- Email: [email protected]
C cathode composite at low and high temperatures
Applied Energy 162 (2016) 1419–1427
Contents lists available at ScienceDirect
Applied Energy journal homepage: www.elsevier.com/locate/apenergy
Study of electrochemical performances of lithium titanium oxide–coated LiFePO4/C cathode composite at low and high temperatures q Chun-Chen Yang a,c,⇑, Jer-Huan Jang a,b, Jia-Rong Jiang a,c a
Department of Chemical Engineering, Ming Chi University of Technology, New Taipei City 243, Taiwan, ROC Department of Mechanical Engineering, Ming Chi University of Technology, New Taipei City 243, Taiwan, ROC c Battery Research Center of Green Energy, Ming Chi University of Technology, New Taipei City 243, Taiwan, ROC b
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
Spherical LiFePO4/C (LFP/C)
The surface modification of the LiFePO4 cathode material is done by a dual-layer coating (LTO and C layers) via a spray dry and a sol–gel method.
composites are prepared by a spray dry method. The 1–5 wt.%Li4Ti5O12 (LTO) coating is carried out on SP-LFP/C. The dual surface coating with LTO and carbon layers greatly improve the cycle life performance at 55 °C at 3 C rate. The LTO and carbon layers on LFP are functions as the electron and ionic conductors, respectively.
a r t i c l e
i n f o
Article history: Received 1 September 2014 Received in revised form 27 January 2015 Accepted 27 January 2015 Available online 5 March 2015 Keywords: Spray dry (SP) LiFePO4 Li4Ti5O12 (LTO) Surface modification Lithium-ion batteries
a b s t r a c t Spherical porous LiFePO4/C (LFP/C) composite materials were prepared by employing a solid-state (SS) method and a spray dry (SP) method. The surface modification was conducted on the spherical LFP/C composite using 1–5 wt.%Li4Ti5O12 (LTO) to improve the rate capability and the cycle stability properties at low temperature (0 °C) and elevated temperature (55 °C). The characteristic properties were examined through X-ray diffraction, micro-Raman spectroscopy, scanning electron microscopy, an AC impedance method, and galvanostatic charge–discharge method. The characteristic properties of the SS-LFP/C, SP-LFP/C, and LTO-coated SP-LPF/C composites were studied and compared. The 3 wt.%LTO-coated SP-LFP/C composite exhibited the most favorable performance among the samples. It exhibited discharge capacities of 150, 141, 131, 110, 103, and 84 mA h g1, at rates of 0.2 C, 0.5 C, 1 C, 3 C, 5 C, and 10 C, respectively. It was also found that the 3 wt.%LTO-coated SP-LFP/C composite shows the best cycling stability performance at
q This article is based on a short proceedings paper in Energy Procedia Volume 161 (2014). It has been substantially modified and extended, and has been subject to the normal peer review and revision process of the journal. This paper is included in the Special Issue of ICAE2014 edited by Prof. J Yan, Prof. DJ Lee, Prof. SK Chou, and Prof. U Desideri. ⇑ Corresponding author at: Department of Chemical Engineering, Ming Chi University of Technology, New Taipei City 243, Taiwan, ROC. Tel.: +886 2 2908 9899; fax: +886 2 2908 5941. E-mail address: [email protected] (C.-C. Yang).
http://dx.doi.org/10.1016/j.apenergy.2015.01.131 0306-2619/Ó 2015 Elsevier Ltd. All rights reserved.
1420
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
elevated temperature of 55 °C. The LTO-coated SP-LFP/C composite was demonstrated to be suitable for high-temperature and high-power application in lithium-ion batteries. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Because of their relatively high specific capacity, lithium-ion batteries are used in electric vehicles, cell phones, laptop computers, digital cameras, renewable-energy storage, and smart grid applications. The LiFePO4 cathode material proposed by Padhi et al. [1] has attracted extensive attention for application in the next generation of rechargeable lithium-ion batteries because of its low cost, environmental friendliness, excellent safety characteristics, high capacity (approximately 170 mA h g1 in theory), and excellent cycling performance. However, its low conductivity in electron and ion transfer constitutes a major barrier for use in commercial applications. The electron conductivity of LiFePO4 is only 109 S cm1 [2] and its lithium-ion diffusivity is 1014 to 1016 cm2 s1 [3]. Three methods can be used to improve the electronic conductivity: (1) carbon coating; (2) doping with supervalent cations; (3) decreasing the particle size. The carbon coating is the most effective among these methods, and is a facile method of improving the conductivity of LiFePO4 materials. LiFePO4 cathode materials can typically be prepared through a solid-state reaction [4,5], sol–gel process [6,7], or hydrothermal process [8–14]. Whittingham et al. [8] revealed that high-quality crystalline LiFePO4 platelets can be obtained using a low-temperature hydrothermal process. A spray dry (SP) method has been widely used to prepare spherical LiFePO4/C cathode materials to enhance the electrochemical performance [15–19]. The low- and high-temperature performance and capacity fading mechanism of LiFePO4/C materials have recently been intensively explored [20–24]. Surface coating is an effective method of solving the problems of high-temperature metal dissolution and cycling deterioration. The spinel Li4Ti5O12 (LTO) surface modification of various cathode materials has become crucial for improving low- and high-temperature electrochemical properties [25–29]. Cong et al. [25] prepared a 3 wt.%Li4Ti5O12 (3%LTO)-coated Li[Li0.2Mn0.54Ni0.13Co0.13]O2 lithium-rich cathode material by using a ball-milled process. The lithium-rich oxides exhibited cycle-performance and coulombic efficiency improvement because LTO provides a numerous sites for insertion of extracted lithium. Yi et al. [26] reported the electrochemical performance of LTO-coated LiMn1.4Ni0.4Cr0.2O4 material as a 5V cathode. They showed that 4 wt.%LTO-coated LiMn1.4Ni0.4Cr0.2O4 particles exhibited the highest electrochemical performance. Yao et al. [27] synthesized Li3.5+xTi5O12-coated and LTO-coated LiMn2O4 cathode materials by using the sol–gel method. Their study demonstrated that the valence of surface Mn for the Li3.5+xTi5O12-coated LiMn2O4 material increased after annealing; the material exhibited a favorable cycling performance at 55 °C. Zhu et al. [28] also prepared an LTO-coated LiMn1.5Ni0.5O4 spinel material by using the sol–gel method followed by high-temperature calcination. They demonstrated that a 3wt.%LTO-coated layer protected the surface of the LiMn1.5Ni0.5O4 active material from an HF attack in the electrolyte. Yi et al. [29] prepared LTO-coated LiCoO2 powders by using a sol– gel method at various sintering temperatures and coating contents. The results demonstrated that the LTO-coated layer on the LiCoO2 suppressed the capacity fading by reducing cell polarization. No studies have been published that investigate LTO-coated LiFePO4/C. In this study, we first prepared solid-state LiFePO4/C (denoted as SS-LFP/C) materials by using a solid-state method. The SP method and post-sintering were used to prepare a spherical porous
SP-LFP/C composite. A suitable amount of glucose was added as a binder and carbon source in the SP processing. To improve the low- and high-temperature performance of the SP-LFP/C material, LTO surface modification was applied to the SP-LFP/C material by using the sol–gel method. For comparison, an SP-LiFePO4/C material without LTO surface modification was also examined. The characteristic properties of the SP–LiFePO4/C cathode material with and without LTO coating were examined through X-ray diffraction (XRD), micro-Raman spectroscopy, scanning electron microscopy (SEM), elemental analysis, and micro-Raman spectroscopy. The electrochemical performances of the Li/LFP coin cell were examined using an automatic galvanostatic charge–discharge unit and a cyclic voltammetry (CV) method. 2. Experimental 2.1. Preparation of SP–LiFePO4/C and LTO-coated SP–LiFePO4/C composites The LiFePO4/C materials (SS-LFP/C) were prepared using a solid state method. The appropriate quantities of the raw materials LiH2PO4, FeC2O42H2O, 5% sucrose, and 5% citric acid (Aldrich) were mixed in acetone and then ball-milled at a rotation speed of 400 rpm for 6 h in a planetary miller with a ball/reactant weight ratio of 6:1. The added carbon precursors in the LiFePO4 materials was maintained at 10 wt.%. The ball-milled mixture precursor was pre-sintered at 350 °C for 4 h and then sintered at 700 °C for 12 h in an Ar/H2 (95:5, v/v) atmosphere. The spherical SP–LiFePO4/C material were prepared using an SP method and a post-sintering process. The appropriate quantities of the as-prepared SS-LiFePO4/C materials and 5% glucose were mixed and dispersed in deionized water. The spherical porous SP-LFP/C material with a double carbon source was prepared using a spray dryer (EYELA, Spray Dryer SD-1000 model, Japan). The spray-dried spherical LFP/C composite precursor was further calcined at 700 °C for 5 h in an Ar atmosphere. The 1–5 wt.%LTO-coated SP-LFP/C composites were synthesized using the sol–gel method. Ti(C4H9O)4 and CH3COOLi2H2O at a Li:Ti stoichiometric molar ratio of 4:5 were dissolved in a solution containing ethanol and distilled water to form a clear solution. A small amount of HNO3 was then added to this solution under continuous stirring to obtain a sol. The spherical SP-LFP/C powder was slowly added to the sol under constant stirring. After hydrolyzing and condensation for 10 h, a black gel formed and was dried at 120 °C for 2 h and then calcined at 700 °C for 3 h in an Ar atmosphere to obtain the final 3%LTO-coated SP-LFP/C composite. 2.2. Characterization The crystal structure of the SS-LFP/C, SP-LFP/C, and LTO-coated SP-LFP/C composite samples was examined using an XRD spectrometer (Philip, X’pert Pro System). The surface morphology was examined using a scanning electron microscope (SEM, Hitachi). The micro-Raman spectra were recorded on a confocal micro-Renishaw with a 632 nm He–Ne laser excitation. The residual carbon content in the sample was analyzed using an elemental analyzer (Perkin Elmer 2400). The electron conductivity of the composite samples was measured using an AC impedance method.
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
2.3. Electrochemical measurements The electrochemical performances of the Li/LiFePO4 half-cell batteries were measured using a two-electrode system (a CR 2032 coin cell assembled in an argon-filled glove box). All LiFePO4/ C electrodes were prepared by mixing active LiFePO4/C materials, Super P, and a poly(vinyl fluoride) binder at a weight ratio of 90:5:5, pasted on an aluminum foil (Aldrich), and finally dried in a vacuum oven at 120 °C for 12 h. The lithium foil (Aldrich) was used as the counter and reference electrode. A microporous PE film was used as a separator. The electrolyte was 1 M LiPF6 in a mixture of EC and DEC (1:1 in v/v, Merck). The half-cell LiFePO4/Li batteries were charged using a constant current profile and discharged using a constant current profile, over a potential range of 2.0–3.8 V (vs. Li/Li+) at varied C rates with an Autolab PGSTAT302N potentiostat. The CV was conducted using an Autolab instrument at a scanning rate of 0.1 mV s1, between 2.5 and 4.2 V. 3. Results and discussion The XRD patterns of the SS-LFP/C, SP-LFP/C, and 3 wt.%LTOcoated SP-LFP/C composite prepared using the solid state method and SP method are shown in Fig. 1. The XRD diffraction patterns revealed that the as-prepared SS-LFP/C, SP-LFP/C, and 3 wt.%LTOcoated SP-LFP/C composite were single-phase materials with an olivine-type structure indexed in the orthorhombic Pnma space group. The diffraction peak of the residual carbon could not be determined in the pattern and may be their low content or amorphous state. The lattice parameters (a, b, c, and cell volume) of the SS-LFP/C, SP-LFP/C, and 3 wt.%LTO-coated SP-LFP/C composites were calculated based on the XRD patterns. The lattice parameters of all LFP/C samples were the same as those of the standard LiFePO4 (JCPDS card number 81-1173, a = 10.33 Å, b = 6.010 Å, c = 4.692 Å, V = 291.35 (Å)3). The LTO and carbon source of glucose and citric acid had no observable influence on the structure of the LFP/C composite materials. No impurity phase occurred. The SEM image of the SS-LFP/C material prepared using the solid state method is shown in Fig. 2(a). Generally, the particle size increased with increasing sintering temperature. The particle size of the SS-LFP/C sample was in the range of 0.5–2 lm. Consequently, the addition of citric acid and sucrose as the carbon source not only decreased the particle size, but also prevented the particle from growing during the sintering process. The SEM image of the spherical SP-LFP/C material prepared using the SP method and sintering at 700 °C, is shown in
(a) SS-LFP/C (b) SP-LFP/C (c) 3%LTO-coated SP-LFP/C
Intensity/ a.u.
