Vikram K.
Bharti
,
Ananya
Gangadharan
,
S. Krishna
Kumar
,
Anil D.
Pathak
and
Chandra S.
Sharma
*
Creative & Advanced Research Based on Nanomaterials (CARBON) Laboratory, Department of Chemical Engineering, Indian Institute of Technology Hyderabad, Kandi-502285, Telangana, India. E-mail: cssharma@che.iith.ac.in
First published on 22nd March 2021
The commercial realization of next-generation lithium–sulfur (Li–S) batteries is mainly hindered due to the unwanted lithium polysulfide shuttling and the insulating nature of the sulfur cathode. In the present work, we aim to overcome these critical challenges by the first-time usage of candle soot carbon as a conducting host as well as an inbuilt interlayer. The Li–S battery thus fabricated delivers an impressive capacity of 1182 mA h g−1 with 92% coulombic efficiency at 0.1C. This excellent electrochemical performance is further maintained in long cycling even at a higher C-rate (1C) and exhibits a capacity of 667 mA h g−1 after 200 cycles with coulombic efficiency ∼95% (an extremely slow capacity decay rate of 0.03% per cycle). Moreover, for a high-sulfur loading (4.5 mg cm−2) electrode the Li–S battery retains 61.3% of the initial capacity after cycling for 150 cycles at 2.0C. Further, to understand the functional mechanism of the carbon interlayer for anchoring lithium polysulfides, first principles calculations are performed based on density functional theory. To the best of our knowledge, this is the first such report on using inexpensive candle soot carbon as a cathode host and as an interlayer that results in outstanding electrochemical performance.
Another important concern for Li–S batteries is the trapping of soluble long-chain lithium polysulfides to retain the battery cycle life without capacity fading. Manthiram et al.24 introduced the concept of an interlayer, a physical barricade between the cathode and the separator, to inhibit the movement of long-chain (higher order) lithium polysulfides from the cathode to the electrolyte. Moreover, the interlayer provides an alternative site for the adsorption of long-chain lithium polysulfides via physisorption or chemisorption, preventing the loss of active material (sulfur). Therefore, the cell can be operated for a long cycle life without capacity loss.20 The pioneering work by X. Gao et al.25–28 shed light on the use of carbon materials with various strategies for enhancing the electrochemical performance. The group has reported the use of a multifunctional globular polypyrrole interlayer27 as a polysulfide blockade, which showed impressive electrochemical performance with 74% capacity retention on cycling at 0.5C. In the following work by the group,28 they designed a porous hollow carbon aerogel using CaCO3 as a template and investigated the same as a cathode for Li–S batteries. The as prepared cathode exhibited excellent cycling with 60.5% capacity retention at 0.1C with a sulfur loading of 2.2 mg cm−2. Moreover, the promising results reported by several research groups29–31 emphasize that interlayer modification of Li–S batteries is a viable and efficient design method to block long-chain lithium polysulfide shuttling.
Herein, our approach is to create a conducting host as well as an inbuilt interlayer (as a polysulfide blockade) from the same source of carbon for enhanced interfacing between the electrode and interlayer. For this purpose, we chose an inexpensive combustion byproduct, candle soot, to play the dual role of an electrode inbuilt interlayer and a conducting host. The importance of this work lies in the development of an inbuilt candle soot interlayer over a candle soot–sulfur composite (SC) which can act as an excellent adhesive interlayer to the cathode and also nullify the polysulfide shuttling to a large extent. The enhancement of the capacity and cycle life with the candle soot interlayer is further verified by the electrochemical performances using cells with the interlayer (C-SC) and without the interlayer (SC). The candle soot interlayer showed improvement in the capacity and cycle life. SC delivered a capacity of 874 mA h g−1 at 0.1C which reduced to 193 mA h g−1 at a current rate of 6.0C, whereas C-SC exhibited an impressive reversible capacity of 1182 and 486 mA h g−1 at current rates of 0.1 and 6.0C, respectively. Moreover, C-SC retained a capacity of 710 mA h g−1 at 1.0C with 94% capacity retention over 200 cycles due to the inbuilt interlayer over the electrode, which provides alternative sites for the adsorption of long-chain lithium polysulfides. Later, the electrochemical performance was studied with SC and C-SC high-sulfur loading (4.5 mg cm−2) electrodes. During long-term cyclic stability testing, C-SC was able to retain 61.3% of the initial capacity after 150 cycles at 2.0C, while SC was able to retain only 16.9% of the initial capacity, reflecting the potential of the inbuilt interlayer. Furthermore, the experimental results were supported with first principles simulation studies, allowing in-depth understanding of the adsorption of higher order polysulfides over the carbon-based candle soot material.
