Spherical Macroporous Carbon Nanotube Particles with Ultrahigh

Jan 4, 2018 - The S-M-CNTP cathode shows a highly reversible capacity of 1343 mA h g–1 at a current density of 0.2 C even at a high sulfur content o...
0 downloads 0 Views 4MB Size
Download PDF
www.acsnano.org

Spherical Macroporous Carbon Nanotube Particles with Ultrahigh Sulfur Loading for Lithium−Sulfur Battery Cathodes Donghee Gueon,† Jeong Tae Hwang,† Seung Bo Yang,‡ Eunkyung Cho,‡ Kwonnam Sohn,‡ Doo-Kyung Yang,‡ and Jun Hyuk Moon*,† †

Department of Chemical and Biomolecular Engineering, Sogang University, Baekbeom-ro 35, Mapo-gu, Seoul 04107, Republic of Korea ‡ LG Chem Research Park, Moonji-ro 188, Yuseong-gu, Daejeon 34122, Republic of Korea S Supporting Information *

ABSTRACT: A carbon host capable of effective and uniform sulfur loading is the key for lithium−sulfur batteries (LSBs). Despite the application of porous carbon materials of various morphologies, the carbon hosts capable of uniformly impregnating highly active sulfur is still challenging. To address this issue, we demonstrate a hierarchical pore-structured CNT particle host containing spherical macropores of several hundred nanometers. The macropore CNT particles (M-CNTPs) are prepared by drying the aerosol droplets in which CNTs and polymer particles are dispersed. The spherical macropore greatly improves the penetration of sulfur into the carbon host in the melt diffusion of sulfur. In addition, the formation of macropores greatly develops the volume of the micropore between CNT strands. As a result, we uniformly impregnate 70 wt % sulfur without sulfur residue. The S-M-CNTP cathode shows a highly reversible capacity of 1343 mA h g−1 at a current density of 0.2 C even at a high sulfur content of 70 wt %. Upon a 10-fold current density increase, a high capacity retention of 74% is observed. These cathodes have a higher sulfur content than those of conventional CNT hosts but nevertheless exhibit excellent performance. Our CNTPs and pore control technology will advance the commercialization of CNT hosts for LSBs. KEYWORDS: CNT particles, lithium sulfur batteries, sulfur infiltration, hierarchical pores, high sulfur contents, sulfur residue performance.3,13−16 Recently, graphene or carbon nanotubes (CNT) have demonstrated advantages because they have a high surface area and electrical conductivity and form an openpore structure.8,17−20 In particular, the highly entangled structure of the CNT not only provides highly electrical conductive properties but also provides fully connected network pores.21−24 The macropores provide sufficient porosity for electrolyte/ion diffusion even after loading with sulfur. Recently, various CNT-based hosts, such as free-standing CNT films,7 vertically aligned flexible CNTs,25 and mesoporous CNTs,6 have been applied to LSBs. Meanwhile, sulfur has been impregnated or composited with carbon hosts using melt diffusion,26 chemical precipitation,27,28 and vapor infusion.29,30 Among these methods, melt diffusion has been widely applied.10,15,31,32 This method involves the injection of sulfur into the carbon host by capillary action or

L

ithium−sulfur batteries (LSBs), a secondary battery using elemental sulfur as the main cathode material, is a promising candidate to overcome the energy storage capacity limitations of existing lithium-ion batteries (LIBs) for next-generation electric vehicles (EVs) and hybrid EVs (HEVs).1,2 For example, EVs require more than twice the cell-level energy density of current LIBs, i.e., 400 Wh kg−1.3,4 The theoretical energy density of a LSB is approximately 2600 Wh kg−1 based on the theoretical capacity of 1675 mAh g−1, which is sufficient to meet these requirements.3,5 Sulfur is inexpensive, abundant, and environmentally nontoxic, but sulfur as an electrode material must overcome low electrical conductivity at the insulator level.6,7 Most studies utilize sulfur impregnated into highly electrically conductive carbon hosts.8−12 Conventional, high pore volume, and surface area mesoporous or microporous carbon materials have been widely used because they can hold large amounts of sulfur.13−16 However, after sulfur impregnation, the pores were easily clogged, resulting in deterioration of electrolyte penetration and consequent lithium ion diffusion, resulting in poor cell © XXXX American Chemical Society

