High sulfur loading in activated bamboo-derived porous carbon as a superior cathode for rechargeable Li–S batteries

1 Introduction

Among various types of Li-ion batteries (LIBs), rechargeable lithium–sulfur (Li–S) batteries have attracted extensive attention due to their high theoretical capacity of 1675 mA h g⁻¹ and theoretical power density of 2600 W h kg⁻¹, which are about five times greater than those of commercialized LIBs currently. Besides, the cathode active material of element sulfur is very abundant in nature, low cost and environmental friendly (Goodenough and Kim, 2009; Liu et al., 2009). All of these advantages endow Li–S batteries as one of the most promising candidates for next generation of high energy density rechargeable batteries. However, in spite of these advantages, the practical application of Li–S batteries is still hampered by several problems, including the stability of lithium metal and the insulating nature of pure sulfur with extraordinary low conductivity of ∼10⁻³⁰ s cm⁻¹ at room temperature (Tarascon and Armand, 2001). It is difficult to have sufficient electrochemical contact between the cathode active materials of sulfur, conductive carbon and the current collector, thus resulting in the insufficient utilization of sulfur (Li et al., 2013a). In addition, the dissolution of long chain lithium polysulfides into the organic electrolyte, which causes irreversible capacity loss and low coulombic efficiency is another huge problem impeding the commercialization of Li–S batteries (Zhang et al., 2014). These troubles will lead to poor cycle stability and short cycle life of Li–S batteries.

To overcome these issues, considerable strategies have been proposed, of which most are generally focused on novel cell configurations, electrolyte additives and especially sulfur composite cathode materials with preferable nanostructure and morphology (Ji and Nazar, 2010). For example, much progress had been made to improving sulfur cathode performance with sulfur/porous carbon composite (Li et al., 2011, 2013a, 2015; Zhang et al., 2010; Manthiram et al., 2012; Jayaprakash et al., 2011), sulfur/graphene composite (Wang et al., 2011; Song et al., 2013; Zhou et al., 2014) and sulfur/polymer composite materials (Li et al., 2012, 2013b; Wu et al., 2011). Recently, sulfur/biomass composites have attracted extensive attention for their unique microstructure and morphology, low cost and easy accessibility (Zhang et al., 2014; Gu et al., 2015a,b; Cheng et al., 2015; Moreno et al., 2014; Wang et al., 2014). Biomass-derived carbon materials have good electronic conductivity, large specific surface area and large pore volume accommodating the volume changes, and they have been intensively applied as the anode materials for Li-ion batteries (Hwang et al., 2008; Fey et al., 2003; Xing et al., 1996). These results imply that the biomass-derived carbon materials may be suitable matrix for loading electroactive sulfur for Li–S batteries. Herein, we synthesized a sulfur-loading carbon matrix derived from natural bamboo waste. To further increase the specific surface area (SSA) and pore volume of the raw bamboo, it was firstly grounded small and soaked with 1.5 M KOH solution for 5 days. The obtained activated bamboo-derived carbon (A_BC) has a large SSA of 1565.4 m² g⁻¹ and a total pore volume of 0.95 cm³ g⁻¹. With this highly porous carbon material, a kind of A_BC/S composite cathode with 86 wt% sulfur was designed, and the obtained A_BC/S composite cathode exhibited stable and superior electrochemical performances for Li–S batteries, which may be attributed to its abundant porous and pipeline structure.

2 Experimental procedure

3 Results and discussion

XRD patterns of pure sulfur, A_BC and A_BC/S-86% are displayed in Fig. 1. By comparison, it is obvious that the A_BC has no any sharp peaks, indicating the A_BC is mainly the amorphous carbon. However, the XRD pattern of the A_BC/S-86% includes sharp peaks matching with the standard value (JCPDF NO. 08-0247) of pure sulfur and the broaden peaks matching with the A_BC at the same degree. Furthermore, the pattern of the A_BC/S also reveals that sulfur exists in a mixed state of both amorphous and crystalline phases (Lai et al., 2009), which may be due to the fact that sulfur content in the composite is rather high, and some sulfur particles still retained on the surface of the composite after heated. Previous research shows that when sulfur content in the composite is low, no clear peaks of the sulfur can be observed because sulfur has become amorphous distributed into the matrix (Yin et al., 2012).

Close

Figure 1

XRD patterns of the standard data of sulfur (brown), A_BC (blue) and A_BC/S-86% composite (red).

