Polythiophene Coated Aromatic Polyimide Enabled Ultrafast and Sustainable Lithium Ion Batteries†‡ Cite This: J

Journal of Materials Chemistry A

COMMUNICATION View Article Online View Journal | View Issue

Polythiophene coated aromatic polyimide enabled ultrafast and sustainable lithium ion batteries†‡ Cite this: J. Mater. Chem. A,2017,5, 24083 Hailong Lyu,ab Jiurong Liu, *a Shannon Mahurin,b Sheng Dai, bc Zhanhu Guo d b Received 7th September 2017 and Xiao-Guang Sun * Accepted 30th October 2017

DOI: 10.1039/c7ta07893e

rsc.li/materials-a

Organic composite electrode materials based on an aromatic poly- 1. Introduction imide (PI) and electron conductive polythiophene (PT) have been prepared by a facile in situ chemical oxidation polymerization method. Although conventional lithium-ion batteries (LIBs) based on The common aromatic structure possessed by both electroactive PI inorganic cathodes have been utilized in a wide range of and electron conductive PT allows intimate contacts, resulting in applications,1–6 they still have several issues restricting their conductive polymeric composites with highly reversible redox reac- forward evolvement, such as safety, power density and tions and good structural stability. It has been demonstrated that the PI sustainability.7 Therefore, efforts have been directed to look for composite material with 30 wt% PT coating (PI30PT) has the optimal alternative, greener and naturally abundant electrode materials combination of good electronic conductivity and fast lithium reaction for LIBs.8,9 As a result, renewable organic electrodes, such as kinetics. The synergistic effect between PI and PT enables a high organosulfur compounds,10,11 organic radical polymers,12–16 reversible capacity of 216.8 mA h g 1 at a current rate of C/10, as well conducting polymers17,18 and organic carbonyl compounds,19–22 as a high-rate cycling stability, that is, a high capacity of 89.6 mA h g 1 have been intensively investigated.23 These organic electrode at a high current rate of 20C with a capacity retention of 94% after materials have some unique properties that are generally not 1000 cycles. The elaborate combination of the high electronic available in inorganic cathode materials. For instance, safer and conductivity of the PT coating and the fabulous redox reaction more exible electrodes can be easily achieved due to the light reversibility of the PI matrix offers an economic way to prepare high weight, exibility and chemical tunability of organic electrode performance lithium ion batteries for sustainable energy storage materials.3,24 More importantly, without toxic heavy metal

Published on 31 October 2017. Downloaded by University of Tennessee at Knoxville 26/12/2017 23:23:53. applications. elements they can be easily recycled as paper with a minimal environmental footprint.25 Nevertheless, serious obstacles still exist for the practical application of organic electrodes, such as poor rate capability due to their low electronic conductivity and

aKey Laboratory for Liquid-Solid Structural Evolution and Processing of Materials, rapid capacity fading during cycling due to the dissolution of 26,27 Ministry of Education, School of Materials Science and Engineering, Shandong active organic cathodes. University, Jinan, Shandong 250061, P. R. China. E-mail: [email protected]; Tel: Among different organic electrodes, aromatic polyimides are +86-531-88390236 very promising candidates with a theoretical capacity b Chemical Sciences Division, Oak Ridge National Laboratory, One Bethel Valley Road, approaching 400 mA h g 1 and a working voltage of around 2.5 Oak Ridge, TN 37831, USA. E-mail: [email protected]; Tel: +1-865-241-8822 V vs. Li/Li+.19 During the discharge process (lithium intake), cDepartment of Chemistry, University of Tennessee, Knoxville, TN 37996, USA dIntegrated Composites Laboratory (ICL), Department of Chemical & Biomolecular aromatic polyimides can stepwise accept two electrons, result- Engineering, University of Tennessee, Knoxville, TN 37996, USA ing in the formation of a delocalized radical anion and dianion 3 † This manuscript has been authored by UT-Battelle, LLC under contract no. (Scheme 1). Although two more electrons can be accepted DE-AC05-00OR22725 with the U.S. Department of Energy. The United States during the discharge process, a lower redox potential coupled Government retains and the publisher, by accepting the article for publication, with poor structural stability usually prevents such an effort.28,29 acknowledges that the United States Government retains a non-exclusive, Different from lithium intercalation in inorganic cathodes, the paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government lithium storage mechanism in aromatic polyimides is a simple 8 purposes. The Department of Energy will provide public access to these results redox reaction, facilitating fast lithium reaction kinetics. To of federally sponsored research in accordance with the DOE Public Access Plan overcome the intrinsic electrical insulation of aromatic poly- (http://energy.gov/downloads/doe-public-access-plan). imides and obtain high rate performance, expensive carbon ‡ Electronic supplementary information (ESI) available. See DOI: additives such as graphene or carbon nanotubes have been used 10.1039/c7ta07893e

This journal is © The Royal Society of Chemistry 2017 J. Mater. Chem. A,2017,5,24083–24090 | 24083 View Article Online Journal of Materials Chemistry A Communication

Scheme 1 Electrochemical redox reactions of PI.