Received: August 17, 2017 Accepted: December 27, 2017

A

DOI: 10.1021/acsnano.7b05869 ACS Nano XXXX, XXX, XXX−XXX

Article

Cite This: ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano evaporation at a temperature above the melting point of sulfur (usually 155 °C). To achieve higher energy density than current LIBs, 70 wt % of active sulfur should be incorporated into the carbon host.33 However, many studies using CNT hosts as well as various porous carbon hosts have typically applied only 40−60 wt % sulfur composite,7,8,19,25 which is lower than this requirement. This may be due to the difficulty of uniform impregnation of sulfur to the carbon host under high sulfur loading conditions. We were able to identify sulfur chunks that did not penetrate into the CNT host when the sulfur content exceeded 60 wt %.8,17 We also observed similar sulfur residues when 70 wt % sulfur was injected into the conventional CNT film. This residue may not be activated upon charge and discharge of the LSB and leads to a lower capacity per sulfur mass. Therefore, in order to obtain the performance of the LSB with substantially higher energy density, uniform sulfur impregnation at high sulfur loading conditions should be achieved. In this study, we introduce a hierarchical pore-structured CNT particle (CNTP) host containing spherical macropores of several hundred nanometers. The spherical macropore was formed using a polymer sphere template of several hundred nanometers in the CNT assembly. Here, we observed that the penetration of sulfur is limited by the morphology characteristics (curvature) of CNTs and is significantly improved by the introduction of spherical macropores. We also observed a significant increase in micropore volume due to loss of aggregation in the CNT assembly with a polymer sphere. As a result, we were able to uniformly impregnate more than 70 wt % of sulfur into the CNT host without residue. The LSB with this high sulfur-loaded macroporous CNTP (M-CNTP) cathode exhibited a high reversible capacity of 1034 mAh/g at a current density of 2 C and capacity retention of 71% over 100 cycles, even at a high content of 70 wt %. This result is superior to the performance of LSB cathodes using previously reported CNT-based hosts. Our results show that LSB performance can be improved through microstructure control of commercial CNTs, which is noteworthy from a commercialization point of view for the CNT host.

Figure 1. (Top) Schematic diagram for preparation of M-CNTP. (Bottom) (a) Low- and (b) high-magnification SEM images of the M-CNTPs. Inset of (a) shows 20 g of the M-CNTPs in a 100 mL vial produced on a laboratory scale via the spray drying process. (c) Magnified SEM image of the M-CNTP surface. (d) Cross-sectional SEM image of a cracked M-CNTP. Inset of (d) shows a magnified SEM image of the cracked M-CNTP (scale bar: 1 μm). (e) Digital images of the powder-dispersed slurry with the same CNT and MCNTP weights. (f) Cross-sectional SEM image of the M-CNTP electrode film on a C-coated Al substrate with a 3 mg/cm2 active material mass loading.

RESULTS AND DISCUSSION The M-CNTPs are produced by rapidly drying an aerosol of the CNT and polymer sphere (PS)-dispersed solution (140 °C), followed by a post heat treatment (500 °C), as described in Figure 1. The dried droplets of the dispersion solution form a tangled CNT structure while maintaining a spherical shape, and the high-temperature heat treatment removes the PSs resulting in spherical macropores. Complete removal of PS was confirmed by thermogravimetric analysis (TGA), as observed in Figure S1. The production rate of M-CNTPs is greater than 10 g/h (see the inset of Figure 1a); the production rate can be easily scaled up using multijet. The scanning electron microscopy (SEM) image of the M-CNTPs is shown in Figure 1a,b. The sizes of the CNTPs are in the range of 3−8 μm. The entangled structure of CNTs on the particle surfaces is confirmed, as observed in Figure 1c. In the internal image of the M-CNTPs observed by fracturing a particle (Figure 1d) reveals the removal of the PSs and the formation of macropores of approximately 700 nm. The CNTP morphology is advantageous for producing a thick film based on the manufacturing process of the electrode host film. Thick host films are essential to achieve high loadings of active sulfur for LSBs. For this purpose, casting a solution

