Magnetron sputtering deposition of TiO₂ particles on polypropylene separators for lithium-ion batteries

Abstract

Magnetron sputtering deposition (MSD) of TiO₂ is applied on a porous polypropylene (PP) separator for lithium ion batteries. The surface morphology, contact angles, thermal properties, and electrolyte uptake of the modified separators are characterized and the electrochemical performances such as ionic conductivity, cycle performance, and high rate discharge capacity are investigated. The TiO₂ MSD-coated separators present suppressed thermal shrinkage which may lead to improved safety of the batteries, and show improvement in wettability with polar electrolytes and cell performance as compared to the bare PP separator.

---

1. Introduction

Lithium-ion batteries (LIBs) are widely used power sources for portable devices due to their high specific energy, long cycle life, low self-discharge rate and high operational voltage.¹ With the expansion of LIBs from portable devices to plug-in electric vehicles (PEVS) and plug-in hybrid vehicles (PHEVS),^2,3 which require high-energy density and high-power density, safety issues must be confronted.⁴ In a lithium-ion battery, the separator is considered a key component to secure the battery safety, whose function is to maintain the physical isolation between the cathode and anode to prevent electrical short circuits, while allowing the transport of ionic charge carriers.⁵

Porous polyolefin membranes such as polyethylene (PE) and polypropylene (PP) are the most dominant separators for LIBs because of their advantages such as chemical stability, mechanical strength, uniform pore structure and distribution and low cost.⁶ Although the PP, PE separators are reliable for portable applications, two major limitations should be overcome for vehicular storage utilization. First, shrinking or melting of polyolefin separators may happen at elevated temperature when the battery is used for a long time under a high current density, which often causes the occurrence of internal short circuit, and explosion of LIBs.⁷ Second, the hydrophobic surface and low surface energy of polyolefin separators result in their poor wettability with liquid polar electrolytes such as ethylene carbonate (EC), dimethyl carbonate (DEC) and propylene carbonate (PC).⁸ These features may decrease the electrolyte uptake of the separators, which results in increased internal resistance of the battery and even worse, leads to potential safety issues.^9,10

Recently, many efforts have been made to improve the poor wettability and thermal stability of polyolefin separators. One possible method is to introduce functional groups or hydrophilic polymers to polyolefin separators by surface grafting^1,11–13 and surface coating via phase inversion process.^14–16 These methods are very efficient in improving the wettability, but oftentimes cannot address the thermal stability issue. Furthermore, the fabrication process is very complicated and incompatible with that of the conventional separator production, and the uniformity of the grafted or coated layer is hard to control.¹⁷ Another approach is to coat inorganic particles including SiO₂, ZrO₂, Al₂O₃, on a porous organic support.^18–21 The inorganic–organic hybrid membranes show both improved thermal stability and wettability with organic electrolyte.²² In addition, the inorganic particles can prevent the penetration of Li dendrites.²³ However, in these cases, polyvinylidene fluoride-hexafluoropropylene (PVDF-HFP) as binders should be used to attach the inorganic particles on the separator surface, which leads to the membranes thicker than desired one and consequently a decrease in energy density of batteries. Moreover, excessive binders could also block up the pores in the separators which results in an increase in the internal resistance of batteries. To overcome these problems, an atomic layer deposition (ALD) method has been developed to coat Al₂O₃ layer on the surface of PP separators without binders.^24,25 Although it is promising in improving the thermal and wetting properties, the multiple self-limiting surface reaction and plasma pretreatment could reduce the pore size and sometimes decrease the mechanical properties of the separators.

Magnetron sputtering deposition (MSD) has been demonstrated as an effective surface modification method with a precise control of the nanoparticles shape, size and distribution by adjusting the process parameters (such as power, pressure, deposition time).²⁶ It is a well-established technology applied in fabrication of hard coatings,²⁷ superconducting film etc.²⁸ In this paper, MSD treatment is performed on the porous PP membranes (Celgard 2400) to form a thin uniform layer of TiO₂ tightly attaching to the PP substrate. Performance superiorities of these modified separators, as compared to bare PP separator, are evaluated in terms of surface morphology, thermal stability, electrolyte wettability, and ionic transport. Based on this characterization, advantageous effects of the modified separators on cell performance such as cycle performance, high rate discharge capacity are also investigated.

