Research Article www.acsami.org
Relaxation-Induced Memory Effect of LiFePO4 Electrodes in Li-Ion Batteries Jianfeng Jia,† Chuhao Tan,† Mengchuang Liu,† De Li,*,†,‡ and Yong Chen*,†,‡ †
State Key Laboratory on Marine Resource Utilization in South China Sea, Hainan Provincial Key Laboratory of Research on Utilization of Si−Zr−Ti Resources, Materials and Chemical Engineering, Hainan University, 58 Renmin Road, Haikou 570228, China ‡ Key Laboratory of Advanced Energy Materials Chemistry (Ministry of Education), Nankai University, Tianjin 300071, China S Supporting Information *
ABSTRACT: In Li-ion batteries, memory effect has been found in several commercial two-phase materials as a voltage bump and a step in the (dis)charging plateau, which delays the two-phase transition and influences the estimation of the state of charge. Although memory effect has been first discovered in olivine LiFePO4, the origination and dependence are still not clear and are critical for regulating the memory effect of LiFePO4. Herein, LiFePO4 has been synthesized by a home-built spray drying instrument, of which the memory effect has been investigated in Li-ion batteries. For assynthesized LiFePO4, the memory effect is significantly dependent on the relaxation time after phase transition. Besides, the voltage bump of memory effect is actually a delayed voltage overshooting that is overlaid at the edge of stepped (dis)charging plateau. Furthermore, we studied the kinetics of LiFePO4 electrode with electrochemical impedance spectroscopy (EIS), which shows that the memory effect is related to the electrochemical kinetics. Thereby, the underlying mechanism has been revealed in memory effect, which would guide us to optimize two-phase electrode materials and improve Li-ion battery management systems. KEYWORDS: Li-ion battery, LiFePO4, memory effect, relaxation, overshooting
1. INTRODUCTION Nowadays, fossil energy consumption has aroused more and more concerns regarding global societies, which have caused serious environmental issues. Although rechargeable batteries have been developed for centuries, they are mainly used in small and auxiliary energy-consuming devices. Fortunately, Li-ion batteries (LIBs) were invented in the end of last century, which are not only applied widely in portable electronic devices but also adopted promisingly in electric vehicles, electricity storage stations, and smart grids.1 To substitute for traditional fossil energy, present LIBS need to be significantly improved in the properties of energy density, power density, rate performance, cycle life, and so on.2 Up to now, few concerns has been paid to the memory effect of LIBs,3 which is a notable disadvantage in nickel cadmium (Ni−Cd) and nickel−metal hydride (Ni−MH) batteries.4−6 In LIBs, memory effect has be found in several commercial two-phase materials7−9 as a voltage bump and step in the (dis)charging plateau, which delays the two-phase transition and influences the estimation of the state of charge. Olivine LiFePO4 cathode material has been commercially used in LIBs because of its excellent electrochemical performances, low raw-materials cost, environmental friendliness, and safety.10−13 However, LiFePO4 also possesses the memory effect, which was first reported in LIBS 4 years ago.7 In the © XXXX American Chemical Society
memory effect, LiFePO4 electrode can memorize a fraction of capacity that undergoes extra (dis)charging cycles, and this memory can be read and released as a voltage bump and step in the consequent (dis)charging plateau. It is well-known that the olivine LiFePO4 takes a two-phase transition between the lithiated phase (Li1−βFePO4, β-phase) and the delithiated phase (LiαFePO4, α-phase), corresponding to the plateau of the (dis)charging curve.14−17 According to the particle-by-particle model,18,19 partial LiFePO4 particles that undergoes extra (dis)charging cycles go through the phase transition first, while the other particles will be left in a certain phase for quite a long time. For the memory effect of LIBs, despite some apparent descriptions, the origination and dependence are still not clear and are critical for regulating the memory effect.1,7,9 In this work, LiFePO4 precursor with a hollow spherical morphology has been synthesized by a home-built spray drying instrument,20 and then calcined in a reductive atmosphere with sucrose additive to obtain hollowed-out LiFePO4 particles. By modifying the relaxation time and the initial voltage overshooting, we investigated the memory effect of as-synthesized Received: April 27, 2017 Accepted: June 28, 2017 Published: June 28, 2017 A
DOI: 10.1021/acsami.7b05852 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces
Figure 1. Schematic illustration of home-built spray-drying instrument.