30–32 for electrode fabrication. For example, a polyimide deriva- hexauorophosphate (LiPF6) were obtained from BASF. All the tive containing a single walled carbon nanotube network chemicals and solvents were used directly without further (PMAQ–SWNT) exhibited a high capacity of 190 mA h g 1 at treatment. a current rate of 0.1C, as well as good rate performance.33 Alternatively, a composite electrode of a polyimide in situ 2.2. Synthesis of N,N0-diamino-1,4,5,8- polymerized on a three-dimensional graphene network (3D- naphthalenetetracarboxylic bisimide (DANTCBI) RGO/PI) showed a high capacity of 175 mA h g 1 at a current As the intermediate, DANTCBI was synthesized via the substi- rate of 0.1C and a low capacity of 40 mA h g 1 at a high current tution reaction between NTCDA and hydrazine. NTCDA (10 g) rate of 5C.34 Unfortunately, large amounts of conductive carbon and ethanol (200 ml) were added into an ice-cooled ask and additives were used in those composites, i.e. 30 wt% of SWNTs stirred to achieve a homogeneous solution under a nitrogen in PMAQ–SWNT and 20 wt% of 3D-RGO in 3D-RGO/PI, which atmosphere. Hydrazine hydrate (20 ml) was then added drop- limited the loading of the active polyimide materials in the wise into the cold reaction mixture. Aer stirring for 1 h, the cathodes and signicantly increased the cost of the electrodes. mixture was heated to reux for 1 h. The resulting dark yellow Besides conductive carbons, conducting polymers are solid was collected by ltration and dried in a vacuum to obtain attractive alternatives to enhance the electron conductivity of 10.6 g of product in 96% yield. 1H NMR (DMSO-d , 400 MHz): polymer composites due to their easy synthesis, high electron 6 d ¼ 5.86 ppm (s, 4H) and d ¼ 8.68 ppm (s, 4H). conductivity, environmental stability, cost effectiveness and unique electrochemical redox properties.35,36 Among different 2.3. Preparation of PI@PT composites conducting polymers, unmodied polythiophenes are insoluble in most organic solvents, which makes them an ideal candidate The precursor PI was synthesized by a simple condensation for the modication of the electronic conductivity of polymer polymerization method. Equimolar NTCDA and DANTCBI were cathodes.37 Herein, a novel polymer composite based on dissolved in warm 4-chlorophenol solvent, followed by heating a conductive polythiophene coated aromatic polyimide (PI@PT) to reux under nitrogen with stirring for 6 h. The product was is prepared by a facile in situ chemical oxidation polymerization ltered, thoroughly washed with methanol, and nally dried at approach. The effect of PT contents on the electrochemical 120 C under vacuum for 12 h. The pristine PI was obtained by Published on 31 October 2017. Downloaded by University of Tennessee at Knoxville 26/12/2017 23:23:53. performance of the PI@PT composite is investigated and opti- heat treatment at 350 C under nitrogen for 8 h. The PI@PT mized. Besides providing an outstanding electronic conduc- composites were synthesized by a typical in situ chemical tivity for the polymer composite, the PT coating also acts as an oxidation polymerization approach. Well-ground PI (1.8 g) and

ionic adsorbent and protective shell for the composite material FeCl3 (0.8 g) were uniformly dispersed in CHCl3 by sonication because of its good electrochemical stability.38,39 Overall, the and stirring for 1 h, and then a solution of thiophene (0.19 ml,  PI@PT composite demonstrates high capacity, long-term equal to 10 wt% in the nal product) and CHCl3 (30 ml) was cycling stability and good rate performance in rechargeable added slowly. The reaction mixture was stirred for 10 h at 0 C LIBs. under nitrogen. The product was washed several times with methanol and collected by ltration. Finally, a red-brown powder was obtained by drying at 80 C under vacuum and 2. Experimental denoted as PI10PT. Similarly, the amount of thiophene was 2.1. Materials increased to 30 wt% and 50 wt%, while maintaining the weight ratio of thiophene to FeCl as 1 : 4, to obtain the products of 1,4,5,8-Naphthalenetetracarboxylic dianhydride (NTCDA, 3 PI30PT and PI50PT, respectively. As a baseline, pure PT was also $98%), hydrazine hydrate (reagent grade, N H 50–60%), 2 4 synthesized following the same procedure without PI. thiophene (C4H4S, $99%) and iron(III) chloride anhydrous (FeCl , 97%) were purchased from Sigma-Aldrich. 4-Chlor- 3 2.4. Characterization ophenol (ClC6H4OH, 99%) was purchased from Alfa Aesar. 1 Chloroform (CHCl3, 99.8%) was purchased from BDH. H nuclear magnetic resonance (NMR) spectra were obtained on