2. Experimental

2.1. Materials

Porous PP separators (Celgard 2400) were purchased from Celgard Company with a thickness of 25 μm and an average pore size of 43 nm. Titanium dioxide target with 99.9% purity was purchased from Lesker company.

2.2. Sample preparation

The porous PP separators were cleaned by ethanol and dried at room temperature before using. The thin TiO₂ layer was deposited by a direct current magnetron sputtering deposition technique using a magnetron sputtering setup (JPGF400B-G, China). The distance between the target and substrate was 5 cm and the input power was 90 W. The vacuum chamber was capable of achieving a base pressure of 5 × 10⁻³ Pa, and the working pressure for sputtering was 1 Pa. Four samples of TiO₂-coated PP separators prepared with 10, 20, 30, and 60 minutes of MSD treatment are designated as 10 MSD, 20 MSD, 30 MSD, and 60 MSD, respectively.

2.3. Characterization

The morphologies of the separators were examined using a field emission scanning electron microscope (FE-SEM, Hitachi S-4800). A thin layer of Pd/Au was sputtered on the samples surface before SEM observation. Energy Dispersive Spectrometer (EDS) linear analyses were carried out using an Inca X-max (Oxford Instruments). For cross-section structure observation, the modified separators were first washed in acetone and then immersed in a low-viscosity epoxy resin. The immersed separators were solidified at 80 °C with 6 h, then the solidified samples were fractured in liquid nitrogen and the section was collected for SEM observation. Contact angles against water on the bare PP and modified separators subjected to various sputtering time were measured by a contact angle goniometer (OCA 40 Micro). The contact angles were calculated through analyzing the shape of the drop and the average value of the measurement was reported. Thermo gravimetric (TG) analysis was performed in air at a heating rate of 20 °C min⁻¹ from 30 °C to 700 °C using a STA409PC thermal analyzer. The thermal shrinkage of the separators was determined by measuring the dimensional change after being subjected to heat treatment at various temperatures for 30 min.

The electrolyte uptake (T) of the separators was determined by immersing the separators in 1 M LiPF₆ (EC/DEC 1 : 1 in volume) electrolyte for 30 min and obtained by:

where W₁ and W₂ are the mass of the dry and wet separators respectively. The ionic conductivity of swollen separators was measured by electrochemical impedance spectroscopy (EIS) carried out on a Zennium CIMPS-1 electrochemical system. The samples were placed between two stainless steels, and spectra were obtained by sweeping in the 0.1 Hz to 100 kHz frequency range with a 5 mV AC amplitude at room temperature. The ionic conductivity (σ) of the separator was calculated by the eqn (2):where L, R and S are the thickness, resistance, and the effective area of the separator, respectively.

To evaluate the cell performances with the modified separators, a 2016-type coin cell was assembled using a Li-metal anode and LiCoO₂ cathode prepared by slurry coating a mixture of LiCoO₂, PVDF binder, carbon black at a weight ratio of 8 : 1 : 1 onto high purity aluminum foil. The electrodes and separators were dried under vacuum at 100 °C and 60 °C for at least 12 h, respectively. All batteries were assembled in a dry box filled with argon. In order to form a stable solid electrolyte interface on the surface of the anode, the first cycle of all test cells were charged and then discharged under a constant current mode at the 0.1C rate. The cycling tests were carried out in the voltage range of 3.0 V to 4.2 V at a constant current with a 0.5C rate. To investigate the discharge rate capacity retention, the cells were charged up to 4.2 V at 0.1C rate and discharged to 3.0 V at 0.1C in first 5 cycles, and then they were charged to 4.2 V at 0.2C rate and discharged to 3.0 V at the rates of 0.2, 0.5, 1, 2, 4, and 8C. The electrochemical stability window of the separators was performed on a working electrode of stain-steel and a reference electrode of lithium metal at a scan rate of 1.0 mV s⁻¹ between 3.0–6.0 V.