Figure 2. SEM images of (a) LiFePO4 precursor by the spray-drying method and (b) as-synthesized LiFePO4 by calcining at 700 °C for 5 h. (c) XRD patterns of LiFePO4 precursor (black line) and as-synthesized LiFePO4 (red line). All diffraction peaks are indexed for olivine LiFePO4. (d) Raman spectrum of as-synthesized LiFePO4. mainly consists of a heating platform, an Al container with thermal insulator, and a quartz nebulizer. The precursor solution was prepared by successively adding 0.016 mol Fe(NO3)3·H2O (Aladdin, 99.99%), 0.016 mol LiH2PO4 (Macklin, 99%), 0.008 mol LiOH·H2O (Macklin, ≥ 99.0%), 0.048 mol HNO3 acid (65−68%), and 0.16 mol H2O2 (Macklin 30 wt %) into 200 mL of deionized water. Here, an extra 5% of lithium can compensate for the lithium evaporation during the process of hightemperature calcination, the nitric acid can suppress the precipitation in the precursor solution, and the hydrogen peroxide solution can passivate the Al foil to resist the acidic corrosion. Then, the temperature of the heating platform was set at 400 °C, 30 min were allowed to pass until the
LiFePO4 in LIBS. As a result, the memory effect is significantly dependent on the relaxation time after phase transition, and the voltage bump is actually a delayed voltage overshooting. With electrochemical impedance spectroscopy (EIS),21 we found that the memory effect might originate from the electrical and ionic kinetics of the LiFePO4 electrode.
2. EXPERIMENTAL SECTION 2.1. Materials Synthesis. LiFePO4 precursors were synthesized by a home-built spray drying instrument, as shown in Figure 1, which B
DOI: 10.1021/acsami.7b05852 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces
Figure 3. Demonstration of memory effect in LiFePO4. (a) For memory effect during charging, the memory-writing process was a discharge to 2.8 V (black), partial charge for 3 h (red), and discharge to 2.8 V (green); the memory-releasing process was a full charge to 4.0 V (blue), where the current rate is C/10 and the x axis of time is better for showing the sequence. (b) Enlarged view of (a) between 3.44 and 3.47 V, where the capacity is adopted in the x axis, in favor of the comparison between different charge curves. (c) For memory effect during discharging, the memory-writing process was a charge to 4.0 V (black), partial discharge for 3 h (red), and a charge to 4.0 V (green); the memory-releasing process was a full discharge to 2.8 V (blue). (d) Enlarged view of (c) between 3.38 and 3.41 V. instrument temperature became stable, and the precursor solution was nebulized pneumatically at a feeding rate of 2 mL·min−1 for about 2 h. The mist dried quickly to form hollow spherical microparticles inside the Al container, most of which sedimentate on the bottom of Al-foilcovered heating platform; meanwhile, the waste humid gas permeated outward through the edge of Al container. After the spray-drying process, the yellowish-brown powder was collected, mixed with 15 wt % sucrose additive, and ground in an agate mortar completely, and the mixture was calcined at high temperatures in a tube furnace filled with Ar + 5% H2 atmosphere to obtain LiFePO4 samples. 2.2. Characterization. The morphology and crystal structure were characterized by scanning electron microscopy (SEM, Phenom ProX) and powder X-ray diffraction (XRD) using Cu Kα radiation (Bruker D2 PHASER), respectively. The vibrational, rotational and other lowfrequency modes were observed by Raman spectroscopy (Thermo Fisher Scientific DXRxi). Electrochemical tests were conducted by using coin-type cells (CR2025). In the working electrode, a composite paste (ϕ 7 mm), containing 80 wt % LiFePO4, 10 wt % Ketjen black, and 10 wt % polytetrafluoroethylene (PTFE), was firmly pressed on an Al mesh (100 mesh) with a mass loading of approximately 5 mg cm−2. The counter electrode of lithium metal was separated from the working electrode by a glass-fiber film (Whatman GF/A), and 1 M LiClO4 in ethylene carbonate−diethyl carbonate (EC/DEC, 1:1 by volume) was used as the electrolyte. After all of the components were dried, the cells were assembled in a glovebox filled with Ar gas (H2O, O2 < 1 PPM). In a highand low-temperature test chamber, galvanostatic measurements were carried out on battery testing systems (Newware CT-9004), and EIS measurements were performed on a CellTest 8 channel potentiostat− galvanostat with a impedance analyzer (Solartron 1470E and 1260A).