Conductive carbon black (Super C45) was purchased from a Bruker Advance 400 MHz spectrometer using DMSO-d6 as the TIMCAL. Battery grade ethylene carbonate (EC), diethyl solvent. The chemical structures of the polymer products were carbonate (DEC), dimethyl carbonate (DMC), and lithium characterized by Fourier transform infrared (FTIR)

24084 | J. Mater. Chem. A,2017,5,24083–24090 This journal is © The Royal Society of Chemistry 2017 View Article Online Communication Journal of Materials Chemistry A

spectroscopy. The morphologies and microstructures of the increasing the amount of PT, the color of the nal product samples were observed on a Hitachi HD-2000 scanning trans- becomes darker, as shown in Fig. 1a–e, suggesting the mission electron microscope (STEM) operating at 200 kV. The successful coating of PI by PT. Fig. 1f shows the FTIR spectra of elemental compositions of the polymer composites were pristine PT and PI, and their composites, PI10PT, PI30PT and analyzed by energy dispersive X-ray spectroscopy (EDS) on PI50PT. For the pristine PT, the strongest absorption peak at 1 a SEM instrument with an EDAX accessory. 782 cm is attributed to the Cb–H out-of-plane stretching 40 vibration of the thiophene ring, indicating the Ca–Ca 39 – 2.5. Electrochemical measurements connection of the thiophene rings in PT. The Ca Ca connection is crucial to achieve high electron conductivity and good Polymer composite electrodes were fabricated by casting well electrochemical performance of the composites, as it has been homogenized slurries of active material, C45, and poly- proved that the Ca–Ca connected polythiophene exhibited the  vinylidene uoride (PVDF) binder with a weight ratio of 6 : 3 : 1 highest electron conductivity among diﬀerent polythiophene  11 1 in N-methylpyrrolidone (NMP) on aluminum foils. A er solvent structures. The weaker absorption peaks at 1490 and 694 cm evaporation, the electrodes were cut into discs with a diameter of the PT can be assigned to the C–H and C–S stretching of the of 12.7 mm and dried thoroughly under vacuum at 120 C for 12 thiophene ring, respectively.17,41 As for the pure PI and PI@PT h. Half-cells for electrochemical measurement were assembled composites, the characteristic vibration absorptions of C]O with the polymer composites as the cathode, lithium metal foils and C–N at 1703 and 1318 cm 1 are observed and fully as both counter and reference electrodes, Celgard 2320 as the consistent with the previously reported spectra, suggesting the separator and 1 M LiPF6 dissolved in EC, DEC, and DMC successful synthesis of the target aromatic polyimide compos- (1 : 1 : 1 vol) as the electrolyte. Coin cells were assembled in an ites.34,42 Meanwhile, the intensities of the aforementioned PT  argon lled glove-box with the oxygen and moisture contents characteristic peaks increase with increasing the PT amount, below 0.5 ppm. The coin cells were cycled galvanostatically at from PI10PT to PI50PT. Furthermore, the spectra of the PI@PT various current rates on a LAND CT2001A battery test system in composites are only superposition of the characteristic a voltage range from 1.8 to 3.2 V. Cyclic voltammetry (CV) was absorption peaks of PI and PT without generating additional carried out on a Biologic VSP instrument with a scan rate of 0.05 new peaks, demonstrating the non-covalent interaction 1 – mV s in a voltage range of 1.8 3.2 V. Electrochemical between the PI and PT coating, thus preserving the integrity of impedance spectra (EIS) were measured on the same instru- the redox-active carbonyl groups in PI or the electrochemical ment with a 10 mV AC bias in a frequency range from 200 kHz to reactivity of the composite electrode. 10 mHz. The morphologies of PI, PI10PT, PI30PT, PI50PT and pure PT were characterized by SEM, as shown in Fig. 2. The pristine PI 3. Results and discussion particle shows an apparent porous surface structure (Fig. 2a). In contrast, the surface of the pure PT particle is compact and The PI@PT composites were synthesized by an in situ oxidation smooth, without any porous structure (Fig. 2e). For the PI@PT polymerization method. To ensure the homogeneous coating of composites, the porous surface structure of the PI particle PT on the surface of PI, the latter was ground into a ne powder gradually disappears with increasing the amount of PT coating, as shown in Fig. 2b–d. This is consistent with the BET