Furthermore, BET surface analysis is performed to confirm the sulfur incorporation inside the carbon host by analyzing the changes in the surface area, pore volume and pore size of candle soot and SC. The N2 sorption isotherm (Fig. 2(a) and (b)) shows type III hysteresis loops for both the samples, revealing unrestricted monolayer formation during the adsorption and desorption process.44 The specific surface area, pore volume and average pore size of candle soot were found to be 341 m2 g−1, 0.872 cm3 g−1 and 5.92 nm, respectively. The specific surface area and pore volume of SC decreased to 12 m2 g−1 and 0.092 cm3 g−1, respectively, and the average pore size increased to 30.41 nm, clearly indicating the successful incorporation of sulfur in the carbon pores after the melt diffusion process.
To calculate the sulfur content in the candle soot–sulfur composites, TGA of candle soot and SC was carried out in a nitrogen atmosphere at a ramp rate of 10 °C min−1 (Fig. 2(c)). The complete weight loss of candle soot was found to be around 650 °C, indicating the purity of the collected soot. SC exhibited a weight loss of ∼70 wt% at ∼250 °C, which is ascribed to the decomposition of sulfur. The second weight loss of 30 wt% at ∼650 °C can be attributed to the decomposition of candle soot present in SC. This is in good agreement with the ratio of candle soot and sulfur employed during melt diffusion and therefore ensures the effectiveness of the synthesis approach. Later on, an adsorption test was employed to visually investigate the polysulfide adsorption capability of candle soot carbon (Fig. 2(d)). A glass vial was filled with DOL/DME solvent (1:1 v/v) containing polysulfides (Li2S6) followed by the addition of candle soot nanoparticles. The perturbation in the solution was observed for 48 h. The polysulfides were completely adsorbed by the candle soot, resulting in a clear and transparent solution after 48 h, indicating that the polysulfide diffusion can be suppressed by using candle soot as an inbuilt interlayer. This prompted us to use candle soot nanoparticles as a sulfur host to prepare a cathode as well as an interlayer, and to further investigate its electrochemical performance in Li–S batteries.
The redox reaction of sulfur in SC and C-SC was investigated through cyclic voltammetry (CV) in the potential window of 1.7 to 3 V (Fig. 3(a)). In the cathodic scan, the peak located at ∼2.4 V corresponds to reduction of elemental sulfur, which results in formation of long-chain lithium polysulfides (Li2Sx, 4 ≤ x ≤ 8) that are soluble in the electrolyte. Another peak located at ∼2.0 V corresponds to the formation of short-chain lithium polysulfides (Li2S2/Li2S) which are insoluble. In the reverse anodic scan, insoluble lithium polysulfides are converted to soluble polysulfides and then oxidized back to elemental sulfur (S8).45 The repeatability of the redox peak with high current in all cycles indicates the reversible lithium ion storage in sulfur (Fig. S3, ESI†).20 The similarity and reproducibility of the CV curves of C-SC with greater electrochemical surface area compared to the CV curve of SC demonstrate that the candle soot inbuilt interlayer allowed a stable electrochemical reaction by nullifying the shuttle behavior, thus preventing the loss of active material. The greater electrochemical surface area with higher current in CV is an indication of the capacity enhancement. In order to further confirm that, galvanostatic charge–discharge (GCD) measurements of SC and C-SC were carried out in the potential window of 1.7 to 3V. Fig. 3(b) illustrates the GCD profiles of SC and C-SC carried out at a current rate of 0.2C. The profiles exhibited two well defined plateaus during discharge which correspond to the formation of long-chain lithium polysulfides and the subsequent reduction to short-chain lithium polysulfides, while the appearance of a single plateau during charging indicates oxidation of lower order polysulfides to sulfur (S8),6,12,13 which are consistent with the CV studies. The initial reversible capacity of SC was found to be 665 mA h g−1 with a coulombic efficiency of 86.7%, whereas C-SC delivered an initial reversible capacity of 1014 mA h g−1 with a coulombic efficiency of 99.4% at 0.2C. This capacity enhancement along with excellent coulombic efficiency highlights the role of the inbuilt candle soot interlayer in trapping the polysulfides as well as efficient utilization of the active material (sulfur).