carbon in LiFePO4 product. According to XRD patterns in Figure 2c, the LiFePO4 precursor is evidently amorphous without any diffraction peaks, and the LiFePO4 product is wellcrystallized because olivine LiFePO4 for all diffraction peaks match well with standard LiFePO4 values (JCPDS 83−2092). Besides, the Raman spectrum of LiFePO4 product is further confirmed its components, as shown in Figure 2d. in which three Raman bands above 900 cm−1 are assigned to intramolecular (PO4)3− symmetric (951 cm−1) and antisymmetric (996 and 1068 cm−1) stretching modes, the bands between 400−650 cm−1 are assigned to combined antisymmetric stretching modes, and the bands between 200−400 cm−1 are assigned to translatory intermolecular modes involving lattice vibrations of (PO4)3− units and Fe2+ ions.22,23 All of the bands match well with reported literature. Galvanostatic measurements in special consequences were conducted to examine the memory effect of as-synthesized LiFePO4. Panels a and c of Figure 3 show a typical memory effect procedure. In the memory-writing process, the LiFePO4 electrode takes a partial cycle after a full discharging or charging, followed by a full charging or discharging to read and release the memory. Compared to a normal (dis)charging curve, the memory-releasing process exhibits as a two-stepped voltage curve, in which the partial capacity with an extra cycle shows a slightly lower overpotential while the rest capacity has an overpotential that is much higher, as shown in Figure 3b,d. Although charging and discharging are apparent symmetrical for such typical two-phase transition, the memory effect is more evident during charging in comparison with discharging, with the step of 1.9 mV for charging and 0.8 mV for discharging. Thus, the memory effect in as-synthesized LiFePO4 has been confirmed. In previous reports7 and here for LiFePO4, the memory-writing process is carried out by the extra partial cycle. Although the discharging cutoff does not change in the extra partial cycle,9 the relaxation time has been prolonged for the remaining partial LiFePO4 electrode, which might be related to the memory effect. Thereafter, a relaxation of 20 h is inserted into the memory-
3. RESULTS AND DISCUSSION Figure 2a shows a spherical microscale particle of LiFePO4 precursor after the spray-drying process. The spherical particles are hollow with a thin shell, and the particle size has a broad distribution, as shown in Figure S1. After calcination with 15 wt % sucrose additive at 700 °C for 5 h in an Ar + 5% H2 atmosphere, the hollow spherical particle becomes a hollowedout particle in the obtained LiFePO4 product, as shown in Figure 2b, and the sucrose additive was converted into ca. 1.5 wt % C
DOI: 10.1021/acsami.7b05852 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces writing process, as shown in Figure S2a,c. As we expected, the relaxation makes the memory effect stronger, as shown in Figure S2b,d. To study the dependence of LiFePO4 electrode on the relaxation time, we examined the charging behavior after a battery was discharged to 2.8 V and had a rest period of 0, 5, and 20 h, respectively, as shown in Figure 4. Evidently, the charge
Figure 4. Electrochemical dependence on the rest time between the discharge and charge processes of LiFePO4. (a) A sequence of three cycles: a discharge to 2.8 V and an immediate charge to 4.0 V (black line); a discharge to 2.8 V and rest for 5 h then a charge to 4.0 V (red line); a discharge to 2.8 V and rest for 20 h then a charge to 4.0 V (green line). (b) Enlarged view of (a) between 3.445 and 3.460 V. Here, the current rate is C/5 for discharging and C/10 for charging.