(a)
(b)
(c)
10
20
30
40
50
60
70
2θ/ degree Fig. 1. XRD patterns of SS-LFP/C, SP-LFP/C, 3 wt.%LTO-coated SP-LFP/C samples.
1421
Fig. 2(b). The particle size of the SP-LFP/C composite sample was in the range of 2–10 lm. Numerous micro-pores formed on the surface of the SP-LFP/C material and acted as a micro-reservoir for the electrolyte. The SP-LFP/C composite with highly porous and spherical morphology can markedly enhance the mass-transport properties. Fig. 2(c) depicts the SEM image of a 3 wt.%LTOcoated SP-LFP/C composite prepared using the sol–gel method. The particle size of the 3 wt.%LTO-coated SP-LFP/C composite was also in the range of 2–10 lm. The LTO surface modification did not exert any influence on the morphology of the SP-LFP/C composite. It was also found that there are many nano-pores (100– 200 nm) formed on the top-surface of the 3%LTO-coated SP-LFP/C material and it can be function as the electrolyte reservoir. The 3%LTO-coated SP-LFP/C composite with a thin LTO oxide barrier layer substantially improved the electrochemical properties at elevated temperature (see later cycling performance). In addition, micro-Raman spectra of the SS-LFP/C, SP-LFP/C, and 3 wt.% LTO-coated SP-LFP/C composites were prepared using the solid state, SP, and sol–gel methods, respectively, and are depicted in Fig. 3. The micro-Raman peaks of the SS-LFP/C material, located at approximately 1328 and 1590 cm1, were attributable to citric acid and sucrose carbon. Raman peaks at approximately 1328 cm1 (D band) and 1590 cm1 (G band) were observed in the composite samples. The micro-Raman peaks of the SP-LFP/C composites, located at approximately 1332 and 1591 cm1, were attributable to residual carbon of glucose and citric acid. Micro-Raman peaks at approximately 1330 cm1 (D band) and 1586 cm1 (G band) were also observed in the 3%LTO-coated SP-LFP/C composites. The broadening of the D (A1g symmetry) band and G (E2g symmetry) band with a strong D band indicated a localized inplane sp2 graphitic crystal domain and disordered sp3 amorphous carbon. The intensity ratio of the D band versus the G band (denoted as R = ID/IG) was employed to estimate the carbon quality of the LiFePO4/C samples. The R value of the 3 wt.% LTO-coated SP-LFP/C and bare SP-LFP/C composites was approximately between 0.97 and 1.07. By comparison, the R value of the SS-LFP/C samples without spray-dry and LTO oxide coating was approximately 1.10. It is well accepted [11] that the discharge capacity and the rate capability of the LiFePO4/C samples are closely related to the intensity ratio of the D band and the G band. Various micro-Raman peaks located approximately in the ranges of 947–950 cm1 and 596– 446 cm1 were identified as resulting from the vibration of the P–O bond, and the peak located at 638 cm1 was identified as resulting from the vibration of the FeOx groups. It was revealed that the R value of the 3%LTO-coated SP-LFP/C shows the lowest value. It may be due to the surface modification and thermal treatment synergistic effect. The thin LTO layer may re-modify parts of carbon defective area; it results in much uniform and dense carbon layer. Another possible reason is that the thermal sintering treatment during the LTO coating process cause to remove some defective-type carbons residuals; therefore, the R value was slightly reduced. The electrochemical properties of LiFePO4/C samples were strongly related to the electronic conductivity, the total residual carbon content, and the value of the ID/IG ratio. Both the electronic conductivity and the total residual carbon content substantially affected the discharge capacity and the rate capability of the LFP/ C composites. The electronic conductivities (re) and the residual carbon (C%) content of all LiFePO4/C samples were approximately 1.10–3.67 104 S cm1 and 3.10–4.2%, respectively. These values are consistent with findings reported in the literatures [1,2]. By contrast, the electronic conductivity of the 3 wt.% LTO-coated LFP/C composites was slightly higher than those of the SS-LFP/C and SP-LFP/C samples. As shown in Fig. 4, this is the CV curves for the SS-LFP/C (by a solid-state ball-milled method), SP-LFP/C (by a spray dry method),
1422
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
Fig. 2. SEM images of (a) SS-LFP/C; (b) SP-LFP/C, and (c) 3 wt.%LTO-coated SP-LFP/C composite.
(a) SS-LFP/C with ID /I G = 1.10 (b) SP-LFP/C with I D /I G = 1.07
D
G
(a) SS-LFP/C (w/o spray dry) (b) 0%LTO-SP-LFP/C (c) 3% LTO-SP-LFP/C (d) 5% LTO-SP-LFP/C
3.0
(c) 3%LTO SP-LFP/C with I D /I G = 0.97
2.5
(b) (c)
2.0
Intensity/ a.u.
(c)
Current/ mA
1.5
(b)
(d)
1.0 0.5
(a)
0.0 -0.5
(a)
-1.0
200
400
600
800
1000 1200 1400 1600 1800 2000
Raman shift/ cm—1 Fig. 3. Micro-Raman spectra of SS-LFP/C, SP-LFP/C, and 3 wt.%LTO-coated SP-LFP/C composites.
3 and 5 wt.% LTO-coated SP-LFP/C (by a spray dry + sol–gel surface modification) composite materials. The oxidation peak (charge) intensity of the SP-LFP/C material is higher than that of the 3 wt.%LTO-coated SP-LFP/C composite. However, the reduction peak (discharge) intensity of the SP-LFP/C material is much lower than that of the 3%LTO-coated SP-LFP/C composite. By comparison, the total integrated area (charge) from the CV curve of the 3 wt.%LTO-coated SP-LFP/C composite is slightly larger than that of the SP-LFP/C material. This indicated that the current efficiency of the 3%LTO-coated SP-LFP/C composite may be slightly larger than that of the SP-LFP/C material. Indeed, the SP-LFP/C material shows a slightly fast kinetic performance. It is because that the extra LTO-coated layer on the surface SP-LFP/C material induces
-1.5 -2.0 2.6
2.8
3.0
3.2
3.4
3.6
3.8
4.0
Potential/ V (vs. Li+/Li) Fig. 4. The CV curves of SS-LFP/C, SP-LFP/C, 3 and 5 wt.%LTO-coated SP-LFP/C samples at 0.1 mV s1.
a barrier to the charge transfer reaction. It results in some polarization (i.e., charge-transfer resistance increase) on the 3 wt.%LTOcoated SP-LFP/C composite. However, we also discovered that the 5 wt.%LTO-coated SP-LFP/C composite electrode showed poor kinetics properties, as compared with the 3 wt.%LTO-coated SPLFP/C electrode. It may be due to too thick LTO-coated layer, exhibiting much lower charge and discharge peak currents. As seen clearly in Fig. 5, the poorest initial cycle charge/discharge performance at 0.1 C at 25 °C among the LFP/C materials is SS-LFP/C; the charge and discharge capacities are ca. 144 and 133 mA h g1, respectively. In contrast, the charge and discharge
1423
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
4000
(A) SS-LFP/C (in 2.0V~3.8V range) 3800 3600
3500
3400
3250
3200
Voltage/ mV
Voltage/ mV
3750
3000 2750
0.2C/0.2C 0.2C/0.5C 0.2C/1C 0.2C/3C 0.2C/5C 0.2C/10C
3000 2800 2600
(a) 2500
2400
(a) SS-LFP (w/o spray dry) (b) 0%LTO-SP-LFP (c) 3%LTO-SP-LFP/C (d) 5%LTO-SP-LFP/C
2250 2000
0
20
40
60
80
100
2200 2000
(d) (b) (c) 120
140
160
0
20
40
60
80
100
120
140
Capacity/ mAh g-1
180
Capacity/ mAh g-1
(B) SP-LFP/C in 2.0V~3.8V range 4000
Fig. 5. The initial charge/discharge curves of SS-LFP/C, SP-LFP/C, and 3 and 5 wt.%LTO-coated SP-LFP/C samples at 0.1 C rate.
Voltage/ mV
3600 3400
0.2C/0.2C 0.2C/0.5C 0.2C/1C 0.2C/3C 0.2C/5C 0.2C/10C
3200 3000 2800 2600 2400 2200 2000 0
20
40
60
80
100
120
140
Capacity/ mAh g-1 (C) 3%LTO-coated SP-LFP/C in 2.0V~3.8V range 4000 3800 3600 3400
Voltage/ mV
capacities of the SP-LFP/C material are ca. 157 and 152 mA h g1, respectively. On the contrary, we find that the charge and discharge capacities of the 3%LTO-coated SP-LFP/C composite material are 170 and 151 mA h g1, respectively. The extra LTO-coated layer on the surface of SP-LFP/C sample may produce the extra barrier layer to cause some polarization. That is why the discharge capacity is slightly reduced. Surprisingly, it was seen that the charge and discharge capacities of the 5 wt.%LTO-coated SP-LFP/C composite are around 152 and 134 mA h g1, respectively. The specific discharge capacity was greatly reduced; it may be due to the too thick oxide barrier layer (LTO) on SP-LFP/C material. However, the oxide barrier layer can fully isolate the cathode active material direct contact with electrolyte (or avoid HF attack). More importantly, the LTO coating layer can also function as an ionic conductor layer; it will help improve the lithium ionic conductivity. It is because that the LTO oxide with the spinel NASCION structure shows the high ionic conductivity. It can markedly improve the high temperature cycle-life performance at 55 °C, in particular, at 3 C discharge rate. The typical charge–discharge profiles of the SS-LFP/C and SPLFP/C, and 3 wt.%LTO-coated SP-LFP/C composite at 25 °C and varying rates of 0.2–10 C are displayed in Fig. 6. All LFP/C samples revealed typical flat potential plateaus at 3.3–3.4 versus Li/Li+ at rates of 0.2–0.5 C. The potential plateau sloped when the discharge rate was higher than 1 C. The bare SP-LFP/C material exhibited specific discharge capacities of 135, 131, 122, 103, 94, and 78 mA h g1 at rates of 0.2, 0.5, 1, 3, 5, and 10 C, respectively. In contrast, the 3%LTO-coated SP-LFP/C composite exhibited specific discharge capacities of 150, 141, 131, 110, 103, and 84 mA h g1 at rates of 0.2, 0.5, 1, 3, 5, and 10 C, respectively. However, the SS-LFP/C sample with irregular morphology and without LTO surface modification only exhibited the specific discharge capacities of 125, 116, 105, 87, 76, and 48 mA h g1 at rates of 0.1, 0.2, 0.5, 1, 3, 5 and 10 C, respectively. Fig. 7 provides a comparison of the rate capability of the SS-LFP/C, SP-LFP/C, and 3 wt.%- and 5 wt.%LTO-coated SP-LFP/C composite at various rates. Both 3D porous spherical morphology (achieved using the SP method) and 3 wt.%LTO surface modification (achieved using the sol–gel method) on the LiFePO4 material substantially improved the highrate capability. It is well accepted [19–24] that the high concentration Fe2+ dissolving from LFP/C material at elevated temperature is one of the
3800
3200 3000 2800 0.2C/0.2C 0.2C/0.5C 0.2C/1C 0.2C/3C 0.2C/5C 0.2C/10C
2600 2400 2200 2000 0
20
40
60
80
100 120 140 160 180
Capacity/ mAh g-1 Fig. 6. The charge–discharge curve of (a) SS-LFP/C, (b) SP-LFP/C, (c) 3 wt.%LTOcoated SP-LFP/C composites at varied rates.