containing a high concentration of dispersed carbon host material is required. Unlike the CNT dispersion process, the CNTPs can be dispersed without entanglement, easily producing a high-concentration dispersion. As shown in Figure 1e, the CNTP dispersion flows well even at a high concentration of 14 wt %, but CNTs become a gel due to the entanglement of CNT strands. Figure 1f shows a thick carbon host film with a thickness of 27 μm produced by casting the 14 wt % CNTP dispersion. The CNTP film exhibits a closed packing of particles and possesses well-defined pores around the particles. Loading sulfur onto the CNTP host for a LSB cathode was performed using a melt-diffusion method. First, we compared the diffusion of sulfur in the PS-templated macroporous CNT films to that in CNT films. A molten sulfur droplet was dropped onto the surface of the sample film, and the diffusion of sulfur and the contact angle with the substrate were measured. (The substrate was maintained at 155 °C as observed in Figure 2a.) Compared to the CNT host, molten sulfur rapidly and uniformly spread on the CNTP film with spherical macropores, penetrating the inside of the host. In addition, the contact angle of sulfur droplets on the B

DOI: 10.1021/acsnano.7b05869 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

Figure 2. Digital images of melted sulfur after 5 min on a 155 °C hot plate and a sulfur droplet on the (a) macroporous CNT assembly (arrow indicates the spreading boundary of the sulfur) and (c) CNT assembly film. SEM images of the (b) S-M-CNTPs and (d) S-CNTPs. The insets in (b) and (d) are the magnified images of the respective samples (scale bar: 1 μm).

macroporous CNTP film is 10° or less, which is much smaller than that of bare CNT films, displaying contact angles of approximately 35°. The uniform and excellent penetration of sulfur in the M-CNTPs can be explained as follows. Sulfur has a relatively high interfacial tension of 55 dyn/cm,34 and the surface of the wet CNT strand has negative curvature. Thus, molten sulfur is difficult to wet on the CNTs. As shown in Figure S2, sulfur is coated in the form of dewetted particles around the CNT strands. In contrast, when spherical macropores are present, capillary action by the large pores, which have positive curvatures, promotes sulfur penetration into the interior. Figure S3 shows transmission electron microscopy (TEM) images before and after sulfur injection to confirm the presence of sulfur inside the pores. The uniform sulfur penetration greatly reduces the sulfur residue on the surface of the CNTP host. In Figure 2b,d, no sulfur residue on the surface of the M-CNTPs after the melt diffusion of sulfur is observed, but the bare CNTPs show bulky sulfur residue. We present the spatial distribution of sulfur in the surface and cross-sectional images of S-M-CNTP in Figure S4 and confirm the impregnation of S uniformly. In addition, in the low magnification SEM image, uniform sulfur penetration was observed on the entire S-M-CNTP film, as shown in Figure S5. We observed that sulfur penetrated into M-CNTP with no residue up to 80 wt % (see Figure S6). The sulfur residues in the sulfur-loaded M-CNTPs (S-MCNTPs) and sulfur-loaded CNTPs (S-CNTPs) were further investigated using Raman spectroscopy, as observed in Figure 3a. The peaks at 147, 188, 215, 250, 434, and 470 cm−1

correspond to the S−S bonds.35 The broad peaks at 1350 and 1585 cm−1 are attributed to the D- and G-bands of carbon, respectively.17,36,37 The S-CNTPs show noticeable sulfur peaks compared to the S-M-CNTPs, confirming the presence of a large amount of sulfur on the CNTP surface and that all the sulfur in the M-CNTP sample penetrates the CNTPs.38 The TGA on both samples enables the quantitative determination of the sulfur content in CNTP and M-CNTP (see Figure 3b). Sulfur begins to evaporate at 290−320 °C. Considering the weight loss above this temperature, the sulfur content of both samples is approximately 70 wt %. The TGA results show that M-CNTP shows a decrease in sulfur mass at higher temperatures than CNTP. As reported in previous studies, this result can be explained by the fact that sulfur in M-CNTP is entrapped in the micropores or uniformly contacted with the carbon surface, resulting in the formation of relatively stable sulfur.39 Furthermore, electrochemical impedance spectroscopy (EIS) analysis also reveals the lower charge-transfer resistance of the S-M-CNTPs compared to the S-CNTPs, which is evidence of the uniform sulfur coating and excellent electrolyte penetration (see Figure S7).40,41 Pore size analysis of the M-CNTPs and CNTPs was performed; pores less than 50 nm were evidenced by the Brunauer−Emmett−Teller (BET) method, and macropores formed by the PSs were investigated using mercury intrusion porosimetry. The M-CNTPs and CNTPs show sharp increases in their BET adsorption curves near a relative pressure of 1.0, indicating the presence of meso- and macropores,6 as observed in Figure 4a,b. The specific surface area of the M-CNTPs is 238 m2 g−1, which is approximately two times higher than the specific surface area of the CNTPs (118 m2 g−1). The pore size distributions using the Barrett−Joyner−Halenda (BJH) method are shown in Figure 4c,d. Both the M-CNTPs and CNTPs have pores of approximately 4 and 40 nm in size, corresponding to the pores between adjacent CNTs and the pores between CNT bundles, respectively, as previously described.6 Note that the 40 nm macropores are not derived from the PS templates. We observed the presence of pores of hundreds of nanometers by the mercury intrusion method (see Figure S8). The MCNTPs possessed approximately 10 times more 4 nm pores than the CNTPs. This may be due to the increased formation of pores due to loosely agglomerated CNT strands when the PSs are incorporated and/or the increased accessibility of the internal micropores through the PS-templated macropores.