3. Results and discussion

Fig. 1 shows the SEM images of the bare and TiO₂ MSD-coated PP separators. The bare PP separator contained nonporous and porous regions (Fig. 1a), and the area of stripe-like narrow pores were about 20 223 nm² which was consistent with the previous report.²⁹ After 10, 20, 30 minutes MSD treatment (sample: 10 MSD, 20 MSD, and 30 MSD), the porous structure and morphology remained unchanged (Fig. 1b–d) and the edge of pores became much smoother. While, for the 60 MSD sample, some pores were blocked by the TiO₂ particles (Fig. 1e). The average pore area and the porosity of TiO₂ MSD-coated separators were listed in Table 1. The average pore area decreased dramatically with the deposition time, but the reduction of the porosity was less than 3.5%. The pore area was calculated from the SEM images which could only show the pores on the membrane surface. Therefore, the TiO₂ particles were most probably deposited on the separator surface (not in the free volumes of subsurface regions), otherwise the porosity would decrease more dramatically after deposition. Therefore, the reduction of porosity after deposition could be neglected. In the magnified SEM image of 30 MSD samples (Fig. 1f), the TiO₂ layer was composed of large amount of nanoparticles whose size was around 5 nm. Under the magnetron sputtering condition of our experiment, the TiO₂ nanoparticles were deposited at a low rate and aggregated in continuous and compact form, which resulted in the formation of a thin and condense layer. The presence of the TiO₂ layer was expected to improve the thermal properties and wettability of the separators.

Fig. 1 SEM images of surface morphologies of the bare PP separator (a), 10 MSD (b), 20 MSD (c), 30 MSD (d and f), 60 MSD (e).

Table 1 Average pore size, porosity, fraction of TiO₂ of bare PP separator and separators subjected to different sputtering time

Sample	Sputtering time (min)	Average pore area (nm^2)	Porosity (%)	Fraction of TiO2 (%)
Bare PP	0	20 223	41.0	0
10 MSD	10	12 691	40.8	1.17
20 MSD	20	9680	40.6	2.96
30 MSD	30	7416	40.2	4.90
60 MSD	60	2990	39.6	8.96

---

In order to measure the thickness of the deposited TiO₂ layer, SEM observations were carried out on the cross section of modified PP separators (Fig. 2). TiO₂ layer could be identified due to its higher electron density than polymeric component. The linear element scanning spectra (EDS) of titanium along the line in Fig. 2 verified the existence of TiO₂ layer and the thickness of the layers subjected to 30 min and 60 min deposition (sample 30 MSD and 60 MSD) were about 68 nm and 131 nm, respectively. The weight percentages of deposited TiO₂ on membrane surface determined by TG analysis were also measured and the results were listed in Table 1. When the time of deposition was not more than 30 min, the percentage of deposited TiO₂ is less than 5.0%. If the PP separator is subjected to 60 min deposition, the percentage is about 8.69%, which implied that the decrease in energy density of batteries should be considered.

Fig. 2 SEM cross section of modified separators: 30 MSD (a), 60 MSD (b) and EDS intensity along the line in (a) and (b).

TG analysis was performed to investigate the thermal stability of the modified and bare PP separators (Fig. 3). Here, we defined the decomposition temperature of a polymer at which it lost its total weight larger than 5%. The bare PP separator started to decompose at 258 °C through the breakdown of C–C and C–H bonds, forming gaseous compound (H₂O and CO₂) and lost its weight.³⁰ In contrast, the decomposition temperatures of sample 10 MSD, 20 MSD, 30 MSD, and 60 MSD were about 364 °C, 378 °C, 383 °C, and 368 °C, respectively. The improvement of thermal stability indicated that a thin layer of TiO₂ could significantly enhance the thermal stability of modified separators. The reason for this enhancement could be attributed to the fact that the tightly coated TiO₂ thin layer prevented the direct exposure of the polymeric component to oxygen (Fig. 1e). For the 60 MSD sample, the decomposition temperature was lower than the other modified ones. The reason might be contributed to the effect of catalytic oxidation which accelerated the decomposition of PP at elevated temperature,³¹ or the partly degradation of PP after long time deposition. When the TiO₂ layer is thin, the barrier effect plays the predominant role. With the increasing sputtering time, the effect of thermal catalytic oxidation of TiO₂ or degradation of PP became the dominant factor.

Fig. 3 TGA curves of bare PP separator and separators subjected to different sputtering time.