Figure 5. EIS spectra of LiFePO4 after different rest times. (a) An electrochemical sequence: discharge to 2.8 V at a rate of C/5 without relaxation and EIS measurement at 3.3 V with AC amplitude of 10 mV (EIS 1#, black); rest for 5 h and EIS measurement at 3.3 V (EIS 2#, red); and rest for 15 h and EIS measurement at 3.3 V (EIS 3#, green), where (dis)charging at rate of C/10 to 3.3 V and potentiostatic for 10 min before each EIS measurement. (b) Corresponding EIS spectra from 106 to 10−2 Hz. The enlarged medium and low-frequency region are shown in the insets.
plateau rises by prolonging the relaxation time. Then, EIS spectra of discharged LiFePO4 were measured after different rest times, as shown in Figure 5. In the high-frequency region, the depressed semicircle is associated with the charge-transfer resistance (Rct), and the tilted line in the low-frequency region is assigned to the Warburg impedance (Zw) of Li-ion diffusion. Clearly, both Rct and Zw become larger by increasing the rest time, and the |Z| at 10−2 Hz has been increased with 56 Ω after 20 of h rest, so the overpotential increment during the subsequence charging process could be estimated as ca. 2.8 mV, according to the Ohm’s law, where the charge−discharge current is ca. 50 μA. Our estimation is close to the measured increment of ca. 3.3 mV, as shown in Figure 4, so the relaxation-induced EIS increment results in the high overpotential of subsequent charging. Accordingly, a schematic diagram of phase transition is proposed to interpret the relaxation-dependent electrochemical kinetics, as shown in Figure 6. During discharging, the LiFePO4 electrode transforms from α-phase to β-phase, then the β-phase gradually changes to β′- and β″-phase in a long-term rest, representing the discharged phase of reduced electrochemical kinetics. Considering that the increment of Zw is much larger than that of Rct, the kinetic variety of the LiFePO4 electrode should be mainly attributed to Li-ion diffusion. Besides, LiFePO4 structure would gradually become stable during the long-term relaxation, which will need more energy to transform into an activated state; that is, the electrochemical kinetics will become lower. Meanwhile, we also examined the discharging behavior after a battery was charged to 4 V and had a rest of 0, 5, and 20 h,
Figure 6. Schematic diagram of phase transition in LiFePO4 with different rest times, in which α represents the delithiated phase and the lithiated phases with different rest times are represented by β, β′, and β″, respectively.
respectively, as shown in Figure S3. Similarly, the discharge plateau drops by prolonging the relaxation time. Figure S4 shows the EIS spectra of charged LiFePO4 after different rest times, in which both Rct and Zw increases slightly, in contrast to the discharged result. The EIS result is consistent with the discharging behavior, indicating a low relaxation dependence of electrochemical kinetics. As illustrated in Figure 6, the α-phase gradually changes to α′- and α″-phase in a long-term rest, representing that the electrochemical kinetics is slightly decreased of the charged phase. Therefore, the memory effect D
DOI: 10.1021/acsami.7b05852 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces
Figure 7. Memory effect of different LiFePO4 samples, which are calcined at 600 and 800 °C for 5 h, respectively. (a) 600 °C and charge, (b) 600 °C and discharge, (c) 800 °C and charge, and (d) 800 °C and discharge. The measurement procedure is the same as that in Figure 3.
Figure 8. Schematic diagram of phase transitions for different LiFePO4 samples during normal charging and memory effect. (a) Normal charging (red line) without initial overshooting and (b) memory effect (black line) as an elevated step for the LiFePO4 calcined at 600 °C. (c) Normal charging with initial overshooting and (d) memory effect (black line) with an evident bump for the LiFePO4 calcined at 800 °C.