main reasons for the occurrence of a capacity-fading mechanism. According to many research results [19–24], it was clearly indicated that the HF generated from the LiPF6 electrolyte was responsible for the dissolution of Fe during cycling. The uncoated LFP/C cathode material in the LiPF6 electrolyte reacted with the HF and thus
1424
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
160
(a) 0%LTO-SP-LFP/C at 25oC (b) 0%LTO-SP-LFP/C at 55oC (c) 0%LTO-SP-LFP/C at 0oC (d) 3%LTO-SP-LFP/C at 25oC (e) 3%LTO-SP-LFP/C at 55oC (f) 3%LTO-SP-LFP/C at 0oC (g) 5%LTO-SP-LFP/C at 25oC (h) 5%LTO-SP-LFP/C at 55oC o (i) 5%LTO-SP-LFP/C at 0 C
140
160 140
(b)
(e)
80
(b) (d)
60
(a)
40 20
LFP at 1C/3C o at 55, 25, and 0 C (b)
100
(a) SS-LFP/C (b) 0%LTO-SP-LFP/C (c) 3%LTO coted SP-LFP/C (d) 5%LTO coted SP-LFP/C
Capacity/ mAh g-1
Capacity/ mAh g-1
120
120 (h)
(d)
100
(a) (g)
80 60 (f)
40
0
0.2C 0.5C 1C 3C 5C 10C 1-3cycle 1-3cycle 1-3cycle 1-3cycle 1-3cycle 1-3cycle
(i)
(c)
20 20
Fig. 7. The rate capability of SS-LFP/C, SP-LFP/C, and 3 and 5 wt.%LTO-coated SP-LFP/C samples at 0.2–10 C rates.
40
60
80
100
120
140
160
180
200
Number of cycles Fig. 8. The cycling and CE performance for SS-LFP/C, SP-LFP/C, and 3 and 5 wt.%LTOcoated SP-LFP/C composites at 1 C/3 C rate at 25 and 55 °C.
led to a gradual capacity loss. LTO surface coating can prevent Fe2+ ions from directly contacting electrolyte and therefore greatly decrease the capacity loss [24,25]. Differently, LTO has a 3D spinel structure that also acts as an ionic conductor and shows high lithium-ion conductivity. Unlike pure oxides (e.g., Al2O3, ZrO2), they tend to block the ionic channel of LFP. By contrast, LTO can enhance the lithium ionic conductivity of LiFePO4/C cathode material. Moreover, it was found that the discharge capacity and the coulombic efficiency of the 0 wt.%, 3 wt.%, and 5 wt.%LTO coated SP-LFP/C composite at 3 C discharge rate were approximately 100-110 mA h g1 and 93–97%, respectively during 200 cycling tests at 25 °C, as shown in Fig. 8. No significant difference was observed. As a result, their fading rates of the 0 wt.%-, 3 wt.%and 5 wt.%LTO-coated SP-LFP/C composites were approximately 0.01–0.02 mA h cycle1 at 25 °C. More interestingly, the electrochemical performance of the SPLFP/C material with and without LTO coating at elevated temperature (55 °C) was evaluated and compared in details. We found that the specific discharge capacity of the SP-LFP/C material (without LTO coating) at 3 C and 55 °C is markedly decreased, i.e., from 148 to 67 mA h g1 with the fading rate of 0.405 mA h cycle1 for 200 cycles. In contrast, the discharge capacity of the 3 wt.%LTOcoated SP-LFP/C composite at 3 C and 55 °C for 200 cycles is smoothly decreased, i.e., from 123 to 117 mA h g1 with much lower fading rate of 0.028 mA h cycle1. As shown in Fig. 8, the specific discharge capacity of the 5 wt.%LTO-coated SP-LFP/C composite at 55 °C is modestly reduced, from 133 to 97 mA h g1 with the fading rate of 0.180 mA h cycle1 for 200 cycles. In fact, we was found that the 3 wt.%LTO coated SP-LFP/C composite exhibited the best performance at elevated temperature. By comparison the low temperature performance, we found that Furthermore, the discharge capacities and coulombic efficiencies of the 3 wt.%- and 5 wt.%LTO-coated SP-LFP/C composites at 0 °C were varied at 45–42 mA h g1 and 95–97%, respectively. The fading rates of the 3 wt.%- and 5 wt.%LTO-coated SP-LFP/C composites were about 0.015–0.020 mA h cycle1 at 0 °C. As a matter of fact, all SP-LFP/C samples with and without LTO surface modification exhibited pretty similar electrochemical performance, with the fading rate of 0.015–0.026 mA h cycle1 at 0 °C. It was seen clearly that the fading rate (0.405 mA h cycle1) of the bare SP-LFP/C cathode material at 55 °C showed much higher
than that of the bare SP-LFP/C at 0 °C (0.0260 mA h cycle1). It can be seen clearly that the capacity fading will be accelerated at elevated temperature. However, the fading rate of the bare SPLFP/C material (0.405 mA h cycle1) without LTO surface modification was considerably higher than that of the 3%LTO-coated SPLFP/C composite (0.028 mA h cycle1) at elevated temperature (55 °C). In conclusion, it was revealed that the LTO surface modification on LiFePO4/C material is effective way to reduce the fading rate at elevated temperature. We found that the optimum amount of LTO coating needed to produce stable cycling performance at elevated temperature (55 °C) was 3 wt.% of the SP-LFP/C material. The AC impedance spectroscopy was used to study the interface properties of the SS-LFP/C sample and the 3 wt.%LTO-coated SPLFP/C composite samples. The AC spectra of the 3 wt.%LTO-coated SP-LFP/C composite at temperatures of 25, 0, 10, and 20 °C, at an open circuit potential, are shown in Fig. 9(a). Each AC plot consisted of one semicircle at a higher frequency, followed by a linear portion at a lower frequency. The lower frequency region of the straight line was considered as the Warburg impedance. It is for long-range lithium-ion diffusion in bulk phase. The Rb indicates the bulk resistance at the electrolyte; Rct is the charge transfer resistance at the active material interface; CPE represents the double layer capacitance and certain surface film capacitances. The lithium chemical diffusion coefficients (Di) of the electrode were calculated based on Eq. (1) [19–24]:
Di ¼
2 1 RT 2 2 AF rw C
ð1Þ
where rw is the Warburg impedance coefficient (obtained a slope from plot of Zre vs. w0.5, as seen in Fig. 9(b)); Di is the lithium diffusion coefficient; R is the gas constant; T is the absolute temperature; F is the Faraday constant; A is the area of the electrode; and C is the molar concentration of Li+ ions (CLi+ = 1.0 103 mol cm3). Tables 1 and 2 show the calculated Rb, Rct, Di, and jo parameters for the SS-LFP/C sample and the 3 wt.%LTO-coated SPLFP/C composite. The Rb values of the SS-LFP/C sample at temperatures of 25, 0, 10, and 20 °C were 5.14, 7.25, 9.11, and 12.19 ohm, respectively. The lower temperature was in the test,
1425
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
(a)
(b)
1000
1000
o
(a) 20 C o (b) 0 C o (c) -10 C (d) -20oC
o
(a) 20 C (b) 0oC (c) -10oC (d) -20oC
900
900 800
800
700 600
600
-Z''/ ohm
-Z''/ ohm
700
500
500
(d)
400
400 (d)
(c)
300
300
(b) (c)
200
200
(b)
(a) (a)
100
100
0
0 0
200
400
600
800 1000 2001 1400 1600 1800 2000 2200 2400
0
200
400
Z'/ ohm
600
800
1000
Z'/ ohm 1500
2500
(d)
o
(a) 20 C o (b) 0 C (c) -10oC (d) -20oC
2250 2000
1250
(a) 20oC o (b) 0 C o (c) -10 C (d) -20oC
(d)
1750
1000
1250
Z'/ ohm
Zre /ohm
1500
(c)
(c)
750
1000 (b)
750
(b)
500
(a)
250
500 (a)
250 0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
0
W-1/2
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
W--1/2 Fig. 9. (a) The AC impedance spectra of SP-LFP/C composite and Z0 vs. w1/2 curve; (b) The AC impedance spectra of 3 wt.%LTO-coated SP-LFP/C composite and Z0 vs. w1/2 curve.
