Efficient electrospinning fabrication and the underlying formation mechanism of one-dimensional monoclinic Li2FeSiO4 nanofibers

Wenheng Zhang, Longwei Liang, Yan Ju, Yang Liu, Linrui Hou and Changzhou Yuan*
School of Material Science & Engineering, University of Jinan, Jinan, 250022, P. R. China. E-mail: mse_yuancz@ujn.edu.cn; ayuancz@163.com

Received 17th July 2019 , Accepted 14th September 2019

First published on 16th September 2019

The purposeful exploration of efficient methodologies towards synthesizing phase-pure multi-compositional nanostructures is an everlasting topic for material scientists. In this work, we first devise a smart strategy to fabricate high-purity Li2FeSiO4 nanofibers (LFSNFs) with a P21/n structure via controllable electrospinning and subsequent calcination in air and N2 atmospheres. Systematic physicochemical characterization confirms the significant roles of the regulation of polyvinyl pyrrolidone (PVP) content via pretreatment in air, and concentrations of inorganic salts in the formation of one-dimensional LFSNFs. Specifically, the excess PVP-derived carbon content will result in the carbon-thermal reduction of high-valence Fe species, and high concentrations of precursors will increase the diameter of the nanofibers, and even induce the conversion from nanofibers to nanobelts. Finally, the optimum synthetic parameters are convincingly proposed. More significantly, we strongly believe that the unusual formation mechanism of the LFSNFs will have enormous potential in enriching synthetic methods of polyoxyanion nano-architectures.

1. Introduction

More recently, a new class of polyoxyanion compounds, Li2MSiO4 (M = Fe, Mn, Co), especially Li2FeSiO4 (LFS), has attracted considerable attention as potential cathodes for lithium-ion batteries, due to the abundance and low cost of both Si and Fe, and their large capacity of about 332 mA h g−1.1,2 It has been reported that LFS has at least three different crystalline structures, namely the monoclinic structure with the space group P21/n,3 and two orthorhombic structures with space groups Pmn21 and Pmnb.4,5 As verified by first-principles calculations, the structure with the monoclinic space group is found to be the most stable polymorph with two diffusion channels for Li-ion diffusion, and the stability plays an important role in understanding the multi-electron process of LFS.4–8

Of especial note, complex polymorphism means that a complex synthesis process and harsh synthesis conditions are always required to obtain high-purity LFS. Although various methods, such as sol–gel,9 hydrothermal reactions10 and spray pyrolysis,11 have been widely exploited to obtain fine LFS powder, these approaches are commonly very time-consuming, even requiring several days for some cases.12,13 For instance, Yang et al.12 reported that LFS nanorods (NRs) of P21/n structure were synthesized by hydrothermal treatment at 200 °C for 6 days. Xu and co-workers13 revealed that hollow LFS spheres were prepared via a template-free hydrothermal method at 180 °C for 4 days. More importantly, the high-valence Fe species tend to be reduced into impure Fe in a N2 atmosphere.8–13 Therefore, it remains greatly challenging to explore simple yet efficient methodologies for preparing LFS with high purity and controllable micro-structures.

Besides, one-dimensional (1D) nano-architectures, including nanofibers (NFs), NRs and nanobelts (NBs), have attracted enormous interest, and have been extensively utilized in energy-related applications, owing to their unique merits.14,15 As is well known to all, the electrospinning technique is potentially straightforward and cost effective towards producing 1D NFs with diameters from less than 3 nm to over 1 mm.16–18 In the electrospinning process, the starting polymer solutions with several inorganic salts are pumped to a capillary, and in situ spun to fibers with Coulomb forces. As a result, inorganic NFs with high homogeneity can be obtained after calcination.

Herein, we, to the best of our knowledge, first devised the electrospinning fabrication method of monoclinic LFSNFs along with the subsequent fine calcination in air and N2 atmospheres in sequence. Detailed characterization demonstrated that the pretreatment of the NF-shape precursor in air guaranteed the formation of phase-pure LFS, avoiding the reduction of Fe3+ to metallic Fe, and that the suitable inorganic salt concentrations ensured the formation of the NF structure. Accordingly, the optimized synthesis conditions and underlying formation mechanism of 1D LFSNFs were reasonably put forward here.