Normally, the electrochemical kinetics is dependent on the operation temperature, so we compared the memory effect at different temperatures of 15, 25, and 35 °C. As shown in Figure S8, the charging plateau drops down from 3.47 to 3.45 V, and the step of the memory effect becomes low as the temperature is increased from 15 to 35 °C; that is, the overpotential of LiFePO4 is reduced, and the memory effect is weakened at the high temperature, which can be attributed to the fact that both Rct and Zw are normally reduced by elevating the temperature. Actually, the kinetic issues have been discussed in a few early literatures. In anatase TiO2 with different levels of oxygen vacancies, it has been demonstrated that the slower kinetics related to hindered Li-ion mobility is causing an increase of the memory effect.24 In Al-doped Li4Ti5O12, the memory effect is due to the poor electrochemical kinetics when discharged to a
of LiFePO4, as well as its asymmetry of its charging and discharging, can be attributed to the reduced electrochemical kinetics after long-term relaxation. Besides the relaxation time, the current density is also studied during the galvanostatic measurements. As shown in Figure S6, the LiFePO4 electrode is discharged to 2.8 V at a rate of 1, 0.2, and 0.05 C, respectively, and then the subsequent charging overpotential heightens evidently. Similarly, the discharging overpotential also increases when the precharging current is reduced, as shown in Figure S7. For the small rate, the LiFePO4 electrode takes a long time to transform between two phases, equivalent to having a long-term relaxation after (dis)charging, which verifies the reduced electrochemical kinetics after longterm relaxation. E
DOI: 10.1021/acsami.7b05852 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces lower cutoff potential.9 In Li-excess olivine LiFePO4, the presence of LiFe and the absence of FeLi unlock the restrictive Li-ion diffusion and induce faster kinetics, resulting in a significantly reduced memory effect.25 Thus, the electrode kinetics is a common origin of memory effect among those two-phase electrode materials. Next, the calcination temperature of LiFePO4 is tailored to study its impact on the memory effect. The same LiFePO4 precursor with 15 wt % sucrose additive was calcined at different temperatures for 5 h in an Ar + 5% H2 atmosphere, and the primary crystallites of 800 °C have a largest grain size, compared with those of 600 and 700 °C, as shown in Figure S9a,b. According to XRD patterns in Figure S9c, both LiFePO4 products are well crystallized as olivine LiFePO4, and Raman spectra are also confirmed their components, as shown in Figure S9d. The memory effect of these two samples are illustrated in Figure 7. Because the memory effect is weak for discharging, we carefully compared the memory effect for charging between two samples of 600 and 800 °C. With the higher calcining temperature of 800 °C, the initial overshooting is quite noticeable in the beginning of normal charging curve, which is hardly observed for 600 °C. During memory releasing, the charging curve appears as a two-stepped plateau for 600 °C, while a voltage bump is overlaid at the step edge for 800 °C. This result indicates that initial voltage overshooting can influence the memory effect of LiFePO4. In some early literature, the relationship between memory effect and initial overshooting have been discussed. In the pioneer work of memory effect in LIBs, the initial overshooting is attributed to some resistance involved with the spinodal decomposition or nucleation, and the memory effect is considered as the delayed overshooting when the phase transition takes place in sequential groups of LiFePO4 particles.7 In anatase TiO2, the overshooting feature follows the same dependence on the specific surface area as that for the memory effect, indicating that the overshooting is significantly related to the memory effect.8 However, the voltage bump and step of memory effect are still not clearly interpreted by the electrode kinetics and initial overshooting. Combining the memory effect and initial voltage overshooting, a new phase-transition model is proposed to describe the phase transition during normal charging and memory releasing. For LiFePO4 with low temperatures, the β-phase of small crystallites is successively transformed into the α-phase during charging, resulting in a smooth plateau in the two-phase region, as shown in Figure 8a. During memory releasing, the β-phase without relaxation first takes the phase transition at a low potential, and then the rest β′-phase with relaxation completes the phase transition at a high potential, so a voltage step is formed in the middle of the smooth plateau, as shown in Figure 8b. For LiFePO4 of high temperature, a voltage overshooting starts the successive phase transition of big crystallites, which should be attributed to the nucleation process, as shown in Figure 8c. After high-temperature calcination, the well-crystallized LiFePO4 has fewer specific surface and atomic defects, which play as nucleation sites, so the voltage overshooting is necessary to facilitate phase transition in each crystallite. During memory releasing, the initial voltage overshooting seems to shift to the middle of the charging plateau. Actually, the just-discharged βphase first takes the phase transition without an initial voltage overshooting, possibly owing to the residual α-nuclei and the unrelaxed crystal structure. In the middle of charging plateau, the rest β′-phase starts the phase transition from a nucleation
process, so a voltage bump emerges followed by a heightened step, as shown in Figure 8d. Therefore, the voltage bump of memory effect is actually a delayed voltage overshooting that is overlaid on the stepped plateau of memory effect.