Table 1 Impedance parameters derived using equivalent circuit model and lithium diffusion coefficient (Di) for SS-LFP/C material. Temp (°C)
Param Rb (X)
25 0 10 20
5.14 7.25 9.11 12.19
Table 2 Impedance parameters derived using equivalent circuit model and lithium diffusion coefficient (Di) for 3 wt.%LTO-coated SP-LiFePO4/C composite. Temp (°C)
Rct (X) 123.5 615.0 1133.0 1862.0
Di (cm2 s1) 14
2.61 10 3.00 1015 9.71 1016 3.65 1016
the higher was the value of Rb. This was because of the lower ionic conductivity of the electrolyte. The values of Rct and Di for the SS-LFP/C samples varied considerably. The Rct values of the SSLFP/C sample at temperatures of 25, 0, 10, and 20 °C were 123.5, 615.0, 1133, and 1862 ohm, respectively. The Di values of
25 0 10 20
Param Rb (X)
Rct (X)
Di (cm2 s1)
5.67 8.19 8.85 11.43
135.2 195.7 315.4 622.1
1.80 1012 1.00 1013 7.00 1014 1.88 1014
the SS-LFP/C sample at temperatures of 25, 0, 10, and 20 °C were 2.61 1014, 3.00 1015, 9.71 1015, and 3.65 1016 cm2 s1, respectively. Accordingly, the Rb values of the 3 wt.%LTO-coated SP-LFP/C composite at temperatures of 25, 0, 10, and 20 °C were 5.67, 8.19, 8.85, and 11.43 ohm, respectively. The lower temperature
1426
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
-0.5 -0.6 (a)
Log (1/Rb) /ohm-1
-0.7
E a = 13.19 kJ mol
-1
(b)
-0.8 -0.9
E a = 10.40 kJ mol
-1
-1.0 -1.1 -1.2 (a) SS-LFP/C (b) 3%LTO-coated SP-LFP/C
-1.3 -1.4 3.2
3.4
3.6
3.8
4.0
4.2
Log (1/Rct) /ohm-1
1000/T (K-1) -1.8 -1.9 -2.0 -2.1 -2.2 -2.3 -2.4 -2.5 -2.6 -2.7 -2.8 -2.9 -3.0 -3.1 -3.2 -3.3 -3.4
(a) SS-LFP (b) 3%LTO-coated SP-LFP/C
E a,Rct = 23.07 kJ mol
E a,Rct = 42.12 kJ mol
-1
(a)
-1
(b)
3.2
3.4
3.6
3.8
4.0
4.2
10, and 20 °C were 135.2, 195.7, 315.4, and 622.1 ohm, respectively. The Di values of the 3 wt.%LTO-coated SP-LFP/C composite at temperatures of 25, 0, 10, and 20 °C were 1.80 1012, 1.00 1013, 7.00 1014, and 1.88 1014 cm2 s1, respectively. This demonstrates that the low-temperature performance of the LTO-coated SP-LFP/C composite with spherical morphology and LTO surface modification was more favorable than that of the SSLFP/C with an irregular shape and without LTO surface modification. It was found that the Rct and Di values of the 3 wt.%LTO-coated SP-LFP/C composite were considerably higher than those of the SSLFP/C material. This was because of the improvement in transport properties, in particular, for Rct and Di. Fig. 10(a) and (b) shows the Arrhenius plot of the SS-LFP/C and 3 wt.%LTO-coated SP-LFP/C composite based on Rb and Rct values at various test temperatures, (i.e., 25, 0, 10, and 20 °C). According to the plot, the activation energy values of the SS-LFP/C and 3 wt.%LTO-coated SP-LFP/C composite can be determined. The activation energy values (Ea,Rb) of the SS-LFP/C sample and 3 wt.%LTOcoated SP-LFP/C composite based on Rb were 13.19 and 10.40 kJ mol1, respectively. This Ea,Rb value is closely related to the current collector, active material conductivity. It seems about the magnitude order for two as-prepared LFP cathode materials. However, the activation energy values (Ea,Rct) of the SS-LFP/C sample and 3 wt.%LTO-coated SP-LFP/C composite based on Rct were 42.12 and 23.07 kJ mol1, respectively. This Ea,Rct value was closely related to the charge transfer at the interface of the electrode and electrolyte. The high Ea,Rct value of the SS-LFP/C sample (ca. 42.12 kJ mol1) indicated a less favorable electrochemical performance, compared with that of the 3 wt.%LTO-coated SP-LFP/C composite (ca. 23.07 kJ mol1). Consequently, through controlling (spherical) morphology by an SP process and doing the surface modification (LTO-coating) on the LFP/C material, the electrochemical performance and rate capability at a lower temperature can improve. Fig. 11 shows the schematic diagram for the preparation LTO-coated SP-LFP/C composite through a spray dry process and LTO surface modification process.
1000/T (K-1) Fig. 10. The Arrhenius plot for: (a) SS-LFP/C and (b) 3 wt.%LTO-coated SP-LFP/C composite.
was in the test, the higher was the value of Rb. This was because of the lower ionic conductivity of the electrolyte. The values of Rct and Di for the composite samples varied considerably. The Rct values of 3 wt.%LTO-coated SP-LFP/C composite at temperatures of 25, 0,
4. Conclusion The SP-LFP/C sample was first prepared using the solid-state and the SP methods. The surface modification was further conducted on the SP-LFP/C material by using LTO coating material to improve the rate capability and cycling stability properties at low temperature (0 °C) and elevated temperature (55 °C). The charac-
Fig. 11. Schematic diagram of fabrication LTP-coated SP-LFP/C composite via a spray dry (SP) process and a LTO surface modification method.
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
teristic properties were examined by employing XRD, micro-Raman spectroscopy, SEM, the AC impedance method, and the galvanostatic charge–discharge method. Consequently, the 3 wt.%LTO-coated SP-LiFePO4/C composite exhibited the highest specific discharge capacities of 150, 141, 131, 110, 103, and 84 mA h g1 at rates of 0.2, 0.5, 1, 3, 5, and 10 C, respectively, at 25 °C. It was also demonstrated that the 3 wt.%LTO-coated SPLFP/C composite exhibited the lowest fading rate at 55 °C (0.028 mA h cycle1) at 3 C rate for 200 cycles test, as compared with the bare SP-LFP/C material (0.405 mA h cycle1). As a result, the bare SP-LFP/C sample showed a less a favorable high-rate capability at elevated temperature. According to our experimental results, the fading rate of the LiFePO4 material can be greatly reduced through LTO surface modification. It was found that the optimum amount of LTO coating needed to produce stable cycling performance was 3 wt.%. In fact, the LTO material with spinel structure plays an important role in improving the performance at elevated temperature. It is because that the LTO coating material here has two functions, the first one is as an ionic conductor layer to keep high Li+ ionic conductivity. The second one is as a barrier lay to prevent Fe2+ dissolution. The LTO-coated SP-LFP/C composite can be a good candidate for lithium-ion batteries in high-power and high-temperature applications. Acknowledgement Financial support from the National Science Council, Taiwan (Project No: NSC 102-2632-E-131-001-MY3) is gratefully acknowledged. References [1] Padhi AK, Nanjundaswamy KS, Goodenough JB. Phospho-olivines as positiveelectrode materials for rechargeable lithium batteries. J Electrochem Soc 1997;144:1188–94. [2] Chung SY, Chiang YM. Microscale measurements of the electrical conductivity of doped LiFePO4. Electrochem Solid State Lett 2003;6:A278–81. [3] Prosini PP, Lisi M, Zane D, Pasquali M. Determination of the chemical diffusion coefficient of lithium in LiFePO4. Solid State Ionics 2002;148:45–51. [4] Meethong N, Huang HYS, Speakman SA, Carter WC, Chiang YM. Strain accommodation during phase transformations in olivine-based cathodes as a materials selection criterion for high-power rechargeable batteries. Adv Funct Mater 2007;17:1115–23. [5] Yamada A, Chung SC, Hinokuma K. Optimized LiFePO4 for lithium battery cathodes. J Electrochem Soc 2001;148:A224–9. [6] Yang JS, Xu JJ. Nonaqueous sol–gel synthesis of high-performance LiFePO4. Electrochem Solid State Lett 2004;7:A515–8. [7] Sides CR, Croce F, Young VY, Martin CR, Scrosati B. A high-rate, nanocomposite LiFePO4/carbon cathode. Electrochem Solid State Lett 2005;8:A484–7. [8] Yang SF, Zavalij PY, Whittingham MS. Hydrothermal synthesis of lithium iron phosphate cathodes. Electrochem Commun 2001;3:505–8. [9] Chen JJ, Whittingham MS. Hydrothermal synthesis of lithium iron phosphate. Electrochem Commun 2006;8:855–8.
1427
[10] Chen J, Wang S, Whittingham MS. Hydrothermal synthesis of cathode materials. J Power Sources 2007;174:442–8. [11] Chen JJ, Vacchio MJ, Wang SJ, Chernova N, Zavalij PY, Whittingham MS. The hydrothermal synthesis and characterization of olivines and related compounds for electrochemical applications. Solid State Ionics 2008;178:1676–93. [12] Doeff MM, Hu YQ, McLarnon F, Kostecki R. Effect of surface carbon structure on the electrochemical performance of LiFePO4. Electrochem Solid State Lett 2003;6:A207–9. [13] Huang XJ, Yan SJ, Zhao HY, Zhang L, Guo R, Chang CK, et al. Electrochemical performance of LiFePO4 nanorods obtained from hydrothermal process. Mater Charact 2010;61:720–5. [14] Teng F, Santhanagopalan S, Lemmens R, Geng XB, Patel P, Meng DD. In situ growth of LiFePO4 nanorod arrays under hydrothermal condition. Solid State Sci 2010;12:952–5. [15] Liao LX, Zuo PJ, Ma YL, Chen XQ, An YX, Gao YZ, et al. Effects of temperature on charge/discharge behaviors of LiFePO4 cathode for Li-ion batteries. Electrochim Acta 2012;60:269–73. [16] Wu L, Zhong SK, Liu JQ, Lv F, Wan K. High tap-density and high performance LiFePO4/C cathode material synthesized by the combined sol spray-drying and liquid nitrogen quenching method. Mater Lett 2012;89:32–5. [17] Konarova M, Taniguchi I. Preparation of carbon coated LiFePO4 by a combination of spray pyrolysis with planetary ball-milling followed by heat treatment and their electrochemical properties. Powder Technol 2009;191:111–6. [18] Konarova M, Taniguchi I. Synthesis of carbon-coated LiFePO4 nanoparticles with high rate performance in lithium secondary batteries. J Power Sources 2010;195:3661–7. [19] Zhong SK, Wu L, Zheng JC, Liu JQ. Preparation of high tap-density 9LiFePO4. Li3V2(PO4)3/C composite cathode material by spray drying and post-calcining method. Powder Technol 2012;219:45–8. [20] Liao XZ, Ma ZF, Gong Q, He YS, Pei L, Zeng LJ. Low-temperature performance of LiFePO4/C cathode in a quaternary carbonate-based electrolyte. Electrochem Commun 2008;10:691–4. [21] Yang KR, Deng ZH, Suo JS. Synthesis and characterization of LiFePO4 and LiFePO4/C cathode material from lithium carboxylic acid and Fe3+. J Power Sources 2012;201:274–9. [22] Zhang YC, Wang CY, Tang XD. Cycling degradation of an automotive LiFePO4 lithium-ion battery. J Power Sources 2011;196:1513–20. [23] Song HS, Cao Z, Chen X, Lu H, Jia M, Zhang ZA, et al. Capacity fade of LiFePO4/graphite cell at elevated temperature. J Solid State Electr 2013;17:599–605. [24] Kim JH, Woo SC, Park MS, Kim KJ, Yim T, Kim JS, et al. Capacity fading mechanism of LiFePO4-based lithium secondary batteries for stationary energy storage. J Power Sources 2013;229:190–7. [25] Cong LN, Gao XG, Ma SC, Guo X, Zeng YP, Tai LH, et al. Enhancement of electrochemical performance of Li[Li0.2Mn0.54Ni0.13Co0.13]O2 by surface modification with Li4Ti5O12. Electrochim Acta 2014;115:399–406. [26] Yi TF, Shu J, Zhu YR, Zhou AN, Zhu RS. Structure and electrochemical performance of Li4Ti5O12-coated LiMn1.4Ni0.4Cr0.2O4 spinel as 5 V materials. Electrochem Commun 2009;11:91–4. [27] Yao J, Shen C, Zhang P, Ma CA, Gregory DH, Wang L. Spinel-Li3.5+xTi5O12 coated LiMn2O4 with high surface Mn valence for an enhanced cycling performance at high temperature. Electrochem Commun 2013;31:92–5. [28] Zhu YR, Yi TF, Zhu RS, Zhou AN. Increased cycling stability of Li4Ti5O12-coated LiMn1.5Ni0.5O4 as cathode material for lithium-ion batteries. Ceram Int 2013;39:3087–94. [29] Yi TF, Shu J, Wang Y, Xue J, Meng J, Yue CB, et al. Effect of treated temperature on structure and performance of LiCoO2 coated by Li4Ti5O12. Surf Coat Technol 2011;205:3885–9.