Published on 31 October 2017. Downloaded by University of Tennessee at Knoxville 26/12/2017 23:23:53. and well dispersed in chloroform in which the monomer thio- ‡ phene was soluble. The surface coating was proceeded slowly measurement, as shown in Fig. S1. The surface areas of PI, under the catalysis of FeCl at a low temperature of 0 C. With PI10PT, PI30PT, PI50PT and PT are 74.3, 32.6, 24.2, 13.9 and 3 13.7 m2 g 1, respectively. It is noted that the surface coating of PT in the PI50PT sample is so thick that it covers the entire PI particle (Fig. 2d), similar to pure PT (Fig. 2e). This compact and non-porous morphology of PI50PT will impede the penetration of the liquid electrolyte, aﬀecting the diﬀusion of lithium ions as well as rate capability as will be shown later. To further investigate the polymeric composites, EDS elemental analysis was also carried out. Fig. 3a shows a typical EDS mapping of the PI30PT sample. As can be seen from the gure, a uniform distribution of carbon and sulfur whereas almost no nitrogen and oxygen can be found within the sample area, conrming that PT was indeed coated on the surface of PI. Fig. 3b shows the EDS spectra of PT, PI, PI10PT, PI30PT and PI50PT. The peaks of nitrogen and oxygen can be clearly seen in the PI sample, while they are hardly observed in the other composite samples. In contrast, the sulfur peak in the PI@PT composites gradually increases with increasing the content of Fig. 1 Photographs of (a) PI, (b) PI10PT, (c) PI30PT, (d) PI50PT and (e) PT. It is noted that the spectrum of the PI50PT sample is almost PT powders; (f) FTIR spectra of PI, PI10PT, PI30PT, PI50PT and PT. identical to that of pure PT, suggesting that the PI particles in

This journal is © The Royal Society of Chemistry 2017 J. Mater. Chem. A,2017,5,24083–24090 | 24085 View Article Online Journal of Materials Chemistry A Communication

Fig. 2 SEM images of (a) PI, (b) PI10PT, (c) PI30PT, (d) PI50PT, and (e) PT.

corresponding to a step-wise formation of a radical anion and dianion (Scheme 1).19 In comparison, the corresponding doublets for both PI and PI10PT are not well dened, which can be attributed to their low electronic conductivity due to no or less content of highly conductive PT coating, resulting in slow charge transfer kinetics between the radical anion and the dianion.43 It is noticed that the peak currents of PI50PT are much lower than those of PI30PT, attributed to poor Li-ion diffusivity because of a too thick PT coating in PI50PT. Furthermore, the onset potentials of the rst reduction peak and the rst oxidation peak during the discharge/charge process are 2.41 and 2.75 V, 2.29 and 2.78 V, 2.21 and 2.83 V, and 2.28 and 2.80 V for the PI, PI10PT, PI30PT and PI50PT electrodes, respectively. The higher potential during discharge and lower potential during charge for the PI30PT electrode Published on 31 October 2017. Downloaded by University of Tennessee at Knoxville 26/12/2017 23:23:53. suggest that it has the lowest polarization among these polymeric electrodes, resulting in high utilization efficiency of the polymer cathode and high specic capacity as will be demonstrated later. Moreover, the CV of pristine PT shows no peaks in the voltage range between 1.8 V and 3.2 V (Fig. S2‡), conrming Fig. 3 (a) SEM image of PI30PT, its total as well as individual EDS that the PT contribution in the PI@PT composites in the tested mappings of elements C, N, O and S; (b) EDS spectra of PT, PI, PI10PT, PI30PT and PI50PT. voltage range is only electron conductivity. To further understand the effect of the polythiophene coating on the Li-ion diffusion in the electrodes, CVs of the PI, PI10PT, PI30PT and PI50PT composite electrode were obtained PI50PT are totally covered by PT. These results further conrm at various scan rates in the range of 0.05–2.0 mV s 1, as shown the successful synthesis of the target polymer composites of in Fig. S3a–d.‡ The Li-ion diffusion coefficients are calculated PI@PT. using the Randles–Sevcik equation (eqn (1)):44,45 Fig. 4a shows the cyclic voltammograms (CVs) of the PI, PI10PT, PI30PT and PI50PT composite electrodes in the voltage I ¼ 269000n3/2AD1/2Cn1/2 p (1) range of 1.8–3.2 V at a scan rate of 0.05 mV s 1. Two pairs of well-resolved redox peaks can be observed for the PI30PT elec- where Ip is the peak current (A), n is the electron concentration trode, involving two reduction peaks at 2.39 and 2.58 V and two per molecule during the redox reactions, A is the surface area of corresponding oxidation peaks at 2.52 and 2.77 V, respectively. the electrodes (cm2), D is the diffusion coefficient of lithium  A reversible two-electron redox reaction is con rmed by the ions (cm2 s 1), C is the bulk concentration of lithium ions in the doublets during both lithiation and de-lithiation processes, electrodes (mol cm 3) and n is the scan rate (V s 1). A linear