4. CONCLUSIONS In this investigation, LiFePO4 is synthesized by a home-built spray drying instrument, and its memory effect in LIBS is studied by modifying the relaxation time and the initial voltage overshooting. As a result, the memory effect is significantly dependent on the relaxation time after phase transition, which can be attributed to the reduced electrochemical kinetics after long-term relaxation, and the voltage bump is actually a delayed voltage overshooting overlaid on the stepped plateau of memory effect. Because the underlying mechanism has been revealed in the memory effect, it would guide us to optimize two-phase electrode materials and improve Li-ion battery management systems.
■
ASSOCIATED CONTENT
* Supporting Information S
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b05852. Figures showing sample characterization, memory effects for different relaxation times and operation temperatures, electrochemical dependence on rest time and current rate, EIS spectra, and a phase-transition diagram. (PDF)
■
AUTHOR INFORMATION
Corresponding Authors
*E-mail:
[email protected]. *E-mail:
[email protected]. Tel: +86-898-66277929. ORCID
De Li: 0000-0003-2836-6808 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was financially supported by the NSFC (grant nos. 51162006, 51362009, and 21603048), Science and technology development special fund project (grant no. ZY2016HN07), Key Science & Technology Project (grant no. ZDXM2015118), International Science & Technology Cooperation Program of Hainan (grant no. KJHZ2015-02), Natural Science Foundation of Hainan (grant no. 20165186), and the Hainan University’s Scientific Research Foundation (grant no. kyqd1545).
■
REFERENCES
(1) Armand, M.; Tarascon, J. M. Building Better Batteries. Nature 2008, 451 (7179), 652−657. (2) Tarascon, J. M.; Armand, M. Issues and Challenges Facing Rechargeable Lithium Batteries. Nature 2001, 414 (6861), 359−367. (3) Nishi, Y. Lithium Ion Secondary Batteries; Past 10 Years and the Future. J. Power Sources 2001, 100 (1−2), 101−106. (4) Davolio, G.; Soragni, E. The ’Memory Effect’ on Nickel Oxide Electrodes: Electrochemical and Mechanical Aspects. J. Appl. Electrochem. 1998, 28 (12), 1313−1319. (5) Sato, Y.; Takeuchi, S.; Kobayakawa, K. Cause of the Memory Effect Observed in Alkaline Secondary Batteries Using Nickel Electrode. J. Power Sources 2001, 93 (1−2), 20−24. (6) Huggins, R. A. Mechanism of the Memory Effect in ″Nickel″ Electrodes. Solid State Ionics 2006, 177 (26−32), 2643−2646. F
DOI: 10.1021/acsami.7b05852 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces (7) Sasaki, T.; Ukyo, Y.; Novak, P. Memory Effect in a Lithium-Ion Battery. Nat. Mater. 2013, 12 (6), 569−575. (8) Madej, E.; La Mantia, F.; Schuhmann, W.; Ventosa, E. Impact of the Specific Surface Area on the Memory Effect in Li-Ion Batteries: The Case of Anatase TiO2. Adv. Energy Mater. 