Contents lists available at ScienceDirect
Applied Energy journal homepage: www.elsevier.com/locate/apenergy
Study of electrochemical performances of lithium titanium oxide–coated LiFePO4/C cathode composite at low and high temperatures q Chun-Chen Yang a,c,⇑, Jer-Huan Jang a,b, Jia-Rong Jiang a,c a
Department of Chemical Engineering, Ming Chi University of Technology, New Taipei City 243, Taiwan, ROC Department of Mechanical Engineering, Ming Chi University of Technology, New Taipei City 243, Taiwan, ROC c Battery Research Center of Green Energy, Ming Chi University of Technology, New Taipei City 243, Taiwan, ROC b
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
Spherical LiFePO4/C (LFP/C)
The surface modification of the LiFePO4 cathode material is done by a dual-layer coating (LTO and C layers) via a spray dry and a sol–gel method.
composites are prepared by a spray dry method. The 1–5 wt.%Li4Ti5O12 (LTO) coating is carried out on SP-LFP/C. The dual surface coating with LTO and carbon layers greatly improve the cycle life performance at 55 °C at 3 C rate. The LTO and carbon layers on LFP are functions as the electron and ionic conductors, respectively.
a r t i c l e
i n f o
Article history: Received 1 September 2014 Received in revised form 27 January 2015 Accepted 27 January 2015 Available online 5 March 2015 Keywords: Spray dry (SP) LiFePO4 Li4Ti5O12 (LTO) Surface modification Lithium-ion batteries
a b s t r a c t Spherical porous LiFePO4/C (LFP/C) composite materials were prepared by employing a solid-state (SS) method and a spray dry (SP) method. The surface modification was conducted on the spherical LFP/C composite using 1–5 wt.%Li4Ti5O12 (LTO) to improve the rate capability and the cycle stability properties at low temperature (0 °C) and elevated temperature (55 °C). The characteristic properties were examined through X-ray diffraction, micro-Raman spectroscopy, scanning electron microscopy, an AC impedance method, and galvanostatic charge–discharge method. The characteristic properties of the SS-LFP/C, SP-LFP/C, and LTO-coated SP-LPF/C composites were studied and compared. The 3 wt.%LTO-coated SP-LFP/C composite exhibited the most favorable performance among the samples. It exhibited discharge capacities of 150, 141, 131, 110, 103, and 84 mA h g1, at rates of 0.2 C, 0.5 C, 1 C, 3 C, 5 C, and 10 C, respectively. It was also found that the 3 wt.%LTO-coated SP-LFP/C composite shows the best cycling stability performance at
q This article is based on a short proceedings paper in Energy Procedia Volume 161 (2014). It has been substantially modified and extended, and has been subject to the normal peer review and revision process of the journal. This paper is included in the Special Issue of ICAE2014 edited by Prof. J Yan, Prof. DJ Lee, Prof. SK Chou, and Prof. U Desideri. ⇑ Corresponding author at: Department of Chemical Engineering, Ming Chi University of Technology, New Taipei City 243, Taiwan, ROC. Tel.: +886 2 2908 9899; fax: +886 2 2908 5941. E-mail address: [email protected] (C.-C. Yang).
http://dx.doi.org/10.1016/j.apenergy.2015.01.131 0306-2619/Ó 2015 Elsevier Ltd. All rights reserved.
1420
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
elevated temperature of 55 °C. The LTO-coated SP-LFP/C composite was demonstrated to be suitable for high-temperature and high-power application in lithium-ion batteries. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Because of their relatively high specific capacity, lithium-ion batteries are used in electric vehicles, cell phones, laptop computers, digital cameras, renewable-energy storage, and smart grid applications. The LiFePO4 cathode material proposed by Padhi et al. [1] has attracted extensive attention for application in the next generation of rechargeable lithium-ion batteries because of its low cost, environmental friendliness, excellent safety characteristics, high capacity (approximately 170 mA h g1 in theory), and excellent cycling performance. However, its low conductivity in electron and ion transfer constitutes a major barrier for use in commercial applications. The electron conductivity of LiFePO4 is only 109 S cm1 [2] and its lithium-ion diffusivity is 1014 to 1016 cm2 s1 [3]. Three methods can be used to improve the electronic conductivity: (1) carbon coating; (2) doping with supervalent cations; (3) decreasing the particle size. The carbon coating is the most effective among these methods, and is a facile method of improving the conductivity of LiFePO4 materials. LiFePO4 cathode materials can typically be prepared through a solid-state reaction [4,5], sol–gel process [6,7], or hydrothermal process [8–14]. Whittingham et al. [8] revealed that high-quality crystalline LiFePO4 platelets can be obtained using a low-temperature hydrothermal process. A spray dry (SP) method has been widely used to prepare spherical LiFePO4/C cathode materials to enhance the electrochemical performance [15–19]. The low- and high-temperature performance and capacity fading mechanism of LiFePO4/C materials have recently been intensively explored [20–24]. Surface coating is an effective method of solving the problems of high-temperature metal dissolution and cycling deterioration. The spinel Li4Ti5O12 (LTO) surface modification of various cathode materials has become crucial for improving low- and high-temperature electrochemical properties [25–29]. Cong et al. [25] prepared a 3 wt.%Li4Ti5O12 (3%LTO)-coated Li[Li0.2Mn0.54Ni0.13Co0.13]O2 lithium-rich cathode material by using a ball-milled process. The lithium-rich oxides exhibited cycle-performance and coulombic efficiency improvement because LTO provides a numerous sites for insertion of extracted lithium. Yi et al. [26] reported the electrochemical performance of LTO-coated LiMn1.4Ni0.4Cr0.2O4 material as a 5V cathode. They showed that 4 wt.%LTO-coated LiMn1.4Ni0.4Cr0.2O4 particles exhibited the highest electrochemical performance. Yao et al. [27] synthesized Li3.5+xTi5O12-coated and LTO-coated LiMn2O4 cathode materials by using the sol–gel method. Their study demonstrated that the valence of surface Mn for the Li3.5+xTi5O12-coated LiMn2O4 material increased after annealing; the material exhibited a favorable cycling performance at 55 °C. Zhu et al. [28] also prepared an LTO-coated LiMn1.5Ni0.5O4 spinel material by using the sol–gel method followed by high-temperature calcination. They demonstrated that a 3wt.%LTO-coated layer protected the surface of the LiMn1.5Ni0.5O4 active material from an HF attack in the electrolyte. Yi et al. [29] prepared LTO-coated LiCoO2 powders by using a sol– gel method at various sintering temperatures and coating contents. The results demonstrated that the LTO-coated layer on the LiCoO2 suppressed the capacity fading by reducing cell polarization. No studies have been published that investigate LTO-coated LiFePO4/C. In this study, we first prepared solid-state LiFePO4/C (denoted as SS-LFP/C) materials by using a solid-state method. The SP method and post-sintering were used to prepare a spherical porous
SP-LFP/C composite. A suitable amount of glucose was added as a binder and carbon source in the SP processing. To improve the low- and high-temperature performance of the SP-LFP/C material, LTO surface modification was applied to the SP-LFP/C material by using the sol–gel method. For comparison, an SP-LiFePO4/C material without LTO surface modification was also examined. The characteristic properties of the SP–LiFePO4/C cathode material with and without LTO coating were examined through X-ray diffraction (XRD), micro-Raman spectroscopy, scanning electron microscopy (SEM), elemental analysis, and micro-Raman spectroscopy. The electrochemical performances of the Li/LFP coin cell were examined using an automatic galvanostatic charge–discharge unit and a cyclic voltammetry (CV) method. 2. Experimental 2.1. Preparation of SP–LiFePO4/C and LTO-coated SP–LiFePO4/C composites The LiFePO4/C materials (SS-LFP/C) were prepared using a solid state method. The appropriate quantities of the raw materials LiH2PO4, FeC2O42H2O, 5% sucrose, and 5% citric acid (Aldrich) were mixed in acetone and then ball-milled at a rotation speed of 400 rpm for 6 h in a planetary miller with a ball/reactant weight ratio of 6:1. The added carbon precursors in the LiFePO4 materials was maintained at 10 wt.%. The ball-milled mixture precursor was pre-sintered at 350 °C for 4 h and then sintered at 700 °C for 12 h in an Ar/H2 (95:5, v/v) atmosphere. The spherical SP–LiFePO4/C material were prepared using an SP method and a post-sintering process. The appropriate quantities of the as-prepared SS-LiFePO4/C materials and 5% glucose were mixed and dispersed in deionized water. The spherical porous SP-LFP/C material with a double carbon source was prepared using a spray dryer (EYELA, Spray Dryer SD-1000 model, Japan). The spray-dried spherical LFP/C composite precursor was further calcined at 700 °C for 5 h in an Ar atmosphere. The 1–5 wt.%LTO-coated SP-LFP/C composites were synthesized using the sol–gel method. Ti(C4H9O)4 and CH3COOLi2H2O at a Li:Ti stoichiometric molar ratio of 4:5 were dissolved in a solution containing ethanol and distilled water to form a clear solution. A small amount of HNO3 was then added to this solution under continuous stirring to obtain a sol. The spherical SP-LFP/C powder was slowly added to the sol under constant stirring. After hydrolyzing and condensation for 10 h, a black gel formed and was dried at 120 °C for 2 h and then calcined at 700 °C for 3 h in an Ar atmosphere to obtain the final 3%LTO-coated SP-LFP/C composite. 2.2. Characterization The crystal structure of the SS-LFP/C, SP-LFP/C, and LTO-coated SP-LFP/C composite samples was examined using an XRD spectrometer (Philip, X’pert Pro System). The surface morphology was examined using a scanning electron microscope (SEM, Hitachi). The micro-Raman spectra were recorded on a confocal micro-Renishaw with a 632 nm He–Ne laser excitation. The residual carbon content in the sample was analyzed using an elemental analyzer (Perkin Elmer 2400). The electron conductivity of the composite samples was measured using an AC impedance method.
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
2.3. Electrochemical measurements The electrochemical performances of the Li/LiFePO4 half-cell batteries were measured using a two-electrode system (a CR 2032 coin cell assembled in an argon-filled glove box). All LiFePO4/ C electrodes were prepared by mixing active LiFePO4/C materials, Super P, and a poly(vinyl fluoride) binder at a weight ratio of 90:5:5, pasted on an aluminum foil (Aldrich), and finally dried in a vacuum oven at 120 °C for 12 h. The lithium foil (Aldrich) was used as the counter and reference electrode. A microporous PE film was used as a separator. The electrolyte was 1 M LiPF6 in a mixture of EC and DEC (1:1 in v/v, Merck). The half-cell LiFePO4/Li batteries were charged using a constant current profile and discharged using a constant current profile, over a potential range of 2.0–3.8 V (vs. Li/Li+) at varied C rates with an Autolab PGSTAT302N potentiostat. The CV was conducted using an Autolab instrument at a scanning rate of 0.1 mV s1, between 2.5 and 4.2 V. 3. Results and discussion The XRD patterns of the SS-LFP/C, SP-LFP/C, and 3 wt.%LTOcoated SP-LFP/C composite prepared using the solid state method and SP method are shown in Fig. 1. The XRD diffraction patterns revealed that the as-prepared SS-LFP/C, SP-LFP/C, and 3 wt.%LTOcoated SP-LFP/C composite were single-phase materials with an olivine-type structure indexed in the orthorhombic Pnma space group. The diffraction peak of the residual carbon could not be determined in the pattern and may be their low content or amorphous state. The lattice parameters (a, b, c, and cell volume) of the SS-LFP/C, SP-LFP/C, and 3 wt.%LTO-coated SP-LFP/C composites were calculated based on the XRD patterns. The lattice parameters of all LFP/C samples were the same as those of the standard LiFePO4 (JCPDS card number 81-1173, a = 10.33 Å, b = 6.010 Å, c = 4.692 Å, V = 291.35 (Å)3). The LTO and carbon source of glucose and citric acid had no observable influence on the structure of the LFP/C composite materials. No impurity phase occurred. The SEM image of the SS-LFP/C material prepared using the solid state method is shown in Fig. 2(a). Generally, the particle size increased with increasing sintering temperature. The particle size of the SS-LFP/C sample was in the range of 0.5–2 lm. Consequently, the addition of citric acid and sucrose as the carbon source not only decreased the particle size, but also prevented the particle from growing during the sintering process. The SEM image of the spherical SP-LFP/C material prepared using the SP method and sintering at 700 °C, is shown in
(a) SS-LFP/C (b) SP-LFP/C (c) 3%LTO-coated SP-LFP/C
Intensity/ a.u.