24086 | J. Mater. Chem. A,2017,5,24083–24090 This journal is © The Royal Society of Chemistry 2017 View Article Online Communication Journal of Materials Chemistry A

Fig. 4 Cyclic voltammograms (CVs) of (a) PI, PI10PT, PI30PT and PI50PT at a scan rate of 0.05 mV s 1; charge–discharge proﬁles of (b) PI, PI10PT, PI30PT and PI50PT at a current rate of C/10.

1/2 relationship between Ip and n was obtained as shown in the half-cells based on PI and PI10PT are 116.0, 67.0, 41.9, 32.0, Fig. S3e.‡ The diffusion coefficients of lithium ions were 23.2, 11.4, 90.2 mA h g 1 and 192.5, 174.6, 143.1, 111.1, 71.1, calculated according to the slopes. It is estimated that the 18.4, 183.4 mA h g 1 at current rates of C/10, C/5, C/2, 1C, 2C, diffusion coefficients of lithium ions in PI, PI10PT, PI30PT and 5C, and C/10, respectively. The comparison of rate performance PI50PT are 6.39 10 13 and 8.35 10 13 cm2 s 1, 4.73 10 11 in Fig. 5a suggests that the PI30PT cathode has the best and 4.79 10 11 cm2 s 1, 3.14 10 10 and 2.92 10 10 cm2 combination of electron conductivity and lithium reaction s 1, and 9.60 10 11 and 7.08 10 11 cm2 s 1 for oxidation kinetics, consistent with its highest lithium diffusion coeffi- and reduction, respectively. The lithium ion diffusion coefficient. The reversible capacities of the PI50PT based cell are cient in the PI10PT electrode is almost two orders magnitude always lower than those of the PI10PT based cell except at 5C, higher than that in the PI electrode, but it is still one order although the former has a higher lithium diffusion coefficient magnitude lower than that in the PI30PT electrode. Compara- than the latter. This is closely related to the thickness of the tively, the Li-ion diffusion coefficient in the PI30PT electrode is surface coating and the different surface morphologies of the more than triple that in the PI50PT electrode, which is double primary composite particles, as shown in Fig. 2. Although the that in the PI10PT electrode. These data conrm that the lithium diffusion coefficient in PI50PT is double that of PI10PT, PI30PT electrode has the highest lithium diffusion coefficients, on average the thickness of the PT coating in PI50PT is about which is consistent with the rate performance shown in Fig. 5. ve times that of PI10PT; therefore, at the same current rate Fig. 4b shows the galvanostatic charge/discharge proles of lithium ions will reach the electroactive PI center faster in the half-cells based on PI, PI10PT, PI30PT and PI50PT at PI10PT than in PI50PT, resulting in higher capacity in the a current rate of C/10. As expected from the CVs in Fig. 4a, a two- former. Published on 31 October 2017. Downloaded by University of Tennessee at Knoxville 26/12/2017 23:23:53. step charge/discharge process can be clearly observed for all the Besides good rate performance, the PI30PT based half-cell PI@PT composite materials. However, only a slope is observed also exhibits better cycling stability, as shown in Fig. 5b. Aer for the PI based cell during the charge process, which is a gradual increase within the initial few cycles, the reversible attributed to the severe polarization due to its low electronic capacity of the PI30PT based cell reaches a high value of 186.6 conductivity. These results are highly consistent with the CV mA h g 1 at a current rate of C/2. Then it decreases slowly with analysis in Fig. 4a, further conrming that the PI30PT electrode cycling and is still as high as 170.9 mA h g 1 aer 300 cycles, delivers low cell polarization and fast lithium reaction kinetics. resulting in a high capacity retention of 92%. In contrast, the Fig. 5 compares the rate and cycling performance of the half- reversible capacities of the PI10PT and PI50PT based cells are cells based on PI, PI10PT, PI30PT and PI50PT, as well as the EIS 140.1 and 106.6 mA h g 1 aer the rst few activation cycles, data of the cells before and aer cycling. A common feature for and then decrease gradually to 101.5 and 71.6 mA h g 1 aer all the polyimide based materials is the initial gradual increase 300 cycles, exhibiting capacity retentions of 72% and 67%, of capacity, which can be attributed to the activation process respectively. The fabulous cycling stability of the PI30PT based because of the gradual penetration of the liquid electrolyte into cell is also supported by the EIS data before and aer cycling as the polymeric composite cathodes.46,47 As shown in Fig. 5a, the shown in Fig. 5c and d, respectively. The total cell impedance PI30PT based cell exhibits reversible capacities of ca. 216.8, before cycling continually decreases with increasing the PT 193.6, 169.1, 151.1, 132.2 and 104.7 mA h g 1 at current rates of content, consistent with the excellent electronic conductivity of C/10, C/5, C/2, 1C, 2C and 5C, respectively, with a high PT. Aer 300 cycles, the total cell impedance of all the cells coulombic efficiency of 100% except for the rate changing increases. However, it only increases from 40.8 U before cycling cycles. The capacity recovers to 205.8 mA h g 1 when the current to 57.8 U aer 300 cycles for the PI30PT based cell, whereas it rate is switched back to C/10, which is equivalent to a 95% increases from 15.2, 107.3, and 188.8 U before cycling to 58.8, capacity retention. As a comparison, the reversible capacities of 286.7, and 417.4 U aer 300 cycles for the PI50PT, PI10PT and PI

