO3 electrodes tested in 0.5M MgSO4 electrolyte showed extra peaks in the XRD patterns along with
AC C
a considerable shift in the peaks to lower angles and peak broadening (see Fig. 4(b)) after the full charge cycle. These new peaks appeared at 2θ =13.76o, 22.58 o, 29.87 o and 44.23 o, which can potentially be associated with the formation of a magnesiated phase (MgxMoO3) as observed in prior reports by Aurbach et al. [27]. These results are similar to the ones obtained with BeSO4. However, the remaining new peaks were much more intense for MgSO4 electrolyte indicating to the possibility of partial entrapment of ions as reported by Aurbach et at [27].
10
ACCEPTED MANUSCRIPT
To further elucidate the electrochemical activity of MoO3 electrodes, we performed XPS studies on the MoO3 samples fully charged/discharged at 6.25 Ag-1 using aqueous 0.5M BeSO4 (pH 2.2)
RI PT
and 0.5M MgSO4 (pH 6.2) electrolytes. Fig 4(c) shows the XPS spectra of MoO3 electrodes after a full charge and discharge with aqueous 0.5M BeSO4 electrolyte. The peaks at ~233.0 eV and ~236.1 eV correspond to the characteristic Mo 3d spin-orbit doublet peaks of MoO3 [11]. After one full charge in 0.5M BeSO4, new peaks appear at ~231.6 and 234.7 eV, between MoO2 and
SC
MoO3, these peaks can be associated with the insertion of Be2+ forming a MoO2-BeO complex.
M AN U
After the completion of first discharge cycle the new peaks remained but reduced quite a bit in intensity. Similarly, the MoO3 samples tested with 0.5M MgSO4 (see Fig. S3) also showed a new peak at ~231.6 eV corresponding to the insertion of Mg2+ with formation of MoO2-MgO complexes. These results are consistent with the extra peaks appeared in the XRD spectra,
TE D
which indicates that the insertion of Be2+ and Mg2+ may not be completely reversible. In fact some of this irreversibility was evident to certain extent even in the columbic efficiency data presented in Fig. 3(d), where the columbic efficiency gradually increased with increasing
AC C
EP
number of cycles and decay in the capacitance.
To gain further insights into the diffusion of both Be2+ and Mg2+ in MoO3, the diffusion coefficients of both ions were evaluated through GITT experiments (Fig S4) using the following equation
! 4 $ = ! √#
11
ACCEPTED MANUSCRIPT Here, - current, - molar volume of the electrode, - number of charges, - Faraday's
constant, - contact area between electrode and electrolyte, dE/dδ - slope of the steady state
voltages, and dE/d√t - slope of the linearized potential E(V) during the current pulse duration of
RI PT
t(s). Figure 4(d) shows the variation of diffusion coefficients of Be2+ and Mg2+ with respect to their concentration in MoO3. The diffusion coefficients of Be2+ and Mg2+ were determined to vary from 1.9 x 10-11 (at x = 0 for BexMoO3) to 6.4 x 10-12 cm2s-1, and 1.3 x 10-11 (at x = 0 for
SC
MgxMoO3) to 4.8 x 10-13 cm2s-1, respectively. The diffusion coefficients of Mg2+ and Be2+ are on
M AN U
the lower side when compared with the diffusion coefficients values reported for Li+ in MoO3 (10-9 to 10-12 cm2s-1)[40–42]. This can be attributed to the larger electrostatic interactions of divalent ions with the lattice structure of electrode material. At x=0, the diffusion coefficients of Be2+ and Mg2+ were almost equal suggesting similar levels of resistance at the liquid
TE D
electrolyte and electrode interface. However, with increasing the x value, the diffusion coefficient of Mg2+ dropped almost by two orders of magnitude, whereas Be2+ showed a minimal drop. The reason for such a contrasting behavior could be related to the diffusion
EP
mechanisms of both ions. According to Ziqin Rong et al. [3], the hopping mechanism of a multivalent ion in a given electrode follows a lower energy path depending on the crystal
AC C
structure of the host as well as preferred coordination number of the multivalent ion. In layer structured electrodes, the diffusion of multivalent ions is expected to follow octahedraltetrahedral-octahedral path as shown in Fig. 4(e). However, the rate of hopping is dictated by the preferred coordination number of the multivalent ion. For example, Li+ and Zn2+ with a coordination number of four can diffuse easily compared to the ions that have higher coordination number [3], for example, Mg2+ and Ca2+, which have preferred coordination
12
ACCEPTED MANUSCRIPT
numbers six and eight, respectively [3,43,44] diffuse slowly compared to Li+ and Zn2+. The crystallography literature [45] suggests that the preferred coordination number of Be2+ is four,
RI PT
and therefore, when compared with Mg2+ the hopping of Be2+ through octahedral-tetrahedraloctahedral path would very likely be favored in layer structured MoO3. On the other hand, the unique features of Be2+ with ionic and covalent nature, unlike Mg2+, would also likely result in weaker electrostatic interactions and thereby probably result in slightly better characteristics.