Figure 3. (a) Raman spectra of sulfur, the S-CNTPs, the S-MCNTPs, and the CNTs. b) TGA curves of sulfur, the S-CNTPs, and the S-M-CNTPs in a nitrogen atmosphere. C

DOI: 10.1021/acsnano.7b05869 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

CNTPs have a high utilization of sulfur, whereas the S-CNTPs display sulfur loss.4,17 This result corresponds to the presence of sulfur residue in the S-CNTPs. The galvanostatic charge−discharge profile of the S-MCNTP cathode was measured at various rates under the voltage window of 1.5−2.8 V versus Li+/Li, as displayed in Figure 5a.

Figure 4. (a) Nitrogen adsorption/desorption isotherms of the MCNTPs and S-M-CNTPs. (b) Nitrogen adsorption/desorption isotherms of the CNTPs and S-CNTPs. (c) Pore size distributions of the M-CNTPs and S-M-CNTPs. (d) Pore size distributions of the CNTPs and S-CNTPs.

Figure 5. (a) Charge/discharge voltage profiles of the S-M-CNTP cathode at various C rates from 0.2 to 2 C. (b) Rate capabilities of the S-CNTP cathode and S-M-CNTP cathode at various current densities from 0.2 to 2 C. Capacity contributions of high-order polysulfide conversion (Q1) and low-order polysulfide conversion (Q2) and the Q2/Q1 ratios at various C rates for the (c) S-CNTP and (d) S-M-CNTP cathodes.

After incorporating sulfur, the BET surface areas of the MCNTPs and CNTPs were reduced to 18 and 10 m2 g−1, respectively. In addition, the BJH pore size distribution results show that the approximately 4 nm peaks almost disappeared in both samples, as observed in Figure 4c,d. The larger 40 nm pores only decreased by 50−70%, indicating that the adsorption of sulfur did not occur in the macropores but mainly in the micropores between the CNTs. As a result, the PS-templated macropores facilitate sulfur diffusion into the interior of the CNTPs and consequently promote adsorption into the internal CNT strands. Sulfur contained in the internal micropores is expected to prevent the elution of the polysulfide intermediates into the electrolyte during charging and discharging. Additionally, the macropores existing after sulfur loading can provide sufficient space to mitigate sulfur volumetric changes and accelerate ion penetration during LSB charging and discharging.42,43 The S-M-CNTPs were evaluated as a cathode of a LSB. The S-CNTP cathode was also tested for comparison. The mass loading of the active material in all samples was 1 mg cm−2. The specific capacity in our results is based on the mass of the pure sulfur content, excluding the CNTs, conductors and binders. First, cyclic voltammetry (CV) of the S-CNTP and S-M-CNTP electrode cells was performed over 10 cycles at a scan rate of 10 μV s−1 cycling between 1.5 and 2.8 V versus Li/Li+. Note that we intentionally used a very slow scan rate of 10 μV s−1 to facilitate higher utilization of sulfur and allow sufficient time for polysulfide shuttling.4 As shown in Figure S9a,b, two reduction peaks and one oxidation peak are observed. According to the multistep redox reaction of the sulfur cathode during discharging/charging, the peak at 2.3 V represents the reduction of elemental sulfur to higher order Li polysulfides (Li2Sn, 4 ≤ n ≤ 8), and the peak at 1.95 V indicates reduction to Li2S2 and Li2S. In the anodic scan, the peak at approximately 2.55 V is attributed to the complete conversion of Li2S and polysulfides into elemental sulfur.4,25,44,45 Unlike the S-CNTPs, the S-M-CNTPs show a stable cyclic voltammogram; the redox peaks overlap during the first 10 cycles, indicating that the S-M-