2. Experimental

2.1 Synthesis of LFS@PVP samples

All the chemicals used in this work were analytical-grade reagents from Sinopharm Chemical Reagent Co., Ltd, and directly used without any further purification. Typically, 2 g of polyvinyl pyrrolidone (PVP, K88–K96, Mw = 1[thin space (1/6-em)]300[thin space (1/6-em)]000) was firstly dissolved in 18.0 mL of absolute ethanol, which was marked as solution A. Then, 0.204 g of LiAC·2H2O, 0.404 g of Fe(NO3)3·9H2O and 0.219 mL of tetraethyl orthosilicate (TEOS) were added into 10.0 mL of N,N-dimethylformamide (DMF) with a molar ratio of Li[thin space (1/6-em)]:[thin space (1/6-em)]Fe[thin space (1/6-em)]:[thin space (1/6-em)]Si = 2[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]1, which was designated as solution B. After vigorous stirring for 12 h, respectively, the two solutions were further well mixed under stirring. Finally, the starting solution was pumped into a plastic syringe with a stainless steel nozzle (0.50 mm in diameter) with a fixed distance of 15 cm from the collector. An electrical potential of 15 kV and a push speed of 0.06 mm min−1 were applied for the following electrospinning. The precursor was obtained after being dried at 60 °C, and denoted as LFS@PVP-1 for convenience. For comparison, three other precursor solutions were prepared with enhanced LiAC·2H2O, Fe(NO3)3·9H2O and TEOS contents up to two, three and four times those in solution B, and the resulting polymer fibers were marked as LFS@PVP-2, LFS@PVP-3 and LFS@PVP-4, respectively. As noted, the other synthetic parameters for the three are the same as those of LFS@PVP-1.

2.2 Synthesis of LFSNF products

The dried LFS@PVP-1 was annealed at 400 °C for 3 h in air with a heating ramp of 2 °C min−1. The intermediate was then annealed at 700 °C with a ramp rate of 2 °C min−1 for 10 h in N2. A black product was achieved, and designated as LFSNF-400-3. In order to explore the effect of pretreatment time on the formation of LFS fibers, the samples were named as LFSNF-400-1, LFSNF-400-2 and LFSNF-400-6, which were pretreated at 400 °C for 1 h, 2 h and 6 h in air, respectively, and the other conditions were kept the same as those for LFSNF-400-3. Additionally, two other samples were also prepared by annealing the intermediate in N2 at 700 °C for 5 and 8 h, respectively.

2.3 Material characterization

The X-ray diffraction (XRD) patterns of the resultant products were collected on a Rigaku Ultima IV (Cu Kα radiation, λ = 1.5046 Å, Japan) from 10 to 80° with a scanning speed of 5° min−1. Morphological features were characterized by field-emission scanning electron microscopy (FESEM, JEOL-6300F), transmission electron microscopy (TEM), scanning TEM (STEM) and high-resolution TEM (HRTEM, JEOL JEM-2010) with energy-dispersive X-ray spectroscopy (EDS). The diameter distribution diagram of the samples was measured via Nano Measurer software. The chemical states of the samples were characterized through X-ray photoelectron spectroscopy (XPS, Thermo ESCALAB 250Xi). Thermogravimetric (TG) analysis and differential scanning calorimetry (DSC) were performed on a NETZSCH STA449 F5 thermal analyzer. Raman spectra were obtained using a Renishaw inVia confocal Raman microspectroscope with 532 nm laser radiation at a laser power of 0.04 mW in the range of 100–3000 cm−1.