SC
This could be the same reason why Mg2+ intercalation did not show much improvement in the
M AN U
capacity in two electrode tests in comparison with the three electrode test data. In addition, higher impedance in the Warburg region along with GITT diffusion measurements also
4. Conclusions
TE D
substantiate that the diffusion of Mg2+ will be slower than Be2+.
The intercalation characteristics of α-MoO3 were investigated with Be2+ and Mg2+ of different
EP
sizes. Be2+ showed the highest capacity for the MoO3/AC asymmetric cell, and it can be ascribed to peculiar characteristics of Be2+ with covalent and ionic nature, and a preferred coordination
AC C
number of four, which may have resulted in a higher diffusion coefficient of Be2+ in MoO3. The high capacitive behavior displayed by Be2+ presents an opportunity for improved energy densities of supercapacitors when compared against Mg2+ and possibly other multivalent ions. However, understanding of new phases with insertion of Be2+ into MoO3 needs further investigation, which will be presented in the future publications.
13
ACCEPTED MANUSCRIPT
Acknowledgements Juan C. Icaza would like to thank the Dept. of Mechanical Engineering at Lamar University for
RI PT
the Doctoral Fellowship. Also, this work is supported by the start-up funds provided by Lamar University to the faculty Dr. Ramesh K. Guduru. XPS analysis was carried out in part at the
SC
Frederick Seitz Materials Research Laboratory Central Research Facilities, University of Illinois.
References
M.A. Kiani, M.F. Mousavi, M.S. Rahmanifar, Synthesis of nano- and micro-particles of
M AN U
[1]
LiMn2O4: Electrochemical investigation and assessment as a cathode in li battery, Int. J. Electrochem. Sci. 6 (2011) 2581–2595. doi:10.1039/c1ee01598b. A. González, E. Goikolea, J.A. Barrena, R. Mysyk, Review on supercapacitors:
TE D
[2]
Technologies and materials, Renew. Sustain. Energy Rev. 58 (2016) 1189–1206. doi:10.1016/j.rser.2015.12.249.
Z. Rong, R. Malik, P. Canepa, G. Sai Gautam, M. Liu, A. Jain, K. Persson, G. Ceder,
EP
[3]
AC C
Materials Design Rules for Multivalent Ion Mobility in Intercalation Structures, Chem. Mater. 27 (2015) 6016–6021. doi:10.1021/acs.chemmater.5b02342.
[4]
P. Canepa, G. Sai Gautam, D.C. Hannah, R. Malik, M. Liu, K.G. Gallagher, K.A. Persson, G.
Ceder, Odyssey of Multivalent Cathode Materials: Open Questions and Future Challenges, Chem. Rev. 117 (2017) 4287–4341. doi:10.1021/acs.chemrev.6b00614.
[5]
C. Zhao, W. Zheng, A Review for Aqueous Electrochemical Supercapacitors, Front. Energy 14
ACCEPTED MANUSCRIPT
Res. 3 (2015) 1–11. doi:10.3389/fenrg.2015.00023. [6]
G. Wang, L. Zhang, J. Zhang, A review of electrode materials for electrochemical
[7]
RI PT
supercapacitors, Chem. Soc. Rev. 41 (2012) 797–828. doi:10.1039/c1cs15060j. R. Guduru, J. Icaza, A Brief Review on Multivalent Intercalation Batteries with Aqueous
[8]
SC
Electrolytes, Nanomaterials. 6 (2016) 41. doi:10.3390/nano6030041.
A.G. Volkov, S. Paula, D.W. Deamer, Two mechanisms of permeation of small neutral
M AN U
molecules and hydrated ions across phospholipid bilayers, Bioelectrochemistry Bioenerg. 42 (1997) 153–160. doi:10.1016/S0302-4598(96)05097-0. [9]
N.A. Chernova, M. Roppolo, A.C. Dillon, M.S. Whittingham, Layered vanadium and molybdenum oxides: batteries and electrochromics, J. Mater. Chem. 19 (2009) 2526–
[10]
TE D
2552. doi:10.1039/b819629j.