These profiles show the first discharge/charge curves at each Crate. Under low current conditions of 0.2 C, the S-M-CNTP electrode exhibits a discharge capacity of 1343 mA h g−1, and the S-CNTP electrode exhibits a capacity of 755 mA h g−1. Additionally, under all current conditions, the S-M-CNTPs display a higher capacity than S-CNTP electrode. As a result, the S-M-CNTPs exhibit a capacity retention rate of 74% more than a 10-fold current density increase, which is much higher than the retention rate of 26% for the S-CNTPs. Despite our host possessing a very high sulfur content of 70 wt %, the retention rate is higher than the rates of other graphene- or CNT-based carbon hosts, which were applied with 45−60 wt % sulfur impregnation (see Table S1). In addition, we investigated the discharge process in detail. In the voltage−capacity profile, a two-stage voltage plateau is observed.4,17,46 The first plateau appears at approximately 2.3 V, which corresponds to the reduction of elemental S8 to higher order polysulfides (Li2S84−). The overall reaction at this stage is S8 + 4Li+ + 4e− → 2Li2S4. The plateau at 2.1 V corresponds to the production of Li2S2 and Li2S through association with additional lithium ions, i.e., 2Li2S4 + 12Li+ + 12e− → 8Li2S.25,45 Ideally, the capacity corresponding to the discharge of the second stage is three times higher than the capacity of the first stage. However, in the second stage, the reaction is relatively sluggish due to solid-state diffusion, and the loss of higher order polysulfides cannot be avoided. As a result, the capacity ratio at each stage is usually less than 3.47 Figure 5c,d shows the total specific capacity, the capacity contributions in the first and second discharge plateaus and the ratio of the capacity in the second stage to the capacity in the first stage. As shown in Figure 5c,d, the S-M-CNTP electrode maintains a high value of 2.42−1.90, whereas the SCNTP electrode is greatly reduced from 1.49 to 0.75 with an D

DOI: 10.1021/acsnano.7b05869 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

Figure 6. (a) Cycling performance of the S-M-CNTP cathode at 0.5 and 2 C rates and the S-CNTP cathode at a 2 C rate and their corresponding Coulombic efficiencies. Capacity contributions of high-order polysulfide conversion (Q1) and low-order polysulfide conversion (Q2) and the Q2/Q1 ratios over 100 lithiation/delithiation cycles for the (b) S-CNTP and )c) S-M-CNTP cathodes.

increasing current. This result confirms the high utilization of uniformly coated sulfur in the S-M-CNTPs and reflects the easy penetration and diffusion of the electrolyte into the interior via the macropores of the M-CNTP host.27 The cycling performances of the S-M-CNTPs and S-CNTPs are shown in Figure 6a. The Coulombic efficiency in each charge/discharge cycle was also plotted. The mass loading of the active material was 1 mg cm−2. The capacity values are calculated based on the sulfur mass. An initial capacity at 0.5 C (discharging or charging the full theoretical capacity in 2 h) of the S-M-CNTPs delivered 1544 and 901 mA h g−1at 100 cycles. The capacity at 100 cycles of the M-CNTP host is superior to those of previous CNT-based hosts, even though a very high sulfur content of 70 wt % is applied and measured at high currents (see Table S1). By contrast, the initial capacity of the S-M-CNTPs and S-CNTPs at 2 C was 1034 and 414 mA h g−1, and the capacity retention at 100 cycles was 71% and 60%, respectively. We observed capacity retention from 4 C to 200 cycles, and the retention of S-M-CNTPs and S-CNTPs were 77% and 40%, respectively (see Figure S10). Therefore, the SM-CNTP electrode clearly exhibits improved capacity retention. The Coulombic efficiency of the S-M-CNTPs remained almost 100% up to 100 cycles, but the efficiency was only 96.7% for the S-CNTPs. The Coulombic efficiency of the S-M-CNTPs remained almost 100% up to 100 cycles, but the efficiency was only 96.7% for the S-CNTPs. Meanwhile, we measured the performance by loading at a level of 4 mg/cm2, as observed in Figure S11. The result showed the initial capacity of 1400 mAh/g and the capacity of 634 mAh/g in 100 cycles at 0.5 C rate. Although there is a decrease in performance at high loading, it is still better than the results of previous literature, considering the 0.5 C rate condition.25,31,48