Thermal shrinkage of separators was an important property related to not only battery performance but also safety. Fig. 4 showed the thermal shrinkages of PP separators subjected to different MSD time. For the bare PP, the thermal shrinkages were about 11.6%, 20.0%, and 37.8% at the temperatures of 120 °C, 140 °C, and 160 °C, respectively. After MSD treatment, the thermal shrinkage was effectively suppressed in the whole measured temperature window. Increasing MSD time could benefit the suppression of thermal shrinkage. For instance, the thermal shrinkages of 30 MSD sample treated at the temperatures of 120 °C, 140 °C, and 160 °C were about 2.5%, 7.5%, and 20.0%, respectively. Although the PP substrate was prone to deform at elevated temperatures, the TiO₂ layer as the skeleton experienced the shrinkage tension which could reduce the structural deformation of the separator.

Fig. 4 Thermal shrinkages of separators at various temperature.

Fig. 5 presents the water contact angle and electrolyte uptake for the bare and TiO₂ MSD-coated separators. The water contact angle decreased obviously after the separators were subjected to MSD deposition. For example, the water contact angle for 60 MSD sample was about 59.3 ± 2.5° compared to 114.5 ± 2.2° for bare PP separator. Due to the polar nature of the deposited particles, the TiO₂ layer effectively changed the hydrophobic surface of PP separators to the hydrophilic one. Therefore, the TiO₂ coated separators presented a higher electrolyte uptake than the bare PP separator. For instance, compared to 144.3 ± 3.0% for bare PP separator, the electrolyte uptake for 30 MSD and 60 MSD samples were 175.3 ± 4.5% and 184.6 ± 4.2%, respectively. The existence of hydrophilic TiO₂ particles could effectively improve the wettability of the separators. For the bare PP separator, the electrolyte filled into the pores and the electrolyte uptake was determined by the porosity of separator. For TiO₂ coated separators, the electrolyte was not only trapped in the pores but also caught by the TiO₂ nanoparticles. The enhanced compatibility with the organic electrolyte compared to bare PP separator resulted in the increasing of ionic conductivity. Fig. 6 showed the Nyquist plots of the liquid electrolyte-soaked bare PP and TiO₂ MSD-coated separators. The ionic conductivities of bare PP separator and separators subjected to 10, 20, 30, and 60 min deposition could be calculated to be 0.07 × 10⁻³ S cm⁻¹, 0.10 × 10⁻³ S cm⁻¹, 0.12 × 10⁻³ S cm⁻¹, 0.48 × 10⁻³ S cm⁻¹, and 0.18 × 10⁻³ S cm⁻¹, respectively. The ionic conductivities increased obviously after deposition of TiO₂, which indicated that the internal resistance of the batteries with these TiO₂ MSD-coated separators could decrease and the performance could be improved. The lower conductivity of 60 MSD than that of 30 MSD sample could be contributed to the fact that some pores were blocked by the TiO₂ particles after 60 min deposition (Fig. 1e).

Fig. 5 Electrolyte uptake and water contact angles of separators subjected to different sputtering time.

Fig. 6 Impedance spectra of bare PP separator and separators subjected to different sputtering time.

To evaluate the potential application of the modified separator in LIBs, their electrochemical performances were investigated in terms of rate capability and cycle performance by cycling LiCoO₂/Li 2016-type coin cells at room temperature. This cell was charged up to 4.2 V and discharged to 3.0 V under constant current mode at charge/discharge rate of 0.5C/0.5C. Fig. 7 depicts the discharge capacities vs. cycle numbers of the test cells with bare PP separator and separators subjected to different sputtering time at C/2. The initial discharge capacities of the cells with MSD-coated separators were 128.7 mA h g⁻¹, 131.6 mA h g⁻¹, 132.5 mA h g⁻¹ and 130.2 mA h g⁻¹ for 10 MSD, 20 MSD, 30 MSD, and 60 MSD, respectively. Their capacities were higher than that of the bare PP (121 mA h g⁻¹) and their stabilities were also improved on the repeated charging and discharging process. Meanwhile, better cycle performances could be provided after 100 cycles. For example, the capacity retention of cells with the 30 and 60 MSD samples were 84.0% and 80.7% after 100 cycles compared to the 72.3% for that with bare PP separator. The better cycle performance of TiO₂ MSD-coated separators could be contributed to the consequence of their improved wettability and electrolyte uptake features.