This journal is © The Royal Society of Chemistry 2017 J. Mater. Chem. A,2017,5,24083–24090 | 24087 View Article Online Journal of Materials Chemistry A Communication Published on 31 October 2017. Downloaded by University of Tennessee at Knoxville 26/12/2017 23:23:53.

Fig. 5 Charge–discharge capacities and coulombic efficiencies of the half-cells based on PI, PI10PT, PI30PT and PI50PT at (a) different current rates and (b) a current rate of C/2; electrochemical impedance spectra of the half-cells based on PI, PI10PT, PI30PT and PI50PT (c) before cycling and (d) after 300 cycles at C/2; (e) charge–discharge capacities and coulombic efficiencies of the half-cells based on PI30PT at different high current rates (hollow circles and solid spheres represent charge and discharge capacities, respectively, while hollow stars represent coulombic efficiencies in (a), (b) and (e)).

based cells, respectively. It is noted that, although the imped- transfer resistance in PI50PT is faster than that in PI30PT at the ance of PI50PT is lower than that of PI30PT before cycling, the same current rate. This also underlines the importance of charge transfer resistance associated with the semi-arc in balancing electron conductivity with lithium diffusion in the PI50PT increases more quickly than that in PI30PT, and it is synthesis of composite electrode materials. even higher than that in PI30PT aer 300 cycles. This To further investigate the potential practical application of phenomenon is closely related to the thickness of the PT the PI30PT composite material, its long-term cycling perfor- coating on the PI@PT composites. Although more PT contrib- mance at high current rates was also evaluated, as shown in utes to higher electron conductivity and lower resistance before Fig. 5e. The reversible capacities are 116.3, 107.1 and 89.6 mA h cycling, unfortunately, a too thick PT coating impedes lithium g 1 at high current rates of 5C, 10C and 20C, respectively, with reaching the electroactive center of PI during cycling. As coulombic efficiencies all near 99.8%. Aer 1000 cycles, demonstrated earlier, the lithium diffusion coefficient in reversible capacities are still as high as 102.5, 90.8 and 84.3 mA PI30PT is more than triple that in PI50PT whereas the thickness hg 1 at 5C, 10C and 20C, leading to high capacity retentions of of the PT coating in the former is nearly half that in the latter; 88%, 85% and 94%, respectively. The superior cycling perfor- therefore, it is not surprising that the increase of the charge mance of the PI30PT based cells can be attributed to the

24088 | J. Mater. Chem. A,2017,5,24083–24090 This journal is © The Royal Society of Chemistry 2017 View Article Online Communication Journal of Materials Chemistry A