(a)
(b)
(c)
10
20
30
40
50
60
70
2θ/ degree Fig. 1. XRD patterns of SS-LFP/C, SP-LFP/C, 3 wt.%LTO-coated SP-LFP/C samples.
1421
Fig. 2(b). The particle size of the SP-LFP/C composite sample was in the range of 2–10 lm. Numerous micro-pores formed on the surface of the SP-LFP/C material and acted as a micro-reservoir for the electrolyte. The SP-LFP/C composite with highly porous and spherical morphology can markedly enhance the mass-transport properties. Fig. 2(c) depicts the SEM image of a 3 wt.%LTOcoated SP-LFP/C composite prepared using the sol–gel method. The particle size of the 3 wt.%LTO-coated SP-LFP/C composite was also in the range of 2–10 lm. The LTO surface modification did not exert any influence on the morphology of the SP-LFP/C composite. It was also found that there are many nano-pores (100– 200 nm) formed on the top-surface of the 3%LTO-coated SP-LFP/C material and it can be function as the electrolyte reservoir. The 3%LTO-coated SP-LFP/C composite with a thin LTO oxide barrier layer substantially improved the electrochemical properties at elevated temperature (see later cycling performance). In addition, micro-Raman spectra of the SS-LFP/C, SP-LFP/C, and 3 wt.% LTO-coated SP-LFP/C composites were prepared using the solid state, SP, and sol–gel methods, respectively, and are depicted in Fig. 3. The micro-Raman peaks of the SS-LFP/C material, located at approximately 1328 and 1590 cm1, were attributable to citric acid and sucrose carbon. Raman peaks at approximately 1328 cm1 (D band) and 1590 cm1 (G band) were observed in the composite samples. The micro-Raman peaks of the SP-LFP/C composites, located at approximately 1332 and 1591 cm1, were attributable to residual carbon of glucose and citric acid. Micro-Raman peaks at approximately 1330 cm1 (D band) and 1586 cm1 (G band) were also observed in the 3%LTO-coated SP-LFP/C composites. The broadening of the D (A1g symmetry) band and G (E2g symmetry) band with a strong D band indicated a localized inplane sp2 graphitic crystal domain and disordered sp3 amorphous carbon. The intensity ratio of the D band versus the G band (denoted as R = ID/IG) was employed to estimate the carbon quality of the LiFePO4/C samples. The R value of the 3 wt.% LTO-coated SP-LFP/C and bare SP-LFP/C composites was approximately between 0.97 and 1.07. By comparison, the R value of the SS-LFP/C samples without spray-dry and LTO oxide coating was approximately 1.10. It is well accepted [11] that the discharge capacity and the rate capability of the LiFePO4/C samples are closely related to the intensity ratio of the D band and the G band. Various micro-Raman peaks located approximately in the ranges of 947–950 cm1 and 596– 446 cm1 were identified as resulting from the vibration of the P–O bond, and the peak located at 638 cm1 was identified as resulting from the vibration of the FeOx groups. It was revealed that the R value of the 3%LTO-coated SP-LFP/C shows the lowest value. It may be due to the surface modification and thermal treatment synergistic effect. The thin LTO layer may re-modify parts of carbon defective area; it results in much uniform and dense carbon layer. Another possible reason is that the thermal sintering treatment during the LTO coating process cause to remove some defective-type carbons residuals; therefore, the R value was slightly reduced. The electrochemical properties of LiFePO4/C samples were strongly related to the electronic conductivity, the total residual carbon content, and the value of the ID/IG ratio. Both the electronic conductivity and the total residual carbon content substantially affected the discharge capacity and the rate capability of the LFP/ C composites. The electronic conductivities (re) and the residual carbon (C%) content of all LiFePO4/C samples were approximately 1.10–3.67 104 S cm1 and 3.10–4.2%, respectively. These values are consistent with findings reported in the literatures [1,2]. By contrast, the electronic conductivity of the 3 wt.% LTO-coated LFP/C composites was slightly higher than those of the SS-LFP/C and SP-LFP/C samples. As shown in Fig. 4, this is the CV curves for the SS-LFP/C (by a solid-state ball-milled method), SP-LFP/C (by a spray dry method),
1422
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
Fig. 2. SEM images of (a) SS-LFP/C; (b) SP-LFP/C, and (c) 3 wt.%LTO-coated SP-LFP/C composite.
(a) SS-LFP/C with ID /I G = 1.10 (b) SP-LFP/C with I D /I G = 1.07
D
G
(a) SS-LFP/C (w/o spray dry) (b) 0%LTO-SP-LFP/C (c) 3% LTO-SP-LFP/C (d) 5% LTO-SP-LFP/C
3.0
(c) 3%LTO SP-LFP/C with I D /I G = 0.97
2.5
(b) (c)
2.0
Intensity/ a.u.
(c)
Current/ mA
1.5
(b)
(d)
1.0 0.5
(a)
0.0 -0.5
(a)
-1.0
200
400
600
800
1000 1200 1400 1600 1800 2000
Raman shift/ cm—1 Fig. 3. Micro-Raman spectra of SS-LFP/C, SP-LFP/C, and 3 wt.%LTO-coated SP-LFP/C composites.
3 and 5 wt.% LTO-coated SP-LFP/C (by a spray dry + sol–gel surface modification) composite materials. The oxidation peak (charge) intensity of the SP-LFP/C material is higher than that of the 3 wt.%LTO-coated SP-LFP/C composite. However, the reduction peak (discharge) intensity of the SP-LFP/C material is much lower than that of the 3%LTO-coated SP-LFP/C composite. By comparison, the total integrated area (charge) from the CV curve of the 3 wt.%LTO-coated SP-LFP/C composite is slightly larger than that of the SP-LFP/C material. This indicated that the current efficiency of the 3%LTO-coated SP-LFP/C composite may be slightly larger than that of the SP-LFP/C material. Indeed, the SP-LFP/C material shows a slightly fast kinetic performance. It is because that the extra LTO-coated layer on the surface SP-LFP/C material induces
-1.5 -2.0 2.6
2.8
3.0
3.2
3.4
3.6
3.8
4.0
Potential/ V (vs. Li+/Li) Fig. 4. The CV curves of SS-LFP/C, SP-LFP/C, 3 and 5 wt.%LTO-coated SP-LFP/C samples at 0.1 mV s1.
a barrier to the charge transfer reaction. It results in some polarization (i.e., charge-transfer resistance increase) on the 3 wt.%LTOcoated SP-LFP/C composite. However, we also discovered that the 5 wt.%LTO-coated SP-LFP/C composite electrode showed poor kinetics properties, as compared with the 3 wt.%LTO-coated SPLFP/C electrode. It may be due to too thick LTO-coated layer, exhibiting much lower charge and discharge peak currents. As seen clearly in Fig. 5, the poorest initial cycle charge/discharge performance at 0.1 C at 25 °C among the LFP/C materials is SS-LFP/C; the charge and discharge capacities are ca. 144 and 133 mA h g1, respectively. In contrast, the charge and discharge
1423
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
4000
(A) SS-LFP/C (in 2.0V~3.8V range) 3800 3600
3500
3400
3250
3200
Voltage/ mV
Voltage/ mV
3750
3000 2750
0.2C/0.2C 0.2C/0.5C 0.2C/1C 0.2C/3C 0.2C/5C 0.2C/10C
3000 2800 2600
(a) 2500
2400
(a) SS-LFP (w/o spray dry) (b) 0%LTO-SP-LFP (c) 3%LTO-SP-LFP/C (d) 5%LTO-SP-LFP/C
2250 2000
0
20
40
60
80
100
2200 2000
(d) (b) (c) 120
140
160
0
20
40
60
80
100
120
140
Capacity/ mAh g-1
180
Capacity/ mAh g-1
(B) SP-LFP/C in 2.0V~3.8V range 4000
Fig. 5. The initial charge/discharge curves of SS-LFP/C, SP-LFP/C, and 3 and 5 wt.%LTO-coated SP-LFP/C samples at 0.1 C rate.
Voltage/ mV
3600 3400
0.2C/0.2C 0.2C/0.5C 0.2C/1C 0.2C/3C 0.2C/5C 0.2C/10C
3200 3000 2800 2600 2400 2200 2000 0
20
40
60
80
100
120
140
Capacity/ mAh g-1 (C) 3%LTO-coated SP-LFP/C in 2.0V~3.8V range 4000 3800 3600 3400
Voltage/ mV
capacities of the SP-LFP/C material are ca. 157 and 152 mA h g1, respectively. On the contrary, we find that the charge and discharge capacities of the 3%LTO-coated SP-LFP/C composite material are 170 and 151 mA h g1, respectively. The extra LTO-coated layer on the surface of SP-LFP/C sample may produce the extra barrier layer to cause some polarization. That is why the discharge capacity is slightly reduced. Surprisingly, it was seen that the charge and discharge capacities of the 5 wt.%LTO-coated SP-LFP/C composite are around 152 and 134 mA h g1, respectively. The specific discharge capacity was greatly reduced; it may be due to the too thick oxide barrier layer (LTO) on SP-LFP/C material. However, the oxide barrier layer can fully isolate the cathode active material direct contact with electrolyte (or avoid HF attack). More importantly, the LTO coating layer can also function as an ionic conductor layer; it will help improve the lithium ionic conductivity. It is because that the LTO oxide with the spinel NASCION structure shows the high ionic conductivity. It can markedly improve the high temperature cycle-life performance at 55 °C, in particular, at 3 C discharge rate. The typical charge–discharge profiles of the SS-LFP/C and SPLFP/C, and 3 wt.%LTO-coated SP-LFP/C composite at 25 °C and varying rates of 0.2–10 C are displayed in Fig. 6. All LFP/C samples revealed typical flat potential plateaus at 3.3–3.4 versus Li/Li+ at rates of 0.2–0.5 C. The potential plateau sloped when the discharge rate was higher than 1 C. The bare SP-LFP/C material exhibited specific discharge capacities of 135, 131, 122, 103, 94, and 78 mA h g1 at rates of 0.2, 0.5, 1, 3, 5, and 10 C, respectively. In contrast, the 3%LTO-coated SP-LFP/C composite exhibited specific discharge capacities of 150, 141, 131, 110, 103, and 84 mA h g1 at rates of 0.2, 0.5, 1, 3, 5, and 10 C, respectively. However, the SS-LFP/C sample with irregular morphology and without LTO surface modification only exhibited the specific discharge capacities of 125, 116, 105, 87, 76, and 48 mA h g1 at rates of 0.1, 0.2, 0.5, 1, 3, 5 and 10 C, respectively. Fig. 7 provides a comparison of the rate capability of the SS-LFP/C, SP-LFP/C, and 3 wt.%- and 5 wt.%LTO-coated SP-LFP/C composite at various rates. Both 3D porous spherical morphology (achieved using the SP method) and 3 wt.%LTO surface modification (achieved using the sol–gel method) on the LiFePO4 material substantially improved the highrate capability. It is well accepted [19–24] that the high concentration Fe2+ dissolving from LFP/C material at elevated temperature is one of the
3800
3200 3000 2800 0.2C/0.2C 0.2C/0.5C 0.2C/1C 0.2C/3C 0.2C/5C 0.2C/10C
2600 2400 2200 2000 0
20
40
60
80
100 120 140 160 180
Capacity/ mAh g-1 Fig. 6. The charge–discharge curve of (a) SS-LFP/C, (b) SP-LFP/C, (c) 3 wt.%LTOcoated SP-LFP/C composites at varied rates.