Fig. 7 Discharge capacities vs. cycle numbers of the test cells with bare PP separator and separators subjected to different sputtering time at C/2.

Fig. 8 shows the rate capabilities of cells with different separators. Cells showed similar capacity retention at 0.1 and 0.2C rate. Afterwards, the higher the discharge rate was applied, the lower the capacity was. Although the discharge capacities for all cells with different separators decreased with the increase in C-rate, the cells with TiO₂ MSD-coated separators displayed a higher discharge capacity than that with the bare PP separator at 1C, 2C, 4C and 8C rate. The performance of the 30 MSD was noticeably enhanced at the high discharge rate (4C and 8C). The improved rate performance of the cells with the TiO₂ MSD-coated separators could be attributed to the low internal resistance and good compatibility with the organic electrolyte.

Fig. 8 Rate capacity tests for the cells with bare PP separator and separators subjected to different sputtering time.

Fig. 9 shows the electrochemical stability window of the electrolyte-soaked PP and MSD-coated separators. For the bare PP separator, a decomposition voltage around 4.7 V (vs. Li/Li⁺) was observed, which is consistent with the reported value.³² For 10 MSD sample, the decomposition voltage appeared at 5.3 V. No significant decomposition of carbonate electrolytes for 20 MSD, 30 MSD and 60 MSD samples took place below 6.0 V. In the whole measured voltage window, the carbonate electrolytes/MSD-coated separator systems showed improved electrochemical stability, which could be ascribed to the impurity absorption of TiO₂ nanoparticles in the electrolyte.^33,34

Fig. 9 Linear sweep voltammograms of the PP separator and separators subjected to different sputtering time at 1.0 mV s⁻¹ from 3.0–6 V.

4. Conclusions

In this work, magnetron sputtering deposition (MSD) approach was developed to modify the porous PP separators for LIBs. The thin MSD-coated TiO₂ layer with negligible effect on the porosity was tightly attached on the surface of PP separators and greatly suppressed thermal shrinkage of the separators, which was expected to partly solve the safety issue for the LIBs. The wettability of TiO₂ MSD-coated separator to the electrolytes was noticeably improved because of the hydrophilic nature of TiO₂ layer. It was also demonstrated that such modified separators presented better thermal stability, cell performances, and electrochemical stability compared to the bare PP separator. Thus, we believed that MSD is a potential way to fabricate inorganic/organic hybrid separators for high-power LIBs used in electric vehicles.

Acknowledgements

This work was financially supported by the Shanghai Leading Academic Discipline Project (No. B603) and the Program of Introducing Talents of Discipline to Universities (No. 111-2-04).