3. Results and discussion

3.1 Structural and physicochemical analysis of LFSNF-400-3

Herein, we have smartly developed an electrospinning method for purposeful fabrication of 1D LFSNFs, as schematically illustrated in Fig. 1. The electrospun LFS@PVP-1 is finely annealed in air to partially remove PVP. The LFSNF-400-3 sample is successfully obtained after subsequently annealing the intermediate in a N2 atmosphere. Fig. 2a shows a typical wide-angle XRD pattern of LFSNF-400-3. The sharp diffraction reflections indicate the good crystallinity of the sample. Owing to the minor difference in reflection positions between Pmn21 and P21/n, as indicated by the red and blue vertical lines in Fig. 2a, the typical reflections in regions b and c are enlarged, and profiled in detail in Fig. 2b and c, respectively. Obviously, the P21/n structure presents three reflections in the 2 theta range from 20 to 23.5°, while only two reflections are visible in the same region for the Pmn21 structure (Fig. 2b), and the absence of the reflection located at 31.6° is also observed in Pmn21 (Fig. 2b). The detected characteristic reflections of LFSNF-400-3 at 20.78° (Fig. 2b) and 31.62° (Fig. 2c) fully verify its typical P21/n structure, where all the LiO4, FeO4 and SiO4 tetrahedra are connected by corner-sharing and half of the three turn to point in the opposite direction, parallel to the b-axis, as demonstrated in Fig. 2d.12,19,20 More importantly, no diffraction signals for impurities can be observed here, indicating the formation of pure-phase LFS. Additionally, when the calcination temperature in N2 is further increased up to 900 °C, the monoclinic structure can still be retained (Fig. S1a, ESI).
image file: c9ce01112a-f1.tif
Fig. 1 Schematic illustration of electrospinning synthesis of LFSNF-400-3.

image file: c9ce01112a-f2.tif
Fig. 2 (a–c) XRD patterns of LFSNF-400-3, (d) P21/n structure diagram of LFS, and XPS spectra of high-resolution elemental (e) Fe 2p and (f) Si 2p for LFSNF-400-3.

XPS analysis is next performed to investigate the elemental compositions and oxidation states of LFSNF-400-3. As detected from the survey spectrum (Fig. S1b, ESI), the elemental Li, Fe, Si, O and C are all identified in LFSNF-400-3, where the C signal may be from the inevitable carbon pollution.21 A typical Fe 2p spectrum and the corresponding fitted profiles of LFSNF-400-3 are depicted in Fig. 2e. Evidently, the two peaks of Fe 2p are centered at binding energies of ∼710.6 and ∼724 eV, respectively, and each part consists of a satellite peak at 713.8 and 730.2 eV, respectively, both of which are characteristic of Fe2+.13,22–24 Particularly, the Fe 2p3/2 peak approximates to ∼710.6 eV (Fe2+), but away from ∼711.08 eV (Fe3+), indicating the main Fe2+ state in the LFSNF-400-3.24 In addition, the Si 2p spectrum (Fig. 2f) corresponds to Si4+ in polysiloxane, revealing the existence of the [SiO4] orthosilicate structure in LFSNF-400-3.23

Representative morphologies of LFS@PVP-1 and LFSNF-400-3 are characterized by FESEM. As displayed in Fig. 3a, LFS@PVP-1 presents numerous NFs of tens of micrometers in length and ∼160–320 nm in diameter (Fig. S2a, ESI), and is randomly oriented with a smooth surface (Fig. 3b). LFSNF-400-3, as shown in Fig. 3c, completely inherits the NF morphology of the precursor LFS@PVP-1. Nevertheless, the diameter of LFSNF-400-3 is greatly reduced to ∼60–130 nm (Fig. S2b, ESI), and the surface becomes somewhat rough (Fig. 3d), as compared to LFS@PVP-1, which should result from the decomposition of organic components and formation of new phases at high temperature.25,26 Further TEM observation (Fig. 3e) reveals that LFSNF-400-3 is made up of numerous primary grains of ∼50 nm in size (the insets in Fig. 3e). The HRTEM image (Fig. 3f) displays nanoscale particles with distinct lattice fringes with a spacing of ∼0.305 nm, corresponding to the (112) plane of LFS. A typical STEM image and corresponding EDS mappings (Fig. 3g) exhibit homogeneous spatial distributions of the elemental Si, O and Fe throughout LFSNF-400-3.

image file: c9ce01112a-f3.tif
Fig. 3 FESEM images of (a and b) LFS@PVP-1 and (c and d) LFSNF-400-3, and (e and f) HRTEM, (g) STEM and corresponding EDS elemental Si, Fe and O mapping images of LFSNF-400-3. The insets in panels (e and f) are the magnified regions as indicated.