To get insight on the enhanced electrochemical performance, electrochemical impedance spectroscopy (EIS) was performed on both the SC and C-SC electrodes (before cycling in Fig. 3(c) and after cycling in Fig. 3(d)) in the frequency range of 0.01 Hz to 1 MHz. Both the electrodes (SC and C-SC) follow similar trends, displaying a semi-circle in the high frequency region, which stands for the charge transfer resistance, and a nearly straight line in the low frequency region, corresponding to the lithium-ion diffusion resistance within the electrode. The smaller semicircle and Warburg line of C-SC show that the charge transfer resistance and lithium ion diffusion resistance in this electrode are minimal when compared to SC.46 This is due to the candle soot inbuilt interlayer acting as an upper current collector and providing good electrical contact with the insulating sulfur, resulting in electron mobility with less resistance.
Further, the electrochemical behaviors of the lithium sulfur battery without the interlayer (SC) and with the interlayer (C-SC) are scrutinized at various current rates. SC delivered reversible capacities of 874, 665, 508, 398, 286 and 193 mA h g−1, whereas C-SC delivered capacities of 1182, 1014, 797, 661, 529 and 486 mA h g−1 at current rates of 0.1, 0.2, 1.0, 2.0, 4.0 and 6.0C, respectively (Fig. 3(e) and (f)). This capacity enhancement from 193 to 486 mA h g−1 at high current rate 6C substantiates the predominance of the interlayer for a faster electrochemical reaction.
Another major parameter in Li–S batteries is the capacity retention at extreme current rates (high & low). This was evaluated by performing rate capability studies. The rate capability test will provide complete information regarding the battery storage capacity from a real application point of view (Fig. 4(a)). SC and C-SC delivered a capacity of ∼850 mA h g−1 and ∼1200 mA h g−1 at 0.1C, respectively. It is to be noted that C-SC retained ∼1200 mA h g−1 after cycling up to 6C when it is switched back to 0.1C but SC failed to retain and the capacity fell to ∼600 mA h g−1 (Fig. 4a). These results further prove that the candle soot inbuilt interlayer played an excellent role in blocking polysulfides and preventing their dissolution into the electrolyte. In addition, it also reveals the compatibility of the candle soot interlayer for practical usage of the battery. Even at a high current rate this interlayer showed its effective adsorption capability towards long-chain lithium polysulfides.
To quantify the cycle life, a long cycling stability test was performed on SC and C-SC at 1.0C as depicted in Fig. 4b. SC could retain a capacity of 382 mA h g−1 after 100 cycles with a coulombic efficiency of 88%. However, when the electrode was modified with the candle soot inbuilt interlayer (C-SC), the cell exhibited an impressive capacity of 700 mA h g−1 with 94% capacity retention even after 200 continuous charge/discharge cycles, exhibiting an extremely low capacity decay rate of 0.03% per cycle. This further revealed the role of candle soot carbon as an inbuilt interlayer in enhancing the electrochemical performance. Later on, the cells were de-crimped post-cycling to visualize the separator condition. Digital photographs of the glass microfiber filter used as a separator for the SC and C-SC samples after cycling are presented in Fig. S4 (ESI†). The separator employed with the SC electrode shows a yellowish color due to polysulfide migration through the separator. In the case of C-SC, even after 200 cycles the separator maintains its structural integrity, unlike the SC sample, which showed dark spots.