Notes and references

1. K. J. Kim, J. H. Kim, M. S. Park, H. K. Kwon, H. Kim and Y. J. Kim, J. Power Sources, 2012, 198, 298.
2. T. H. Cho, M. Tanaka, H. Ohnishi, Y. Kondo, M. Yoshikazu, T. Nakamura and T. Sakai, J. Power Sources, 2010, 195, 4272.
3. X. Huang, J. Power Sources, 2011, 196, 8125.
4. J. Hassoun, S. Panero, P. Reale and B. Scrosati, Adv. Mater., 2009, 21, 4807.
5. M. Xiong, H. Tang, Y. Wang, Y. Lin, M. Sun, Z. Yin and M. Pan, J. Power Sources, 2013, 241, 203.
6. J. L. Shi, L. F. Fang, H. Li, H. Zhang, B. K. Zhu and L. P. Zhu, J. Membr. Sci., 2013, 437, 160.
7. S. I. Tobishima and J. I. Yamaki, J. Power Sources, 1999, 81–82, 882.
8. J. Fang, A. Kelarakis, Y. W. Lin, C. Y. Kang, M. H. Yang, C. L. Cheng, Y. Wang, E. P. Giannelis and L. D. Tsai, Phys. Chem. Chem. Phys., 2011, 13, 14457.
9. K. Gao, X. Hu, C. Dai and T. Yi, J. Mater. Sci. Eng. B, 2006, 131, 100.
10. W. K. Shin and D. W. Kim, J. Power Sources, 2013, 226, 54.
11. J. Y. Lee, Y. M. Lee, B. Bhattacharya, Y. C. Nho and J. K. Park, Electrochim. Acta, 2009, 54, 4312.
12. S. J. Gwon, J. H. Choi, J. Y. Sohn, Y. M. Lim, Y. C. Nho and Y. E. Ihm, J. Ind. Eng. Chem., 2009, 15, 748.
13. S. J. Gwon, J. H. Choi, J. Y. Sohn, S. J. An, Y. E. Ihm and Y. C. Nho, Nucl. Instrum. Methods Phys. Res., Sect. B, 2008, 266, 3387.
14. Y. M. Lee, N. S. Choi, J. A. Lee, W. H. Seol, K. Y. Cho, H. Y. Jung, J. W. Kim and J. K. Park, J. Power Sources, 2005, 146, 431.
15. Y. Lee Min, J. W. Kim, N. S. Choi, J. Lee An, W. H. Seol and J. K. Park, J. Power Sources, 2005, 139, 235.
16. Z. Fu, H. Feng, C. Sun, X. Xiang, W. Wu, J. Luo, Q. Hu, A. Feng and W. Li, J. Solid State Electrochem., 2013, 1.
17. J. L. Shi, L. F. Fang, H. Li, Z. Y. Liang, B. K. Zhu and L. P. Zhu, J. Membr. Sci., 2013, 429, 355.
18. K. Prasanna, C.-S. Kim and C. W. Lee, Mater. Chem. Phys., 2014, 146, 545.
19. K. J. Kim, H. K. Kwon, M.-S. Park, T. Yim, J.-S. Yu and Y.-J. Kim, Phys. Chem. Chem. Phys., 2014, 16, 9337.
20. C. Shi, P. Zhang, L. Chen, P. Yang and J. Zhao, J. Power Sources, 2014, 270, 547.
21. T. Lee, W. K. Kim, Y. Lee, M. H. Ryou and Y. M. Lee, Macromol. Res., 2014, 22, 1190.
22. S. S. Zhang, K. Xu and T. R. Jow, J. Power Sources, 2005, 140, 361.
23. X. Huang, D. Bahroloomi and X. Xiao, J. Solid State Electrochem., 2013, 1.
24. Y. S. Jung, A. S. Cavanagh, L. Gedvilas, N. E. Widjonarko, I. D. Scott, S. H. Lee, G. H. Kim, S. M. George and A. C. Dillon, Adv. Energy Mater., 2012, 2, 1022.
25. H. Chen, Q. Lin, Q. Xu, Y. Yang, Z. Shao and Y. Wang, J. Membr. Sci., 2014, 458, 217.
26. J. Zuo, P. Keil and G. Grundmeier, Appl. Surf. Sci., 2012, 258, 7231.
27. X. Zhang, Y. Qiu, Z. Tan, J. Lin, A. Xu, Y. Zeng, J. J. Moore and J. Jiang, J. Alloys Compd., 2014, 617, 81.
28. Y. Zhang, F. Feng, K. Shi, H. Lu, S. Xiao, W. Wu, R. Huang, T. Qu, X. Wang, Z. Wang and Z. Han, Appl. Surf. Sci., 2013, 284, 150.
29. P. Arora and Z. Zhang, Chem. Rev., 2004, 104, 4419.
30. Y. H. Liao, X. P. Li, C. H. Fu, R. Xu, L. Zhou, C. L. Tan, S. J. Hu and W. S. Li, J. Power Sources, 2011, 196, 2115.
31. J. L. Falconer and K. A. Magrini-Bair, J. Catal., 1998, 179, 171.
32. Q. Xu, Q. Kong, Z. Liu, X. Wang, R. Liu, J. Zhang, L. Yue, Y. Duan and G. Cui, ACS Sustainable Chem. Eng., 2014, 2, 194.
33. H. R. Jung, D. H. Ju, W. J. Lee, X. Zhang and R. Kotek, Electrochim. Acta, 2009, 54, 3630.
34. Y. H. Liao, M. M. Rao, W. S. Li, L. T. Yang, B. K. Zhu, R. Xu and C. H. Fu, J. Membr. Sci., 2010, 352, 95.

---