3.2 Formation mechanism of LFSNFs

In order to investigate the intrinsic formation mechanism of the 1D LFSNF sample, several parallel experiments were systematically performed. Interestingly, without pretreatment in air or pretreatment in N2 in the first step (Fig. 1), a simple mixture of Li2SiO3 and Fe is unexpectedly obtained (Fig. S3a and b, ESI), rather than single-phase LFSNFs, which indicates that reduction of Fe3+ occurs. It is worth mentioning that the absence of pretreatment in air unexpectedly leads to completely different products here. As a result, one question is reasonably raised: what happens in the calcination process, especially in the initial pretreatment process.

On high-temperature calcination, the thermal decomposition of LiAC·2H2O, Fe(NO3)3·9H2O and PVP generally takes place, and leads to the formation of Li2O, Fe2O3 (ref. 27) and C.28 The weak reflection signals of Li2O and Fe2O3 are detected in the intermediate due to their low crystallinities and the partially undecomposed PVP (Fig. S4a, ESI). The Raman spectra of the intermediate (Fig. S4b, ESI) show a strong peak at around 1324 cm−1 (D-band), which is due to the activation of the symmetry forbidden set of modes by the defects in the sp2 electronic network.29,30 The characteristic peak of the G-band is not detected here, suggesting that the carbon material in the intermediate has a low degree of order. In addition, SiO2 is formed by hydrolysis and condensation reactions of TEOS.31 Accordingly, LFS can be rationally derived as described in eqn (1):

4Li2O + 4SiO2 + 2Fe2O3 + C → 4Li2FeSiO4 + CO2 (1)

Therefore, the PVP-derived carbon, as a reducer, has an irreplaceable effect during the above reaction. But as especially noted, if the carbon content was excessive, it will lead to the appearance of impurity Fe,1,32 as expressed by the following equation (eqn (2)):

2Fe2O3 + 3C → 4Fe + 3CO2 (2)

Furthermore, it is worth mentioning that the reduction rate of Fe3+ to Fe is so high that reaction (2) commonly happens along with reaction (1).32 Thus, it is highly necessary and significant to control the specific carbon content before the subsequent calcination process in a N2 atmosphere in order to obtain pure phase LFS and avoid the formation of metallic Fe. As a consequence, the regulation of PVP content in the intermediate should be initially conducted by annealing in air. Accordingly, the annealing temperature during the pretreatment in air is further determined based on the TG-DSC data (Fig. 4). As illustrated in Fig. 4a, the dehydration reaction is first detected at 100 °C for LFS@PVP-1, corresponding to a small endothermic peak. Then, a large mass reduction process, corresponding to the decomposition of PVP and inorganic salts, takes place within the temperature range of 300–450 °C.1 Finally, a large exothermic peak is detected at ∼550 °C with a small amount of mass loss, which is considered to be the main reaction. Fig. 4b further plots the TG-DSC data of K88–K96 in air. Two weight loss rates are evident in K88–K96 from 270 to 450 °C. The slow weight loss rate, corresponding to the low-molecular-weight component in K88–K96, can be observed in the case of <400 °C, while the rapid mass loss above 400 °C is related to the high-molecular-weight counterpart in K88–K96. Considering the thermal analysis above, the pretreatment temperature in air is finally determined to be 400 °C. Besides the annealing temperature, another parameter, i.e., the heat preservation, is also of significance for adjusting the PVP content in the intermediate. Just with annealing at 400 °C in air for 1 or 2 h, the impurities of both Fe and Li2SiO3 can still be observed, and the corresponding reflections of the two become weaker with prolonged heat preservation, as observed from the XRD patterns of LFSNF-400-1 and LFSNF-400-2 (Fig. 5). The signals for the impurities cannot be detected in LFSNF-400-3 (Fig. 2a). Notably, when the heat preservation time is extended to 6 h, the impurity of Li2SiO3 is detected again, and the reflections of Fe2O3 are also observed (Fig. 5, LFSNF-400-6). In addition, when the calcination time in a N2 atmosphere is shortened, such as 5 h, the diffraction signals related to the SiO2 impurity are still observed in the final product (Fig. S5a, ESI), which signifies an incomplete chemical reaction (reaction (1)). With the time up to 8 h, the diffraction signal for the impurity almost disappears (Fig. S5b, ESI). On further extending the time to 10 h, which is the case of LFSNF-400-3 (Fig. 2a), high phase-purity LFS with a P21/n structure can be obtained, as discussed above. Therefore, we believe that successful formation of pure phase LFSNFs can be achieved just by annealing LFS@PVP-1 at 400 °C in air for 3 h, and further at 700 °C in N2 for 10 h.