T. Brezesinski, J. Wang, S.H. Tolbert, B. Dunn, Ordered mesoporous alpha-MoO3 with iso-
EP
oriented nanocrystalline walls for thin-film pseudocapacitors., Nat. Mater. 9 (2010) 146– 151. doi:10.1038/nmat2612.
J. Jiang, J. Liu, S. Peng, D. Qian, D. Luo, Q. Wang, Z. Tian, Y. Liu, Facile synthesis of alpha-
AC C
[11]
MoO3 nanobelts and their pseudocapacitive behavior in an aqueous Li2SO4 solution, J. Mater. Chem. A. 1 (2013) 2588–2594. doi:10.1039/c2ta01120d.
[12]
N. Anbananthan, K.N. Rao, V.K. Venkatesan, Electrochemical formation of oxygendeficient molybdenum oxide, Appl. Surf. Sci. 72 (1993) 189–194. doi:10.1016/01694332(93)90011-Y. 15
ACCEPTED MANUSCRIPT
[13]
I. Shakir, M. Shahid, H.W. Yang, D.J. Kang, Structural and electrochemical characterization of α-MoO3 nanorod-based electrochemical energy storage devices, Electrochim. Acta. 56
[14]
RI PT
(2010) 376–380. doi:10.1016/j.electacta.2010.09.028. B. Mendoza-Sánchez, T. Brousse, C. Ramirez-Castro, V. Nicolosi, P. S. Grant, An
investigation of nanostructured thin film α-MoO3 based supercapacitor electrodes in an
doi:10.1016/j.electacta.2012.11.127.
Z. Cui, W. Yuan, C.M. Li, Q. Zhang, T. Wei, J. Feng, M. Zhang, W. Qian, F. Wei, A.C. Dillon,
M AN U
[15]
SC
aqueous electrolyte, Electrochim. Acta. 91 (2013) 253–260.
Template-mediated growth of microsphere, microbelt and nanorod α-MoO3 structures and their high pseudo-capacitances, J. Mater. Chem. A. 1 (2013) 12926-12931.
[16]
TE D
doi:10.1039/c3ta12688a.
L. Zheng, Y. Xu, D. Jin, Y. Xie, Q. Song, J. Guo, L.Q. Chen, R. Seshadri, G.D. Stucky, Wellaligned molybdenum oxide nanorods on metal substrates: solution-based synthesis and
EP
their electrochemical capacitor application, J. Mater. Chem. 20 (2010) 7135-7143.
[17]
AC C
doi:10.1039/c0jm00744g.
X. Xia, Q. Hao, W. Lei, W. Wang, H. Wang, X. Wang, Reduced-graphene
oxide/molybdenum oxide/polyaniline ternary composite for high energy density supercapacitors: Synthesis and properties, J. Mater. Chem. 22 (2012) 8314-8320. doi:10.1039/c2jm16216d.
[18]
J. Chang, M. Jin, F. Yao, T.H. Kim, V.T. Le, H. Yue, F. Gunes, B. Li, A. Ghosh, S. Xie, Y.H. Lee,
16
ACCEPTED MANUSCRIPT
Asymmetric supercapacitors based on graphene/MnO2 nanospheres and graphene/MoO3 nanosheets with high energy density, Adv. Funct. Mater. 23 (2013)
[19]
RI PT
5074–5083. doi:10.1002/adfm201301851. Y.P. Wu, F. Wang, X. Wang, X. Yuan, Z. Liu, X. Wu, Y. Zhu, L. Fu, A conductive polymer coated MoO3 anode enables an Al-ion capacitor with high performance, J. Mater. Chem.
[20]
SC
A. 4 (2016) 5115–5123. doi:10.1039/C6TA01398H.
I. Shakir, M. Shahid, U.A. Rana, M.F. Warsi, Y. Xie, G. Yu, W. Xing, Z. Lin, W. Wu, T. Li, H.
M AN U
Jin, P. Liu, J. Zhou, C.P. Wong, Z.L. Wang, In situ hydrogenation of molybdenum oxide nanowires for enhanced supercapacitors, RSC Adv. 4 (2014) 8741-8745. doi:10.1039/c3ra44837a.
I. Shakir, M. Shahid, M. Nadeem, D.J. Kang, Tin oxide coating on molybdenum oxide
TE D
[21]
nanowires for high performance supercapacitor devices, Electrochim. Acta. 72 (2012) 134–137. doi:10.1016/j.electacta.2012.04.002. Y. Liu, B. Zhang, Y. Yang, Z. Chang, Z. Wen, Y. Wu, Polypyrrole-coated α-MoO3 nanobelts
EP
[22]
AC C
with good electrochemical performance as anode materials for aqueous supercapacitors, J. Mater. Chem. A. 1 (2013) 13582. doi:10.1039/c3ta12902k.