Export to PPT

TGA was carried out to further confirm the accurate sulfur content in the composite of A_BC/S displayed in Fig. 2. Normally, pure sulfur starts to vaporize at about 200 °C and the rate of evaporation becomes particularly fast between 200 and 300 °C, finally finished at about 330 °C. The weight loss of A_BC/S composite results from the evaporation of sulfur stored in pores of A_BC matrix. The content of sulfur in the composite is measured to be 86 wt%, exceeding most of the previous related reports (Gu et al., 2015a; Ji et al., 2009; Li et al., 2014; Xie et al., 2014; Miao et al., 2013). As we know, a high sulfur loading in the cathode provides higher energy densities and tends to be more practical for its commercial application.

Close

Figure 2

TGA-traces of pure sulfur and A_BC/S-86% composite from room temperature to 500 °C at a heating rate of 10 °C min⁻¹ under N₂ atmosphere.

Export to PPT

As shown in Fig. 3a, abundant pores and pristine vascular bundle structure retained from the raw bamboo can be obviously observed, which can facilitate infiltration of the electrolyte and acts as large storage container when sulfur content is very high (Zhang et al., 2014). The morphology of the A_BC/S-86% composite displayed in Fig. 3b is clearly different from that in Fig. 3a, without any pores that can be observed and the surface of the composite becomes smoother. Moreover, some sulfur particles aggregated on the surface of composite as well, which mainly is due to the fact that sulfur content in A_BC/S composite is relatively high and the pores in the A_BC matrix are fully filled with sulfur. As illustrated in Fig. 3c, abundant micropores in the A_BC can be observed in the high-resolution TEM (HRTEM) image, which is consistent with the N₂ adsorption/desorption results (Fig. 4). Dense of micropores can alleviate the dissolution of long chain lithium polysulfides. At the same time, it can be observed that the sulfur homogeneously dispersed within the pores of A_BC from the TEM image (Fig. 3d), but some large sulfur particles are still observed. In order to further verify the distribution of sulfur in the composite, EDS mapping analysis was carried on the A_BC/S-86% composite, and the corresponding elemental mapping images for sulfur and carbon are shown in Fig. 3f and g, respectively. As illustrated in Fig. 3f and g, the bright yellow spots stand for the element S and the black spots represent the element C. We can find a uniform coverage of sulfur on the external surface throughout the whole area, with only a small amount of bare carbon. These test results demonstrate that vast majority of sulfur has been well diffused into the pores of A_BC with small portion covering on the surface of the composite. The homogeneous distribution of sulfur in the composite can benefit to improving the performance of Li–S batteries.

Close

Figure 3

SEM (a, b) and TEM (c, d) images of A_BC (a, c) and A_BC/S-86% (b, d); (e–g) EDS mapping of the region displayed in (e) for sulfur mapping (f) and carbon mapping (g). The inset of (a) shows photo image of the typical bamboos grown in the central south of China (Xiangtan, Hunan province).

Export to PPT

Close

Figure 4

N₂ adsorption/desorption isotherms of A_BC (a), the A_BC/S-86 wt% (b), and the corresponding pore size distribution calculated by the Barrett–Joyner–Halenda method (c).

Export to PPT

The specific surface area analysis of the A_BC was performed by nitrogen adsorption measurements, and the corresponding pore size distribution was calculated by the Horvath–Kawazoe method. Fig. 4a shows that A_BC has a typical microporous structure (Zhang et al., 2010). The specific surface area of A_BC is 1565.4 m² g⁻¹, mainly contributed by micropores and mesopores within the matrix. As we know, high specific surface area and large pore volume can contribute to accommodating high sulfur content, shorten the distance for charge transport and provide more reactive sites to improve the utilization of active sulfur material (Gu et al., 2015b; Hu et al., 2015). As shown in Fig. 4b, the SSA of A_BC/S (86 wt%) composite is only about 11 m² g⁻¹, and the pore volume decreases from 0.95 cm³ g⁻¹ to 0.028 cm³ g⁻¹ immediately, which is mainly caused by the pores within the A_BC which were filled with sulfur particles (Zhang et al., 2014). The curve of the pore size distribution (Fig. 4c) indicates the A_BC mainly possesses micro-mesopores, displaying a narrow pore size distribution in the range of 1.5–5 nm, which are consistent with the observation in HRTEM image (Fig. 3c). These abundant micropores with high surface area enable sufficient contact between the insulating S and conductive A_BC. Moreover, they can act as porous reactors and restrict the dissolution of long chain lithium polysulfides into the organic electrolyte (Gu et al., 2015a). However, in the curve of A_BC/S (86 wt%) composite displayed in Fig. 4c, none of the pores can be seen, which are highly consistent with the SEM image in Fig. 3b. Due to the strong physical adsorption of capillary force, the sulfur impregnated into the pores of A_BC matrix can maintain stable. The Raman spectrum of A_BC was conducted to further confirm the nature and type of the as-prepared carbon (A_BC), as shown in Fig. 5, the two major Raman bands are located at around 1325 and 1600 cm⁻¹, which are identified as the D band and G band of graphene nanosheets, and the peak height ratio of I_D/I_G is 0.94, indicating the graphitization phase plays a big role in the as-prepared carbon of A_BC (Liu et al., 2013), which can promote the electrical conductivity of A_BC (Gu et al., 2015a).