main reasons for the occurrence of a capacity-fading mechanism. According to many research results [19–24], it was clearly indicated that the HF generated from the LiPF6 electrolyte was responsible for the dissolution of Fe during cycling. The uncoated LFP/C cathode material in the LiPF6 electrolyte reacted with the HF and thus
1424
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
160
(a) 0%LTO-SP-LFP/C at 25oC (b) 0%LTO-SP-LFP/C at 55oC (c) 0%LTO-SP-LFP/C at 0oC (d) 3%LTO-SP-LFP/C at 25oC (e) 3%LTO-SP-LFP/C at 55oC (f) 3%LTO-SP-LFP/C at 0oC (g) 5%LTO-SP-LFP/C at 25oC (h) 5%LTO-SP-LFP/C at 55oC o (i) 5%LTO-SP-LFP/C at 0 C
140
160 140
(b)
(e)
80
(b) (d)
60
(a)
40 20
LFP at 1C/3C o at 55, 25, and 0 C (b)
100
(a) SS-LFP/C (b) 0%LTO-SP-LFP/C (c) 3%LTO coted SP-LFP/C (d) 5%LTO coted SP-LFP/C
Capacity/ mAh g-1
Capacity/ mAh g-1
120
120 (h)
(d)
100
(a) (g)
80 60 (f)
40
0
0.2C 0.5C 1C 3C 5C 10C 1-3cycle 1-3cycle 1-3cycle 1-3cycle 1-3cycle 1-3cycle
(i)
(c)
20 20
Fig. 7. The rate capability of SS-LFP/C, SP-LFP/C, and 3 and 5 wt.%LTO-coated SP-LFP/C samples at 0.2–10 C rates.
40
60
80
100
120
140
160
180
200
Number of cycles Fig. 8. The cycling and CE performance for SS-LFP/C, SP-LFP/C, and 3 and 5 wt.%LTOcoated SP-LFP/C composites at 1 C/3 C rate at 25 and 55 °C.
led to a gradual capacity loss. LTO surface coating can prevent Fe2+ ions from directly contacting electrolyte and therefore greatly decrease the capacity loss [24,25]. Differently, LTO has a 3D spinel structure that also acts as an ionic conductor and shows high lithium-ion conductivity. Unlike pure oxides (e.g., Al2O3, ZrO2), they tend to block the ionic channel of LFP. By contrast, LTO can enhance the lithium ionic conductivity of LiFePO4/C cathode material. Moreover, it was found that the discharge capacity and the coulombic efficiency of the 0 wt.%, 3 wt.%, and 5 wt.%LTO coated SP-LFP/C composite at 3 C discharge rate were approximately 100-110 mA h g1 and 93–97%, respectively during 200 cycling tests at 25 °C, as shown in Fig. 8. No significant difference was observed. As a result, their fading rates of the 0 wt.%-, 3 wt.%and 5 wt.%LTO-coated SP-LFP/C composites were approximately 0.01–0.02 mA h cycle1 at 25 °C. More interestingly, the electrochemical performance of the SPLFP/C material with and without LTO coating at elevated temperature (55 °C) was evaluated and compared in details. We found that the specific discharge capacity of the SP-LFP/C material (without LTO coating) at 3 C and 55 °C is markedly decreased, i.e., from 148 to 67 mA h g1 with the fading rate of 0.405 mA h cycle1 for 200 cycles. In contrast, the discharge capacity of the 3 wt.%LTOcoated SP-LFP/C composite at 3 C and 55 °C for 200 cycles is smoothly decreased, i.e., from 123 to 117 mA h g1 with much lower fading rate of 0.028 mA h cycle1. As shown in Fig. 8, the specific discharge capacity of the 5 wt.%LTO-coated SP-LFP/C composite at 55 °C is modestly reduced, from 133 to 97 mA h g1 with the fading rate of 0.180 mA h cycle1 for 200 cycles. In fact, we was found that the 3 wt.%LTO coated SP-LFP/C composite exhibited the best performance at elevated temperature. By comparison the low temperature performance, we found that Furthermore, the discharge capacities and coulombic efficiencies of the 3 wt.%- and 5 wt.%LTO-coated SP-LFP/C composites at 0 °C were varied at 45–42 mA h g1 and 95–97%, respectively. The fading rates of the 3 wt.%- and 5 wt.%LTO-coated SP-LFP/C composites were about 0.015–0.020 mA h cycle1 at 0 °C. As a matter of fact, all SP-LFP/C samples with and without LTO surface modification exhibited pretty similar electrochemical performance, with the fading rate of 0.015–0.026 mA h cycle1 at 0 °C. It was seen clearly that the fading rate (0.405 mA h cycle1) of the bare SP-LFP/C cathode material at 55 °C showed much higher
than that of the bare SP-LFP/C at 0 °C (0.0260 mA h cycle1). It can be seen clearly that the capacity fading will be accelerated at elevated temperature. However, the fading rate of the bare SPLFP/C material (0.405 mA h cycle1) without LTO surface modification was considerably higher than that of the 3%LTO-coated SPLFP/C composite (0.028 mA h cycle1) at elevated temperature (55 °C). In conclusion, it was revealed that the LTO surface modification on LiFePO4/C material is effective way to reduce the fading rate at elevated temperature. We found that the optimum amount of LTO coating needed to produce stable cycling performance at elevated temperature (55 °C) was 3 wt.% of the SP-LFP/C material. The AC impedance spectroscopy was used to study the interface properties of the SS-LFP/C sample and the 3 wt.%LTO-coated SPLFP/C composite samples. The AC spectra of the 3 wt.%LTO-coated SP-LFP/C composite at temperatures of 25, 0, 10, and 20 °C, at an open circuit potential, are shown in Fig. 9(a). Each AC plot consisted of one semicircle at a higher frequency, followed by a linear portion at a lower frequency. The lower frequency region of the straight line was considered as the Warburg impedance. It is for long-range lithium-ion diffusion in bulk phase. The Rb indicates the bulk resistance at the electrolyte; Rct is the charge transfer resistance at the active material interface; CPE represents the double layer capacitance and certain surface film capacitances. The lithium chemical diffusion coefficients (Di) of the electrode were calculated based on Eq. (1) [19–24]:
Di ¼
2 1 RT 2 2 AF rw C
ð1Þ
where rw is the Warburg impedance coefficient (obtained a slope from plot of Zre vs. w0.5, as seen in Fig. 9(b)); Di is the lithium diffusion coefficient; R is the gas constant; T is the absolute temperature; F is the Faraday constant; A is the area of the electrode; and C is the molar concentration of Li+ ions (CLi+ = 1.0 103 mol cm3). Tables 1 and 2 show the calculated Rb, Rct, Di, and jo parameters for the SS-LFP/C sample and the 3 wt.%LTO-coated SPLFP/C composite. The Rb values of the SS-LFP/C sample at temperatures of 25, 0, 10, and 20 °C were 5.14, 7.25, 9.11, and 12.19 ohm, respectively. The lower temperature was in the test,
1425
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
(a)
(b)
1000
1000
o
(a) 20 C o (b) 0 C o (c) -10 C (d) -20oC
o
(a) 20 C (b) 0oC (c) -10oC (d) -20oC
900
900 800
800
700 600
600
-Z''/ ohm
-Z''/ ohm
700
500
500
(d)
400
400 (d)
(c)
300
300
(b) (c)
200
200
(b)
(a) (a)
100
100
0
0 0
200
400
600
800 1000 2001 1400 1600 1800 2000 2200 2400
0
200
400
Z'/ ohm
600
800
1000
Z'/ ohm 1500
2500
(d)
o
(a) 20 C o (b) 0 C (c) -10oC (d) -20oC
2250 2000
1250
(a) 20oC o (b) 0 C o (c) -10 C (d) -20oC
(d)
1750
1000
1250
Z'/ ohm
Zre /ohm
1500
(c)
(c)
750
1000 (b)
750
(b)
500
(a)
250
500 (a)
250 0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
0
W-1/2
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
W--1/2 Fig. 9. (a) The AC impedance spectra of SP-LFP/C composite and Z0 vs. w1/2 curve; (b) The AC impedance spectra of 3 wt.%LTO-coated SP-LFP/C composite and Z0 vs. w1/2 curve.