[23]
M.E. Spahr, P. Nov?k, O. Haas, R. Nesper, Electrochemical insertion of lithium, sodium,
and magnesium in molybdenum(VI) oxide, J. Power Sources. 54 (1995) 346–351. doi:10.1016/0378-7753(94)02099-O.
[24]
T.S. Sian, G.B. Reddy, S.M. Shivaprasad, Study of Mg ion Intercalation in Polycrystalline 17
ACCEPTED MANUSCRIPT
MoO 3 Thin Films, Jpn. J. Appl. Phys. 43 (2004) 6248–6251. doi:10.1143/JJAP.43.6248. [25]
T.S. Sian, G.B. Reddy, S.M. Shivaprasad, Effect of Microstructure and Stoichiometry on
RI PT
Absorption in Mg Intercalated MoO3 Thin Films, Electrochem. Solid-State Lett. 9 (2006) A120-A122. doi:10.1149/1.2163427. [26]
L. Campanella, G. Pistoia, MoO3: A New Electrode Material for Nonaqueous Secondary
G. Gershinsky, H.D. Yoo, Y. Gofer, D. Aurbach, Electrochemical and Spectroscopic Analysis
M AN U
[27]
SC
Battery Applications, J. Electrochem. Soc. 118 (1971) 1905-1908. doi:10.1149/1.2407864.
of Mg 2+ Intercalation into Thin Film Electrodes of Layered Oxides: V2O5 and MoO3, Langmuir. 29 (2013) 10964–10972. doi:10.1021/la402391f. [28]
W. Tang, L. Liu, S. Tian, L. Li, Y. Yue, Y. Wu, K. Zhu, Aqueous supercapacitors of high
TE D
energy density based on MoO3 nanoplates as anode material, Chem. Commun. 47 (2011) 10058-10060. doi:10.1039/c1cc13474d.
H. Farsi, F. Gobal, H. Raissi, S. Moghiminia, On the pseudocapacitive behavior of
EP
[29]
nanostructured molybdenum oxide, J. Solid State Electrochem. 14 (2010) 643–650.
[30]
AC C
doi:10.1007/s10008-009-0830-5. H. Farsi, F. Gobal, H. Raissi, S. Moghiminia, The pH effects on the capacitive behavior of
nanostructured molybdenum oxide, J. Solid State Electrochem. 14 (2010) 681–686. doi:10.1007/s10008-009-0828-z.
[31]
J.C. Icaza, R.K. Guduru, Effect of ion charges on the electric double layer capacitance of activated carbon in aqueous electrolyte systems, 336 (2016) 360-366. 18
ACCEPTED MANUSCRIPT
doi:10.1016/j.jpowsour.2016.11.017. [32]
C. Wessells, R. Ruffο, R.A. Huggins, Y. Cui, Investigations of the Electrochemical Stability
RI PT
of Aqueous Electrolytes for Lithium Battery Applications, Electrochem. Solid-State Lett. 13 (2010) A59-A61. doi:10.1149/1.3329652. [33]
M. Liu, Z. Rong, R. Malik, P. Canepa, A. Jain, G. Ceder, K.A. Persson, Spinel compounds as
SC
multivalent battery cathodes: a systematic evaluation based on ab initio calculations,
[34]
M AN U
Energy Environ. Sci. 8 (2015) 964–974. doi:10.1039/C4EE03389B.
R.Y. Wang, B. Shyam, K.H. Stone, J.N. Weker, M. Pasta, H.-W. Lee, M.F. Toney, Y. Cui, Reversible Multivalent (Monovalent, Divalent, Trivalent) Ion Insertion in Open Framework Materials, Adv. Energy Mater. 5 (2015) 1401869.
[35]
TE D
doi:10.1002/aenm.201401869.
B. Lee, C.S. Yoon, H.R. Lee, K.Y. Chung, B.W. Cho, S.H. Oh, Electrochemically-induced reversible transition from the tunneled to layered polymorphs of manganese dioxide,
E. Levi, M.D. Levi, O. Chasid, D. Aurbach, A review on the problems of the solid state ions
AC C
[36]
EP
Scientific Reports. 4 (2014) 6066. doi:10.1038/srep06066.
diffusion in cathodes for rechargeable Mg batteries, J. Electroceramics. 22 (2009) 13–19. doi:10.1007/s10832-007-9370-5.