Close

Figure 5

Raman spectrum of the activated bamboo-derived carbon (A_BC).

Export to PPT

Cyclic voltammetry testing was conducted to investigate the electrochemical mechanisms of the as-prepared A_BC/S cathode materials. Fig. 6 shows the CV curves of the A_BC/S-86% cathode at a scan rate of 0.1 mV s⁻¹ for the first three cycles. In the first cycle of the cathode reduction process, two peaks at approximately 2.35 V and 2.0 V, respectively are observed. The first reduction peak (2.35 V) corresponds to the reduction process of elemental sulfur to long chain lithium polysulfides (Li₂S_x, 4 < x < 8), and the other peak (2.0 V) is attributed to the conversion of long chain lithium polysulfides to short chain lithium polysulfides or even to insoluble Li₂S (Wang et al., 2011). In the subsequent anodic scans, only one oxidation peak is seen at around 2.38 V, which can be ascribed to the conversion from short chain lithium polysulfides to long chain lithium polysulfides and sulfur (Li et al., 2013a; Zhou et al., 2013). More importantly, from the second and the third cycles onward, the position and areas of the CV peaks remain rather constant and similar shapes upon cycling, indicating the excellent electrochemical stability of the A_BC/S electrode (Ding et al., 2013; Fu and Manthiram, 2012). As shown in Fig. 7a, the A_BC/S-86% composite cathode cycled at different current rates increased from 0.1 to 1 C and returned back to 0.2 C (1 C = 1675 mA g⁻¹) in the voltage range of 1.0–3.0 V, and the battery exhibited great rate capability and cycling stability. An initial capacity of 1160 mA h g⁻¹ was obtained at 0.1 C, further cycled at 0.2, 0.5 and 1 C rates, and it also showed stable reversible capacities about 930, 695 and 580 mA h g⁻¹, respectively. When the rate returned to 0.2 C, 850 mA h g⁻¹ was obtained again. Fig. 7b shows the charge/discharge profiles of the A_BC/S-86% composite cathodes, two obvious plateaus present in the discharge curves, corresponding to the two different reduction processes for the A_BC/S composite cathode. Furthermore, it is observed that the plateaus become shorter and lower along with the growth of rates and cycling number as shown in Fig. 7b and d. All these discharge curves have a large increased capacity from 1.5 V to 1.0 V, which mainly attributed to the deep reduction from short chain polysulfides to the final products of Li₂S and Li₂S₂. And the long-cycling performance of the A_BC/S-86% cathode at 0.2 C is shown in Fig. 7c, an initial capacity of 930 mA h g⁻¹ is obtained and the capacity decays relatively fast for the initial 30 cycles, which is mainly due to the loss of active sulfur and the dissolution of long chain polysulfides into electrolyte, then the capacity curves tend to be horizontal and a reversible capacity of over 710 mA h g⁻¹ still remained even after one hundred cycles with a stable coulombic efficiency of about 95%. Fig. 7d displays the charge/discharge profiles of the cathode at 0.2 C, and we can see these plateaus are very long and stable, which further confirms the stability and great rate capability of A_BC/S composite cathodes. Moreover, the cycle performance and multi-rate capability comparison of the A_BC/S and pristine sulfur are displayed in Fig. 7e and f, respectively. In order to investigate the changes of electrode morphology before and after cycling, the SEM images of lithium fresh, after first and 100 cycles are shown in Fig. 8a–c, respectively. As shown in Fig. 8a, the lithium fresh is very flat and smooth, and the surface of lithium metal after the first cycle also keeps tight and intact as shown in Fig. 8b. However, it is obviously observed that some thick passivation layers covered on the surface of lithium metal after 100 cycles displayed in Fig. 8c, which are attributed to the deposition of insoluble reduction products, and these passivation layers can protect lithium anode and provide Li-ion pathway, which are beneficial to achieving good rate capability and long-term cycle stability for Li–S batteries (Cheng et al., 2015). Fig. 8d and e shows the morphology of A_BC/S-86% composite before cycling and after 100 cycles, respectively. It can observed that the surface of cathode becomes unfairness and lots of corrosive holes attached to it after 100 cycles, which may be resulted from the loss of active materials and the dissolution of long chain lithium polysulfides along with cycling (see Tables 1 and 2).