image file: c9ce01112a-f4.tif
Fig. 4 TG-DSC curves of (a) LFS@PVP-1 in N2 and (b) K88–K96 in O2 at a ramp rate of 5 °C min−1.

image file: c9ce01112a-f5.tif
Fig. 5 XRD patterns of LFSNF-400-1, LFSNF-400-2 and LFSNF-400-6 as indicated.

Further investigations, as shown in Fig. 6, show that the microscopic morphology of LFS@PVP varies greatly with the concentration of the precursor solution. As for LFS@PVP-2 (Fig. 6a and b), the product is still representatively shaped as NFs, but the diameter obviously increases when compared to that for LFS@PVP-1 (Fig. 3a and b). Interestingly, with the concentration of the precursor solution further increasing, i.e., LFS@PVP-3 (Fig. 6c and d) and LFS@PVP-4 (Fig. 6e and f), the diameter of the NFs further increases along with the appearance of more and more NBs. This interesting phenomenon can be explained by the following two equations:33,34

image file: c9ce01112a-t1.tif(3)
image file: c9ce01112a-t2.tif(4)
where R is the terminal jet radius, c is the concentration, η is the viscosity, γ is the surface tension, Q is the flow rate and I is the electric current. For our case, the electrospinning conditions are all the same just with the exception of the c value. As a result, the R of NFs is directly proportional to the c in our case. However, when the R value increases step by step with the concentration of the precursor solution, the solvent evaporation rate of the inner and outer layers is different, resulting in the collapse of the NF structure and expected formation of the NBs.35 The results suggest that the concentration of precursors plays a critically important role in the formation of the NFs, which allows their growth in a controlled manner. More interestingly, as shown in Fig. 7, all the products obtained via pretreating (400 °C for 3 h in air) and subsequently annealing (700 °C for 10 h in N2) the LFS@PVP-2, LFS@PVP-3 and LFS@PVP-4 samples (Fig. 6) retain their typical monoclinic LFS structure.

image file: c9ce01112a-f6.tif
Fig. 6 FESEM images of (a and b) LFS@PVP-2, (c and d) LFS@PVP-3, and (e and f) LFS@PVP-4.

image file: c9ce01112a-f7.tif
Fig. 7 XRD patterns of the products obtained via pretreating (400 °C for 3 h in air) and subsequently annealing (700 °C for 10 h in N2) (a) LFS@PVP-2, (b) LFS@PVP-3 and (c) LFS@PVP-4.

4. Conclusions

In summary, herein, we first explored a simple yet scalable synthetic strategy, involving electrospinning and subsequent calcination in an air/N2 atmosphere in sequence, to efficiently prepare pure phase LFSNFs with a P21/n structure. Systematic investigations demonstrate that the annealing temperature and heat preservation in an air/N2 atmosphere guarantee the formation of phase-pure LFS, and the concentration of the precursor solution ensures the formation of 1D NF structure. Accordingly, the appropriate synthetic conditions were proposed, and the involved intrinsic formation mechanism of the LFSNFs was rationally elucidated. More importantly, the novel methodology we devised here will have huge potential in enriching synthetic strategies of polyoxyanion nano-structures for energy-related applications.