Table 1 Impedance parameters derived using equivalent circuit model and lithium diffusion coefficient (Di) for SS-LFP/C material. Temp (°C)
Param Rb (X)
25 0 10 20
5.14 7.25 9.11 12.19
Table 2 Impedance parameters derived using equivalent circuit model and lithium diffusion coefficient (Di) for 3 wt.%LTO-coated SP-LiFePO4/C composite. Temp (°C)
Rct (X) 123.5 615.0 1133.0 1862.0
Di (cm2 s1) 14
2.61 10 3.00 1015 9.71 1016 3.65 1016
the higher was the value of Rb. This was because of the lower ionic conductivity of the electrolyte. The values of Rct and Di for the SS-LFP/C samples varied considerably. The Rct values of the SSLFP/C sample at temperatures of 25, 0, 10, and 20 °C were 123.5, 615.0, 1133, and 1862 ohm, respectively. The Di values of
25 0 10 20
Param Rb (X)
Rct (X)
Di (cm2 s1)
5.67 8.19 8.85 11.43
135.2 195.7 315.4 622.1
1.80 1012 1.00 1013 7.00 1014 1.88 1014
the SS-LFP/C sample at temperatures of 25, 0, 10, and 20 °C were 2.61 1014, 3.00 1015, 9.71 1015, and 3.65 1016 cm2 s1, respectively. Accordingly, the Rb values of the 3 wt.%LTO-coated SP-LFP/C composite at temperatures of 25, 0, 10, and 20 °C were 5.67, 8.19, 8.85, and 11.43 ohm, respectively. The lower temperature
1426
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
-0.5 -0.6 (a)
Log (1/Rb) /ohm-1
-0.7
E a = 13.19 kJ mol
-1
(b)
-0.8 -0.9
E a = 10.40 kJ mol
-1
-1.0 -1.1 -1.2 (a) SS-LFP/C (b) 3%LTO-coated SP-LFP/C
-1.3 -1.4 3.2
3.4
3.6
3.8
4.0
4.2
Log (1/Rct) /ohm-1
1000/T (K-1) -1.8 -1.9 -2.0 -2.1 -2.2 -2.3 -2.4 -2.5 -2.6 -2.7 -2.8 -2.9 -3.0 -3.1 -3.2 -3.3 -3.4
(a) SS-LFP (b) 3%LTO-coated SP-LFP/C
E a,Rct = 23.07 kJ mol
E a,Rct = 42.12 kJ mol
-1
(a)
-1
(b)
3.2
3.4
3.6
3.8
4.0
4.2
10, and 20 °C were 135.2, 195.7, 315.4, and 622.1 ohm, respectively. The Di values of the 3 wt.%LTO-coated SP-LFP/C composite at temperatures of 25, 0, 10, and 20 °C were 1.80 1012, 1.00 1013, 7.00 1014, and 1.88 1014 cm2 s1, respectively. This demonstrates that the low-temperature performance of the LTO-coated SP-LFP/C composite with spherical morphology and LTO surface modification was more favorable than that of the SSLFP/C with an irregular shape and without LTO surface modification. It was found that the Rct and Di values of the 3 wt.%LTO-coated SP-LFP/C composite were considerably higher than those of the SSLFP/C material. This was because of the improvement in transport properties, in particular, for Rct and Di. Fig. 10(a) and (b) shows the Arrhenius plot of the SS-LFP/C and 3 wt.%LTO-coated SP-LFP/C composite based on Rb and Rct values at various test temperatures, (i.e., 25, 0, 10, and 20 °C). According to the plot, the activation energy values of the SS-LFP/C and 3 wt.%LTO-coated SP-LFP/C composite can be determined. The activation energy values (Ea,Rb) of the SS-LFP/C sample and 3 wt.%LTOcoated SP-LFP/C composite based on Rb were 13.19 and 10.40 kJ mol1, respectively. This Ea,Rb value is closely related to the current collector, active material conductivity. It seems about the magnitude order for two as-prepared LFP cathode materials. However, the activation energy values (Ea,Rct) of the SS-LFP/C sample and 3 wt.%LTO-coated SP-LFP/C composite based on Rct were 42.12 and 23.07 kJ mol1, respectively. This Ea,Rct value was closely related to the charge transfer at the interface of the electrode and electrolyte. The high Ea,Rct value of the SS-LFP/C sample (ca. 42.12 kJ mol1) indicated a less favorable electrochemical performance, compared with that of the 3 wt.%LTO-coated SP-LFP/C composite (ca. 23.07 kJ mol1). Consequently, through controlling (spherical) morphology by an SP process and doing the surface modification (LTO-coating) on the LFP/C material, the electrochemical performance and rate capability at a lower temperature can improve. Fig. 11 shows the schematic diagram for the preparation LTO-coated SP-LFP/C composite through a spray dry process and LTO surface modification process.
1000/T (K-1) Fig. 10. The Arrhenius plot for: (a) SS-LFP/C and (b) 3 wt.%LTO-coated SP-LFP/C composite.
was in the test, the higher was the value of Rb. This was because of the lower ionic conductivity of the electrolyte. The values of Rct and Di for the composite samples varied considerably. The Rct values of 3 wt.%LTO-coated SP-LFP/C composite at temperatures of 25, 0,
4. Conclusion The SP-LFP/C sample was first prepared using the solid-state and the SP methods. The surface modification was further conducted on the SP-LFP/C material by using LTO coating material to improve the rate capability and cycling stability properties at low temperature (0 °C) and elevated temperature (55 °C). The charac-
Fig. 11. Schematic diagram of fabrication LTP-coated SP-LFP/C composite via a spray dry (SP) process and a LTO surface modification method.
C.-C. Yang et al. / Applied Energy 162 (2016) 1419–1427
teristic properties were examined by employing XRD, micro-Raman spectroscopy, SEM, the AC impedance method, and the galvanostatic charge–discharge method. Consequently, the 3 wt.%LTO-coated SP-LiFePO4/C composite exhibited the highest specific discharge capacities of 150, 141, 131, 110, 103, and 84 mA h g1 at rates of 0.2, 0.5, 1, 3, 5, and 10 C, respectively, at 25 °C. It was also demonstrated that the 3 wt.%LTO-coated SPLFP/C composite exhibited the lowest fading rate at 55 °C (0.028 mA h cycle1) at 3 C rate for 200 cycles test, as compared with the bare SP-LFP/C material (0.405 mA h cycle1). As a result, the bare SP-LFP/C sample showed a less a favorable high-rate capability at elevated temperature. According to our experimental results, the fading rate of the LiFePO4 material can be greatly reduced through LTO surface modification. It was found that the optimum amount of LTO coating needed to produce stable cycling performance was 3 wt.%. In fact, the LTO material with spinel structure plays an important role in improving the performance at elevated temperature. It is because that the LTO coating material here has two functions, the first one is as an ionic conductor layer to keep high Li+ ionic conductivity. The second one is as a barrier lay to prevent Fe2+ dissolution. The LTO-coated SP-LFP/C composite can be a good candidate for lithium-ion batteries in high-power and high-temperature applications. Acknowledgement Financial support from the National Science Council, Taiwan (Project No: NSC 102-2632-E-131-001-MY3) is gratefully acknowledged. References [1] Padhi AK, Nanjundaswamy KS, Goodenough JB. Phospho-olivines as positiveelectrode materials for rechargeable lithium batteries. J Electrochem Soc 1997;144:1188–94. [2] Chung SY, Chiang YM. Microscale measurements of the electrical conductivity of doped LiFePO4. Electrochem Solid State Lett 2003;6:A278–81. [3] Prosini PP, Lisi M, Zane D, Pasquali M. Determination of the chemical diffusion coefficient of lithium in LiFePO4. Solid State Ionics 2002;148:45–51. [4] Meethong N, Huang HYS, Speakman SA, Carter WC, Chiang YM. Strain accommodation during phase transformations in olivine-based cathodes as a materials selection criterion for high-power rechargeable batteries. Adv Funct Mater 2007;17:1115–23. [5] Yamada A, Chung SC, Hinokuma K. Optimized LiFePO4 for lithium battery cathodes. J Electrochem Soc 2001;148:A224–9. [6] Yang JS, Xu JJ. Nonaqueous sol–gel synthesis of high-performance LiFePO4. Electrochem Solid State Lett 2004;7:A515–8. [7] Sides CR, Croce F, Young VY, Martin CR, Scrosati B. A high-rate, nanocomposite LiFePO4/carbon cathode. Electrochem Solid State Lett 2005;8:A484–7. [8] Yang SF, Zavalij PY, Whittingham MS. Hydrothermal synthesis of lithium iron phosphate cathodes. Electrochem Commun 2001;3:505–8. [9] Chen JJ, Whittingham MS. Hydrothermal synthesis of lithium iron phosphate. Electrochem Commun 2006;8:855–8.
1427
[10] Chen J, Wang S, Whittingham MS. Hydrothermal synthesis of cathode materials. J Power Sources 2007;174:442–8. [11] Chen JJ, Vacchio MJ, Wang SJ, Chernova N, Zavalij PY, Whittingham MS. The hydrothermal synthesis and characterization of olivines and related compounds for electrochemical applications. Solid State Ionics 2008;178:1676–93. [12] Doeff MM, Hu YQ, McLarnon F, Kostecki R. Effect of surface carbon structure on the electrochemical performance of LiFePO4. Electrochem Solid State Lett 2003;6:A207–9. [13] Huang XJ, Yan SJ, Zhao HY, Zhang L, Guo R, Chang CK, et al. Electrochemical performance of LiFePO4 nanorods obtained from hydrothermal process. Mater Charact 2010;61:720–5. [14] Teng F, Santhanagopalan S, Lemmens R, Geng XB, Patel P, Meng DD. In situ growth of LiFePO4 nanorod arrays under hydrothermal condition. Solid State Sci 2010;12:952–5. [15] Liao LX, Zuo PJ, Ma YL, Chen XQ, An YX, Gao YZ, et al. Effects of temperature on charge/discharge behaviors of LiFePO4 cathode for Li-ion batteries. Electrochim Acta 2012;60:269–73. [16] Wu L, Zhong SK, Liu JQ, Lv F, Wan K. High tap-density and high performance LiFePO4/C cathode material synthesized by the combined sol spray-drying and liquid nitrogen quenching method. Mater Lett 2012;89:32–5. [17] Konarova M, Taniguchi I. Preparation of carbon coated LiFePO4 by a combination of spray pyrolysis with planetary ball-milling followed by heat treatment and their electrochemical properties. Powder Technol 2009;191:111–6. [18] Konarova M, Taniguchi I. Synthesis of carbon-coated LiFePO4 nanoparticles with high rate performance in lithium secondary batteries. J Power Sources 2010;195:3661–7. [19] Zhong SK, Wu L, Zheng JC, Liu JQ. Preparation of high tap-density 9LiFePO4. Li3V2(PO4)3/C composite cathode material by spray drying and post-calcining method. Powder Technol 2012;219:45–8. [20] Liao XZ, Ma ZF, Gong Q, He YS, Pei L, Zeng LJ. Low-temperature performance of LiFePO4/C cathode in a quaternary carbonate-based electrolyte. Electrochem Commun 2008;10:691–4. [21] Yang KR, Deng ZH, Suo JS. Synthesis and characterization of LiFePO4 and LiFePO4/C cathode material from lithium carboxylic acid and Fe3+. J Power Sources 2012;201:274–9. [22] Zhang YC, Wang CY, Tang XD. Cycling degradation of an automotive LiFePO4 lithium-ion battery. J Power Sources 2011;196:1513–20. [23] Song HS, Cao Z, Chen X, Lu H, Jia M, Zhang ZA, et al. Capacity fade of LiFePO4/graphite cell at elevated temperature. J Solid State Electr 2013;17:599–605. [24] Kim JH, Woo SC, Park MS, Kim KJ, Yim T, Kim JS, et al. Capacity fading mechanism of LiFePO4-based lithium secondary batteries for stationary energy storage. J Power Sources 2013;229:190–7. [25] Cong LN, Gao XG, Ma SC, Guo X, Zeng YP, Tai LH, et al. Enhancement of electrochemical performance of Li[Li0.2Mn0.54Ni0.13Co0.13]O2 by surface modification with Li4Ti5O12. Electrochim Acta 2014;115:399–406. [26] Yi TF, Shu J, Zhu YR, Zhou AN, Zhu RS. Structure and electrochemical performance of Li4Ti5O12-coated LiMn1.4Ni0.4Cr0.2O4 spinel as 5 V materials. Electrochem Commun 2009;11:91–4. [27] Yao J, Shen C, Zhang P, Ma CA, Gregory DH, Wang L. Spinel-Li3.5+xTi5O12 coated LiMn2O4 with high surface Mn valence for an enhanced cycling performance at high temperature. Electrochem Commun 2013;31:92–5. [28] Zhu YR, Yi TF, Zhu RS, Zhou AN. Increased cycling stability of Li4Ti5O12-coated LiMn1.5Ni0.5O4 as cathode material for lithium-ion batteries. Ceram Int 2013;39:3087–94. [29] Yi TF, Shu J, Wang Y, Xue J, Meng J, Yue CB, et al. Effect of treated temperature on structure and performance of LiCoO2 coated by Li4Ti5O12. Surf Coat Technol 2011;205:3885–9.