Close

Figure 6

Cyclic voltammograms of the A_BC/S-86% composite cathode cycled between 1.0 and 3.0 V (vs. Li/Li⁺), recorded at a potential scanning rate of 0.1 mV s⁻¹.

Export to PPT

Close

Figure 7

Electrochemical performances of A_BC/S-86% cathode for Li–S batteries cycled between 1.0 and 3.0 V versus Li/Li⁺: (a) rate capability; (b) charge–discharge profiles at different current densities; (c) cycle performance of the A_BC/S-86% cathode at 0.2 C; (d) charge–discharge profiles of the battery at 0.2 C; (e, f) cycling performance and multi-rate capability of A_BC/S and pristine sulfur cathodes.

Export to PPT

Close

Figure 8

Surface morphologies of Li fresh anode (a), after the first one cycle (b), and after 100 cycles (c); surface morphologies of A_BC/S-86% cathode before (d) and after 100 cycles (e).

Export to PPT

To make a further investigation, electrochemical impedance spectroscopy (EIS) was measured to study the dynamics for lithium insertion and extraction during cycling. As shown in Fig. 9, after the first one cycle, the impedance spectrum is composed of a medium-to-high frequency semicircle and a long inclined line (Warburg impedance) in the low frequency region. The semicircle is attributed to the charge transfer process at the interface between the electrolyte and sulfur electrode, and the straight line represents the Li ions diffusion resistance within the cathode (Zhou et al., 2014; Liu et al., 2013). After 100 cycles, the impedance spectrum added a new small resistance, including two different semicircles followed by a short sloping line. Generally, the high-frequency semicircle results from the interfacial charge transfer process, and the semicircle in the medium frequency range is related to the solid–electrolyte-interface (SEI) film resistance caused by the formation of Li₂S or Li₂S₂ on the surface of cathode, and the straight line is attributed to the diffusion of Li ions in the matrix (Zhou et al., 2014; Yuan et al., 2009). The EIS results show that after the first cycle, the resistance of Rct is about 34 ohm, demonstrating that the composite of A_BC/S has great electrical conductivity, which is beneficial to the rate capability of the battery. However, after 100 cycles, the resistance of the cell is larger than that after the first cycle, which may be resulted from the dissolution of part long chain lithium polysulfides during both discharge and charge processes, reducing the migration speed of Li-ions in the electrolyte. As the activated bamboo carbon has a high surface area of 1565.4 m² g⁻¹, total pore volume of 0.95 cm³ g⁻¹ and abundant micro-mesopores, it can store a high content of sulfur and prevent long chain lithium polysulfides from dissolving into the organic electrolyte (Wang and Kaskel, 2012).

Close

Figure 9

Nyquist plots of the A_BC/S after the first one cycle (green) and the 100th cycle (red), measured with the frequency ranged from 100 kHz to 0.1 Hz.

Export to PPT

4 Conclusions

In summary, we have successfully synthesized a novel type of carbon from natural bamboo waste through simple activation and pyrolysis routes. The as-prepared A_BC possesses total pore volume of 0.95 cm³ g⁻¹ and high specific surface area up to 1565.4 m² g⁻¹. With these carbon materials, the cathode of A_BC/S composite with a high sulfur loading (86 wt%) was designed and exhibited good rate capability and cycling stability, with an initial discharge capacity of 1160 mA h g⁻¹ at 0.1 C, and even after one hundred cycles at 0.2 C, 710 mA h g⁻¹ is still retained. Further cycling at 0.5 C and 1 C, reversible capacities of 695 mA h g⁻¹ and 580 mA h g⁻¹ can be reserved with a high Coulombic efficiency, respectively.