optimal content of PT which not only provides good electron 3B.Häupler, A. Wild and U. S. Schubert, Adv. Energy Mater., conductivity but also successfully acts as a protection shell to 2015, 5, 1402034. maintain the structural integrity of the polyimide composite, 4 B. K. Guo, X. Q. Yu, X. G. Sun, M. F. Chi, Z. A. Qiao, J. Liu, preventing it from decomposing and dissolving in the liquid Y. S. Hu, X. Q. Yang, J. B. Goodenough and S. Dai, Energy electrolyte during cycling via the strong p–p interactions Environ. Sci., 2014, 7, 2220–2226. between the polyimide matrix and the PT coating. Table S1‡ 5 H. Lyu, J. Liu, S. Qiu, Y. Cao, C. Hu, S. Guo and Z. Guo, J. further compares the battery performance of PI30PT with those Mater. Chem. A, 2016, 4, 9881–9889. of reported PI composite materials. As can be seen from the 6 J. B. Goodenough and Y. Kim, Chem. Mater., 2010, 22, 587– table, PI30PT possesses high specic capacity at both low and 603. high current rates, especially the long-term cycling stability at 7 V. Etacheri, R. Marom, R. Elazari, G. Salitra and D. Aurbach, high current rates is unprecedented, suggesting the great Energy Environ. Sci., 2011, 4, 3243. potential of PI30PT for practical application. In addition, with 8 Y. Liang, Z. Tao and J. Chen, Adv. Energy Mater., 2012, 2, 742– electron conductive polymer PT replacing expensive carbon 769. materials such as single-wall carbon nanotubes and 3D reduced 9 J. Chen, N. W. Ding, Z. F. Li, Q. Zhang and S. W. Zhong, graphene oxide, the total polymeric composite PI30PT repre- Progr. Chem., 2015, 27, 1291–1301. sents a promising low-cost cathode material for “green and 10 J. Wang, J. Yang, J. Xie and N. Xu, Adv. Mater., 2002, 14, 963– sustainable” lithium ion batteries. 965. 11 F. Wu, S. Wu, R. Chen, J. Chen and S. Chen, Electrochem. 4. Conclusion Solid-State Lett., 2010, 13, A29. 12 W. Guo, Y.-X. Yin, S. Xin, Y.-G. Guo and L.-J. Wan, Energy Novel polymeric composites based on coating aromatic poly- Environ. Sci., 2012, 5, 5221–5225. imides with an electron conducting polythiophene have been 13 K. Nakahara, K. Oyaizu and H. Nishide, Chem. Lett., 2011, 40, prepared by a facile in situ chemical oxidation polymerization 222–227. approach for application in rechargeable lithium ion batteries. 14 X. L. Wei, W. Xu, M. Vijayakumar, L. Cosimbescu, T. B. Liu, The optimal PT coating, 30 wt% (PI30PT), enables high electron V. Sprenkle and W. Wang, Adv. Mater., 2014, 26, 7649–7653. conductivity and fast lithium reaction kinetics. Therefore, the 15 T. Janoschka, M. D. Hager and U. S. Schubert, Adv. Mater., PI30PT composite electrode delivers not only a reversible specic 2012, 24, 6397–6409. capacity of 216.8 mA h g 1 at a low current rate of C/10 but also 16 K. Nakahara, S. Iwasa, M. Satoh, Y. Morioka, J. Iriyama, a remarkable high-rate cyclability, achieving a high capacity of M. Suguro and E. Hasegawa, Chem. Phys. Lett., 2002, 359, 89.6 mA h g 1 at 20C with a capacity retention of 94% aer 1000 351–354. cycles. These superior electrochemical properties result from the 17 L. Liu, F. Tian, X. Wang, Z. Yang, M. Zhou and X. Wang, elaborate synergy of the stable redox reversibility of PI and the React. Funct. Polym., 2012, 72,45–49. high electronic conductivity of PT. Overall, PI30PT is proven to be 18 S. R. Sivakkumar and D.-W. Kim, J. Electrochem. Soc., 2007, a promising cathode material candidate for practical application 154, A134. in “green and sustainable” lithium ion batteries. 19 Z. Song, H. Zhan and Y. Zhou, Angew. Chem., Int. Ed., 2010, 49, 8444–8448. Published on 31 October 2017. Downloaded by University of Tennessee at Knoxville 26/12/2017 23:23:53. Conflicts of interest 20 W. Xu, A. Read, P. K. Koech, D. H. Hu, C. M. Wang, J. Xiao, A. B. Padmaperuma, G. L. Graff, J. Liu and J. G. Zhang, J. There are no conicts to declare. Mater. Chem., 2012, 22, 4032–4039. 21 B. F. Pan, J. H. Huang, Z. X. Feng, L. Zeng, M. N. He, Acknowledgements L. Zhang, J. T. Vaughey, M. J. Bedzyk, P. Fenter, Z. C. Zhang, A. K. Burrell and C. Liao, Adv. Energy Mater., This work was supported by the ORNL laboratory-directed 2016, 6, 1600140. research and development (LDRD) program. The electron 22 Y. l. Liang, Y. Jing, S. Gheytani, K. Y. Lee, P. Liu, A. Facchetti microscopy work was performed through a user project supported and Y. Yao, Nat. Mater., 2017, 16, 841–848. by the ORNL's Center for Nanophase Materials Sciences, which is 23 S. Muench, A. Wild, C. Friebe, B. Haupler, T. Janoschka and sponsored by the US Department of Energy, Office of Science, U. S. Schubert, Chem. Rev., 2016, 116, 9438–9484. Scientic User Facility Division. J. Liu and H. Lyu acknowledge the 24 Z. P. Song and H. S. Zhou, Energy Environ. Sci., 2013, 6, 2280– nancial support from the National Natural Science Foundation 2301. of China (No. 51572157). H. Lyu also acknowledges support from 25 A. F. Shestakov, O. V. Yarmolenko, A. A. Ignatova, the China Scholarship Council (CSC) program. A. V. Mumyatov, K. J. Stevenson and P. A. Troshin, J. Mater. Chem. A, 2017, 5, 6532–6537. References 26 P. Sharma, D. Damien, K. Nagarajan, M. M. Shaijumon and M. Hariharan, J. Phys. Chem. Lett., 2013, 4, 3192–3197. 1 B. Dunn, H. Kamath and J. M. Tarascon, Science, 2011, 334, 27 M. Lee, J. Hong, H. Kim, H. D. Lim, S. B. Cho, K. Kang and 928–935. C. B. Park, Adv. Mater., 2014, 26, 2558–2565. 2 J.-M. Tarascon and M. Armand, Nature, 2001, 414, 359–367.