Conflicts of interest

There are no conflicts to declare.


The authors acknowledge the financial support from the National Natural Science Foundation of China (No. 51772127 and 51772131), Taishan Scholars (No. ts201712050), Major Program of Shandong Province Natural Science Foundation (ZR2018ZB0317), Natural Science Doctoral Foundation of Shandong Province (ZR2018BEM018) and Collaborative Innovation Center of Technology and Equipment for Biological Diagnosis and Therapy in Universities of Shandong.


  1. Y. Fujita, H. Iwase, K. Shida, J. Liao, T. Fukui and M. Matsuda, J. Power Sources, 2017, 361, 115–121 CrossRef CAS.
  2. R. Dominko, M. Bele, M. Gaberscek, A. Meden, M. Remskar and J. Jamnik, Electrochem. Commun., 2006, 8, 217–222 CrossRef CAS.
  3. S.-I. Nishimura, S. Hayase, R. Kanno, M. Yashima, N. Nakayama and A. Yamada, J. Am. Chem. Soc., 2008, 130, 13212–13213 CrossRef CAS PubMed.
  4. G. Mali, C. Sirisopanaporn, C. Masquelier, D. Hanzel and R. Dominko, Chem. Mater., 2011, 23, 2735–2744 CrossRef CAS.
  5. C. Sirisopanaporn, A. Boulineau, D. Hanzel, R. Dominko, B. Budic, A. R. Armstrong, P. G. Bruce and C. Masquelier, Inorg. Chem., 2010, 49, 7446–7451 CrossRef CAS PubMed.
  6. A. Nyten, S. Kamali, L. Haggstrom, T. Gustafsson and J. O. Thomas, J. Mater. Chem., 2006, 16, 2266–2272 RSC.
  7. A. Saracibar, A. Van der Ven and M. E. Arroyo-de Dompablo, Chem. Mater., 2012, 24, 495–503 CrossRef CAS.
  8. R. B. Araujo, R. H. Scheicher, J. S. de Almeida, A. F. da Silva and R. Ahuja, Solid State Ionics, 2013, 247, 8–14 CrossRef.
  9. W. Wang, H. Liang, L. Zhang, S. V. Savilov, J. Ni and L. Li, Nano Res., 2017, 10, 229–237 CrossRef CAS.
  10. C.-H. Hsu, Y.-W. Shen, L.-H. Chien and P.-L. Kuo, J. Nanopart. Res., 2015, 17, 54 CrossRef.
  11. Y. Fujita, T. Hira, K. Shida, M. Tsushida, J. Liao and M. Matsuda, Ceram. Int., 2018, 44, 11211–11217 CrossRef CAS.
  12. J. Yang, L. Hu, J. Zheng, D. He, L. Tian, S. Mu and F. Pan, J. Mater. Chem. A, 2015, 3, 9601–9608 RSC.
  13. Y. Xu, W. Shen, A. Zhang, H. Liu and Z. Ma, J. Mater. Chem. A, 2014, 2, 12982–12990 RSC.
  14. D. Li, Y. Mao, H.-T.-H. Yong, X. Fan, L. Gou, M. Tian, K. Zhao, R. Sun, W. Zhang and G. Chen, Nanoscale, 2016, 8, 12202–12214 RSC.
  15. J. F. Sun, L. Z. Guo, X. Sun, J. Y. Zhang, L. R. Hou, L. Li, S. H. Yang and C. Z. Yuan, Batteries Supercaps, 2019, 2, 1–23 CrossRef.
  16. K. Bachtin, D. Kramer, V. S. K. Chakravadhanula, X. Mu, V. Trouillet, M. Kaus, S. Indris, H. Ehrenberg and C. Roth, J. Power Sources, 2018, 396, 386–394 CrossRef CAS.