2014, 4 (17), 1400829. (9) Li, D.; Sun, Y.; Liu, X. Z.; Peng, R. W.; Zhou, H. S. Doping-Induced Memory Effect in Li-Ion Batteries: The Case of Al-Doped Li4Ti5O12. Chem. Sci. 2015, 6 (7), 4066−4070. (10) Padhi, A. K.; Nanjundaswamy, K. S.; Goodenough, J. B. PhosphoOlivines as Positive-Electrode Materials for Rechargeable Lithium Batteries. J. Electrochem. Soc. 1997, 144 (4), 1188−1194. (11) Kang, B.; Ceder, G. Battery Materials for Ultrafast Charging and Discharging. Nature 2009, 458 (7235), 190−193. (12) Wang, Y. G.; He, P.; Zhou, H. S. Olivine LiFePO4: Development and Future. Energy Environ. Sci. 2011, 4 (3), 805−817. (13) Yuan, L. X.; Wang, Z. H.; Zhang, W. X.; Hu, X. L.; Chen, J. T.; Huang, Y. H.; Goodenough, J. B. Development and Challenges of LiFePO4 Cathode Material for Lithium-Ion Batteries. Energy Environ. Sci. 2011, 4 (2), 269−284. (14) Nishimura, S.; Kobayashi, G.; Ohoyama, K.; Kanno, R.; Yashima, M.; Yamada, A. Experimental Visualization of Lithium Diffusion in Li(x)FePO(4). Nat. Mater. 2008, 7 (9), 707−711. (15) Delmas, C.; Maccario, M.; Croguennec, L.; Le Cras, F.; Weill, F. Lithium Deintercalation in LiFePO(4) Nanoparticles Via a DominoCascade Model. Nat. Mater. 2008, 7 (8), 665−671. (16) Yamada, A.; Koizumi, H.; Nishimura, S. I.; Sonoyama, N.; Kanno, R.; Yonemura, M.; Nakamura, T.; Kobayashi, Y. Room-Temperature Miscibility Gap in Li(x)FePO(4). Nat. Mater. 2006, 5 (5), 357−360. (17) Malik, R.; Abdellahi, A.; Ceder, G. A Critical Review of the Li Insertion Mechanisms in Lifepo4 Electrodes. J. Electrochem. Soc. 2013, 160 (5), A3179−A3197. (18) Dreyer, W.; Jamnik, J.; Guhlke, C.; Huth, R.; Moskon, J.; Gaberscek, M. The Thermodynamic Origin of Hysteresis in Insertion Batteries. Nat. Mater. 2010, 9 (5), 448−453. (19) Li, D.; Liu, X. Z.; Zhou, H. S. The Size-Dependent Phase Transition of LiFePO4 Particles During Charging and Discharging in Lithium-Ion Batteries. Energy Technol. 2014, 2 (6), 542−547. (20) Ni, L.; Zheng, J. G.; Qin, C. C.; Lu, Y. W.; Liu, P. X.; Wu, T. F.; Tang, Y. F.; Chen, Y. F. Fabrication and Characteristics of Spherical Hierarchical LiFePO4/C Cathode Material by a Facile Method. Electrochim. Acta 2014, 147, 330−336. (21) Hjelm, A. K.; Lindbergh, G. Experimental and Theoretical Analysis of LiMn2O4 Cathodes for Use in Rechargeable Lithium Batteries by Electrochemical Impedance Spectroscopy (EIS). Electrochim. Acta 2002, 47 (11), 1747−1759. (22) Zhao, Y.; Peng, L. L.; Liu, B. R.; Yu, G. H. Single-Crystalline LiFePO4 Nanosheets for High-Rate Li-Ion Batteries. Nano Lett. 2014, 14 (5), 2849−2853. (23) Burba, C. M.; Frech, R. Raman and Ftir Spectroscopic Study of LixFePO4 (0 < X < 1). J. Electrochem. Soc. 2004, 151 (7), A1032−A1038. (24) Ventosa, E.; Loffler, T.; La Mantia, F.; Schuhmann, W. Understanding Memory Effects in Li-Ion Batteries: Evidence of a Kinetic Origin in TiO2 Upon Hydrogen Annealing. Chem. Commun. 2016, 52 (77), 11524−11526. (25) Park, K.-Y.; Park, I.; Kim, H.; Yoon, G.; Gwon, H.; Cho, Y.; Yun, Y. S.; Kim, J.-J.; Lee, S.; Ahn, D.; Kim, Y.; Kim, H.; Hwang, I.; Yoon, W.S.; Kang, K. Lithium-Excess Olivine Electrode for Lithium Rechargeable Batteries. Energy Environ. Sci. 2016, 9 (9), 2902−2915.
G
DOI: 10.1021/acsami.7b05852 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