This journal is © The Royal Society of Chemistry 2017 J. Mater. Chem. A,2017,5,24083–24090 | 24089 View Article Online Journal of Materials Chemistry A Communication

28 X. Y. Han, G. Y. Qing, J. T. Sun and T. L. Sun, Angew. Chem., 38 W. Xiao, J. S. Chen, Q. Lu and X. W. Lou, J. Phys. Chem. C, Int. Ed., 2012, 51, 5147–5151. 2010, 114, 12048–12051. 29 W. Luo, M. Allen, V. Raju and X. L. Ji, Adv. Energy Mater., 39 F. Wu, J. Chen, R. Chen, S. Wu, L. Li, S. Chen and T. Zhao, J. 2014, 4, 1400554. Phys. Chem. C, 2011, 115, 6057–6063. 30 Z. Song, T. Xu, M. L. Gordin, Y. B. Jiang, I. T. Bae, Q. Xiao, 40 F. Zhang, Y. Shi, Z. Zhao, W. Song and Y. Cheng, Appl. Catal., H. Zhan, J. Liu and D. Wang, Nano Lett., 2012, 12, 2205–2211. B, 2014, 150–151, 472–478. 31 Y. Huang, K. Li, J. Liu, X. Zhong, X. Duan, I. Shakir and Y. Xu, 41 E. Tahmasebi, Y. Yamini, M. Moradi and A. Esrali, Anal. J. Mater. Chem. A, 2017, 5, 2710–2716. Chim. Acta, 2013, 770,68–74. 32 H. Wu, S. A. Shevlin, Q. Meng, W. Guo, Y. Meng, K. Lu, Z. Wei 42 C.-P. Constantin, M.-D. Damaceanu, M. Bruma and and Z. Guo, Adv. Mater., 2014, 26, 3338–3343. M. Pinteala, Polym. Int., 2015, 64, 361–372. 33 H. P. Wu, Q. Yang, Q. H. Meng, A. Ahmad, M. Zhang, 43 Z. Song, H. Zhan and Y. Zhou, Chem. Commun., 2009, 448– L. Y. Zhu, Y. G. Liu and Z. X. Wei, J. Mater. Chem. A, 2016, 450, DOI: 10.1039/b814515f. 4, 2115–2121. 44 Z. H. Bi, M. P. Paranthaman, B. K. Guo, R. R. Unocic, 34 Y. Meng, H. Wu, Y. Zhang and Z. Wei, J. Mater. Chem. A, H. M. Meyer, C. A. Bridges, X. G. Sun and S. Dai, J. Mater. 2014, 2, 10842. Chem. A, 2014, 2, 1818–1824. 35 P. Bujak, I. Kulszewicz-Bajer, M. Zagorska, V. Maurel, 45 M. Stromme, G. A. Niklasson and C. G. Granqvist, Solid State I. Wielgus and A. Pron, Chem. Soc. Rev., 2013, 42, 8895–8999. Commun., 1995, 96, 151–154. 36 A. Pron, P. Gawrys, M. Zagorska, D. Djurado and 46 C. Chen, X. Zhao, H.-B. Li, F. Gan, J. Zhang, J. Dong and R. Demadrille, Chem. Soc. Rev., 2010, 39, 2577–2632. Q. Zhang, Electrochim. Acta, 2017, 229, 387–395. 37 L. Groenendaal, G. Zotti, P. H. Aubert, S. M. Waybright and 47 S. R. Sivakkumar, D. R. MacFarlane, M. Forsyth and J. R. Reynolds, Adv. Mater., 2003, 15, 855–879. D.-W. Kim, J. Electrochem. Soc., 2007, 154, A834. Published on 31 October 2017. Downloaded by University of Tennessee at Knoxville 26/12/2017 23:23:53.