  17. L. W. Liang, W. H. Zhang, D. K. Denis, J. Y. Zhang, L. R. Hou, Y. Liu and C. Z. Yuan, J. Mater. Chem. A, 2019, 7, 11915–11927 RSC.
  18. H. Niu, J. Zhang, Z. Xie, X. Wang and T. Lin, Carbon, 2011, 49, 2380–2388 CrossRef CAS.
  19. C. Sirisopannaporn, R. Dominko, C. Masquenlier, A. R. Armstrong, G. Mali and P. G. Bruce, J. Mater. Chem., 2011, 21, 17823–17831 RSC.
  20. Y. Zeng, H.-C. Chiu, M. Rasool, N. Brodusch, R. Gauvin, D.-T. Jiang, D. H. Ryan, K. Zaghib and G. P. Demopoulos, Chem. Eng. J., 2019, 359, 1592–1602 CrossRef CAS.
  21. X. Wang, W. Li, X. Wang, J. Zhang, L. Sun, C. Gao, J. Shang, Y. Hu and Q. Zhu, RSC Adv., 2017, 7, 50753–50759 RSC.
  22. J. Yang, X. Kang, L. Hu, X. Gong and S. Mu, J. Mater. Chem. A, 2014, 2, 6870–6878 RSC.
  23. L. Yi, G. Wang, Y. Bai, M. Liu, X. Wang, M. Liu and X. Wang, J. Alloys Compd., 2017, 721, 683–690 CrossRef CAS.
  24. X. Lu, H. Wei, H.-C. Chiu, R. Gauvin, P. Hovington, A. Guerfi, K. Zaghib and G. P. Demopoulos, Sci. Rep., 2015, 5, 8599 CrossRef CAS PubMed.
  25. H. Hou, C. Dong, L. Wang, F. Gao, G. Wei, J. Zheng, X. Cheng, B. Tang and W. Yang, CrystEngComm, 2013, 15, 2002–2008 RSC.
  26. H. Hou, L. Wang, F. Gao, G. Wei, B. Tang, W. Yang and T. Wu, J. Am. Chem. Soc., 2014, 136, 16716–16719 CrossRef CAS PubMed.
  27. L. Yi, G. Wang, Y. Bai, M. Liu, X. Wang, M. Liu and X. Wang, J. Alloys Compd., 2017, 721, 683–690 CrossRef CAS.
  28. R. Hongtong, P. Thanwisai, R. Yensano, J. Nash, S. Srilomsak and N. Meethong, J. Alloys Compd., 2019, 804, 339–347 CrossRef CAS.
  29. Q. Liang, Y. Zheng, C. Du, Y. Luo, J. Zhao, H. Ren, J. Xu and Q. Yan, ACS Nano, 2018, 12, 12902–12911 CrossRef CAS PubMed.
  30. Y. Guo, X. Zeng, Y. Zhang, Z. Dai, H. Fan, Y. Huang, W. Zhang, H. Zhang, J. Lu, F. Huo and Q. Yan, ACS Appl. Mater. Interfaces, 2017, 9, 17172–17177 CrossRef CAS PubMed.
  31. R. A. Assink and B. D. Kay, J. Non-Cryst. Solids, 1988, 99, 359–370 CrossRef CAS.
  32. B. Shao and I. Taniguchi, J. Power Sources, 2012, 199, 278–286 CrossRef CAS.
  33. Y. M. Shin, M. M. Hohman, M. P. Brenner and G. C. Rutledeg, Polymer, 2001, 42, 9955–9967 CrossRef CAS.
  34. S. V. Fridrikh, J. H. Yu, M. P. Brenner and G. C. Rutledeg, Phys. Rev. Lett., 2003, 90, 144502 CrossRef PubMed.
  35. H. Liu, W. Yang, L. Wang, H. Hou and F. Gao, CrystEngComm, 2017, 19, 6252–6258 RSC.


Electronic supplementary information (ESI) available. See DOI: 10.1039/c9ce01112a
Theses authors contributed equally to this work.

This journal is © The Royal Society of Chemistry 2019