Furthermore, the shuttle factor (f) was used to evaluate the extent of the shuttle effect for the SC and C-SC electrodes. The coulombic efficiency and shuttle factor are related according to the formula:47, where Ceff is the coulombic efficiency and f is the shuttle factor. The iterative calculation for the shuttle factor was carried out using the Newton–Raphson method (Matlab R2019). Fig. 4(c) depicts the plot of the shuttle factor with the cycle number (corresponding to the cyclic stability) and Fig. 4(d) depicts the plot of the shuttle factor at various current rates (corresponding to the rate performance). The SC electrode showed an increase in the shuttle factor with cycling and maintained an average shuttle factor value of 0.63, while the C-SC electrode showed decreased shuttling and maintained a significantly low average value of 0.14. This also indicated that there was a drastic reduction in polysulfide shuttling due to the candle soot inbuilt interlayer (Fig. 4(c)) and this was found to be consistent with the cyclic stability (Fig. 4(b)). The shuttle factor reduces with an increase in the current rate (C-rate) for C-SC as compared to SC (Fig. 4(d)). The shuttle factor study also suggested that the use of candle soot played a significant role in trapping the polysulfides within the cathode and hence prevented the loss of active material.
Meanwhile, to quantify the capability of the inbuilt interlayer, the electrochemical performance was tested using a high sulfur areal loading of 4.5 mg cm−2 in the SC and C-SC electrodes. The CV profiles (Fig. 5(a)) of SC and C-SC revealed two peaks during the cathodic scan ascribed to long-chain lithium polysulfides followed by subsequent formation of short-chain lithium polysulfides, while one oxidation peak during the anodic scan for oxidation of short-chain polysulfides to S8. Meanwhile, the area under the CV curve of C-SC was significantly higher, which may be related to improved charge-storage capability (Fig. 5(a)). Fig. 5(b) illustrates the GCD profiles of C-SC and SC measured at 0.1C. The discharge profile exhibited two plateaus ascribed to formation of long-chain lithium-polysulfides followed by short-chain lithium-polysulfides. However, during charging the profile exhibited a single plateau ascribed to oxidation of short-chain lithium polysulfides to S8. The SC electrode exhibited a reversible capacity of 839 mA h g−1 with a coulombic efficiency of 58.1% while C-SC delivered a capacity of 1013 mA h g−1 with a coulombic efficiency of 98.9%. The improvement in capacity can be ascribed to the nullifying effect of the inbuilt interlayer. Later, the long-term cyclic stability was tested using Li–S cells with SC and C-SC electrodes at 2C for 150 cycles (Fig. 5(c)). Interestingly, the cell with the C-SC electrode delivered excellent cycling for 150 cycles with an initial reversible capacity of 411 mA h g−1 and capacity retention of 61.3%, while the cell with SC was able to retain 15.9% of the initial capacity.
In order to understand the interactions between long-chain lithium polysulfides and the candle soot inbuilt carbon interlayer (C), first-principles calculations based on density functional theory (DFT) were performed. The schematic (Fig. 6(a)) represents the binding energy calculation between polysulfide Li2S8 and carbon. The binding energy of carbon with polysulfides Li2S4, Li2S6, and Li2S8 is −78.76, −52.51, and −26.25 kJ mol−1, respectively. Here, we observed promising results in the interaction energies of carbon with lithium polysulfides. This indicated the strong interlayer properties of the carbon-based material (candle soot) towards soluble polysulfides. The calculated result also demonstrated that the interaction between the polysulfides and carbon layer is thermodynamically favorable and more stable. Therefore, the carbon-based candle soot inbuilt interlayer not only acted as a physical trapper but also chemically anchored higher order lithium polysulfides.