[37]
H.D. Yoo, I. Shterenberg, Y. Gofer, G. Gershinsky, N. Pour, D. Aurbach, Mg rechargeable batteries: an on-going challenge, Energy Environ. Sci. 6 (2013) 2265-2279. doi:10.1039/c3ee40871j. 19
ACCEPTED MANUSCRIPT
[38]
C.W. Bock, J.P. Glusker, Organization of water around a beryllium cation, Inorg. Chem. 32 (1993) 1242–1250. doi:10.1021/ic00059a036. T. Tsumura, M. Inagaki, Lithium insertion/extraction reaction on crystalline MoO3, Solid
RI PT
[39]
State Ionics. 104 (1997) 183–189. doi:10.1016/S0167-2738(97)00418-9. [40]
G. Guzman, B. Yebka, J. Livage, C. Julien, Lithium intercalation studies in hydrated
SC
molybdenum oxides, Solid State Ionics. 86–88 (1996) 407–413. doi:10.1016/0167-
[41]
M AN U
2738(96)00338-4.
C. Julien, G.A. Nazri, Transport properties of lithium-intercalated MoO3, Solid State Ionics. 68 (1994) 117–123. doi:10.1016/0167-2738(94)90245-3.
[42]
C. Julien, A. Khelfa, J.P. Guesdon, A. Gorenstein, Lithium intercalation in MoO3: A
TE D
comparison between crystalline and disordered phases, Appl. Phys. A Solids Surfaces. 59 (1994) 173–178. doi:10.1007/BF00332212. M. Wenger, T. Armbruster, Crystal chemistry of lithium: oxygen coordination and
EP
[43]
bonding, Eur. J. Mineral. 3 (1991) 387–400. doi:10.1127/ejm/3/2/0387. I.D. Brown, What factors determine cation coordination numbers?, Acta
AC C
[44]
Crystallographica. Sect. B Struct. Sci. 44 (1988) 545–553. doi:10.1107/S0108768188007712.
[45]
W.W. Emmett, Geological Survey Professional Paper, J. Hydrol. 2 (1965) 274-275. http://dx.doi.org/10.1016/0022-1694(65)90048-x..
20
ACCEPTED MANUSCRIPT
SC
RI PT
(a)
AC C
EP
TE D
M AN U
(b)
1
ACCEPTED MANUSCRIPT
Figure 1.
(a) X-ray diffraction pattern of as-received commercial MoO3 powders, (b) TEM image of MoO3 powders with selective area diffraction pattern and high-
AC C
EP
TE D
M AN U
SC
commercial MoO3 powders.
RI PT
resolution lattice image, and (c) particle size distribution of as-received
2
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure 2.
(a) CV scans of MoO3 in three electrode configuration. The inset – magnified view of CV scan with aqueous 0.5M MgSO4, (b) CV scans with dilute sulfuric acid, and (c) variation of capacity with respect to the scan rate.
1
AC C
(e)
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
1
ACCEPTED MANUSCRIPT
Figure 3.
(a) CV scans of MoO3 Vs AC asymmetric cells, (b) specific capacity of MoO3 with respect to the scan rate, and (c) impedance of BeSO4 and MgSO4 electrolytes,
RI PT
and zoom in view shown in the top-right corner, (d) cyclability of specific capacity of MoO3, and (e) the equivalent circuit model used to fit the impedance
AC C
EP
TE D
M AN U
SC
spectroscopy data.
2
ACCEPTED MANUSCRIPT
st
RI PT
1 discharging to 0V
st
1 charging to 1V
Mo (6+) (233) Mo (6+) (236.1)
(e)
AC C
EP
Mo (5+) (234.7)
TE D
Mo (5+) (231.6)
M AN U
SC
Pristine
1
ACCEPTED MANUSCRIPT
Figure 4.
XRD patterns of pristine MoO3, MoO3 after the 1st charge and MoO3 after the 1st discharge at 6.25 Ag-1 for (a) 0.5M BeSO4 and (b) 0.5M MgSO4-pH 6.2. (c) XPS spectra in the Mo 3d binding energy region of α-MoO3 after charging/discharging
RI PT
in 0.5M BeSO4 (d) Variation of diffusion coefficients of Be2+ (electrolyte -0.5M BeSO4) and Mg2+ (0.5M MgSO4: pH-6.2) as a function of concentration Be2+ in
SC
MoO3, (e) octahedral-tetrahedral-tetrahedral diffusion topology in layer
AC C
EP
TE D
M AN U
structured electrode.
2
ACCEPTED MANUSCRIPT
Highlights
Comparison of intercalation behavior of divalent ions - Be2+ and Mg2+ in α-MoO3 Effect of divalent ion charge density on the capacitance of α-MoO3 Effect of divalent ion coordination number on diffusion path topology
AC C
EP
TE D
M AN U
SC
RI PT
• • •