Further, we investigated the possible interaction of an adsorbed polysulfide carbon layer (e.g., C–Li2S4) with other soluble polysulfides (e.g., Li2S6 and Li2S8). Fig. 6(b)–(d) represent the calculated possible interaction energies of polysulfide carbon complexes with other polysulfides and the exact possible set of reactions is shown in eqn (S1) (ESI†). Three cases of interactions were considered for each polysulfide. For example, the interaction of the Li2S4 polysulfide can be considered as: (i) Li2S4 interaction with only the carbon layer, (ii) Li2S4 polysulfide interaction with the Li2S6 polysulfide adsorbed carbon layer, and (iii) Li2S4 polysulfide interaction with the Li2S8 polysulfide adsorbed carbon layer. Similarly, other interactions were considered for the remaining polysulfides Li2S6 and Li2S8.
It is interesting to note that the interaction energy of polysulfide Li2S4 with the bare carbon layer is very low (−78.76 kJ mol−1) compared to that of the polysulfide adsorbed carbon layer (−315.06 and −262.55 kJ mol−1 for C–Li2S6 and C–Li2S8, respectively). Similar trends are observed with the other polysulfides (Li2S6 and Li2S8). Thus, this computational study thermodynamically proved that the pristine carbon layer possessed adsorption tendencies with long-chain lithium polysulfides. However, the carbon layer with adsorbed polysulfides acted as a more effective interlayer and can further lower the extent of polysulfide shuttling in Li–S batteries.
This work demonstrates the excellent stability of the electrode in long-term cycling with interlayer modification and also stands ahead in comparison with other forms of carbon–sulfur composites and interlayer modified cells as reported previously, given in Table 1.
Cathode | Sulfur content (%) | Interlayer | Initial capacity (mA h g−1) | Capacity retentiona (%) | Cycle number | C-Rate | Ref. |
---|---|---|---|---|---|---|---|
a Capacity retention = (reversible capacity of last cycle/reversible capacity of first cycle) × 100. | |||||||
rGO coated hollow yeast carbon–sulfur composite | 50.3 | — | 1000 | 65 | 200 | 0.1 | 15 |
Cherry pit carbon–sulfur composite | 40.2 | — | 550 | 75 | 200 | 0.1 | 17 |
Silk cocoon carbon–sulfur composite | 48.4 | — | 1300 | 62 | 80 | 0.5 | 19 |
Hair derived carbon–sulfur composite | 69.0 | — | 1113 | 89 | 300 | 0.2 | 46 |
Banana peel carbon–sulfur composite | 60.0 | — | 600 | 67 | 250 | 1.0 | 20 |
Fern carbon–sulfur composite | 66.4 | — | 1377 | 55 | 100 | 0.2 | 48 |
Ni, S co-doped rice popcorn carbon–sulfur composite | 76.1 | — | 1256 | 65 | 500 | 0.2 | 23 |
Studies with interlayer | |||||||
Bare sulfur | 60.0 | Luffa sponge derived carbon | 1000 | 80 | 500 | 2.0 | 20 |
Bare sulfur | 70.0 | Bamboo char derived carbon | 813 | 74 | 300 | 1.0 | 22 |
Bare sulfur | 70.0 | Polyacrylonitrile spun CNF | 1134 | 41 | 200 | 0.2 | 49 |
Bare sulfur | 60.0 | MoO3 decorated CNF | 1142 | 53 | 500 | 0.2 | 50 |
Bare sulfur | 70.0 | Modified carbon paper | 1100 | 58 | 200 | 0.2 | 51 |
rGO–sulfur composite | 70.0 | Co–Fe bimetallic sulfide decorated on carbon paper | 1125 | 56 | 400 | 0.2 | 52 |
CNT–sulfur composite | 75.0 | MnO2–GO–CNT composite | 813 | 80 | 200 | 0.5 | 53 |
Candle soot carbon–sulfur composite (SC) | 69.3 | Candle soot inbuilt interlayer | 710 | 94 | 200 | 1.0 | Present study |
The improvement in the electrochemical performance is mainly attributed to the confinement of higher order polysulfides within the C-SC electrode by the development of the inbuilt candle soot interlayer which minimizes polysulfide dissolution into the electrolyte and gives effective utilization of active material (sulfur).
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1ma00115a |
This journal is © The Royal Society of Chemistry 2021 |