The S-M-CNTPs showed excellent specific capacity, reversibility, and retention compared to the S-CNTPs. To analyze the excellent capacity characteristics of the M-CNTP host, we identified the capacity of each of the two-stage plateaus and the ratio of these two plateaus in specific cycles, as observed in Figure 6b,c. The S-M-CNTPs maintained a high ratio of 2.4 at 100 cycles. By contrast, the S-CNTPs showed a very low ratio (0.8−0.9) in the second cycle and then recovered to 1.63. The S-M-CNTPs maintained a stable contribution to the second-stage capacity, which involved a relatively sluggish reaction, indicating the sufficient reaction of polysulfides and effective utilization of sulfur during charge and discharge. Therefore, these results indicate that the high volume of micropores in the M-CNTPs play a role in reducing loss by confining sulfur and polysulfides, i.e., the intermediates, in the charge−discharge process. Briefly, the excellent electrochemical performance of the MCNTPs is due to the efficient pore structure of the M-CNTPs and uniform sulfur incorporation. The well-developed micropores of the M-CNTPs and the impregnated sulfur in these pores stably contain sulfur and its derivatives during charging and discharging, resulting in excellent capacity characteristics. In addition, the spherical macroporous structure delivers excellent rate characteristics by providing sufficient pores for electrolyte diffusion even after sulfur loading.42

CONCLUSIONS The impregnation of a high sulfur content is essential for high energy density LSBs. CNTs are a promising sulfur host with multiscale porous structures and high electrical conductivity. However, previous studies have only applied sulfur contents on the order of 40−60% and have frequently reported nonuniform E

DOI: 10.1021/acsnano.7b05869 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

determined from the desorption branches of the isotherms using the BJH method.

sulfur impregnation and low LSB capacity. To address these shortcomings, we fabricated CNTPs containing spherical macropores formed from a PS template. In the BET analysis, the PS template not only introduced macropores but also greatly developed micropores between the CNT strands. We observed that molten sulfur exhibits poor surface wetting properties on the CNT assembly and that wetting could be greatly improved by introducing macropores into the CNTs. The uniform sulfur-loaded M-CNTP cathode showed a high capacity of 1334 mA h g−1 at a current density of 0.2 C even at a high sulfur content of 70 wt %. Upon a 10-fold current density increase, a high capacity retention of 71% was observed. This study reports a higher sulfur content than studies using conventional CNT hosts but nevertheless report excellent performance. In practice, CNTs are a commercialized material, and our assembly and pore control technology will advance the commercialization of CNT hosts for LSBs.

ASSOCIATED CONTENT S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsnano.7b05869. Figures S1−S11 and Table S1 as described in the text (PDF)

AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]. ORCID

Jun Hyuk Moon: 0000-0002-4776-3115 Notes

The authors declare no competing financial interest.

EXPERIMENTAL SECTION Preparation of the CNTPs with Spherical Macropores. PSs with a uniform size (ca. 700 nm) were synthesized via dispersion copolymerization of styrene. An aqueous dispersion of CNTs (3 wt %, purchased from JEIO Co.) was used without further purification. The CNTs were multiwalled with an outer diameter of 20 nm and were used without further purification or treatment. A 3 wt % mixture of PSs and CNTs with an appropriate weight ratio was prepared in deionized (DI) water. The M-CNTPs were obtained using a spraydryer. The particle size could be modified by controlling various factors, including the concentration of the initial dispersion, the nozzle size, and the aerosol production pressure. The aqueous dispersion of the PS/CNT mixture was sprayed at 140 °C. The obtained PS/CNT spherical particles were heat-treated in a furnace under inert conditions at 500 °C for 2 h to remove the PS template. Characterization of Electrochemical Properties and LSB. We obtained a mixture of the active material (80 wt %), a carbon-based conductive agent (DB-100, 10 wt %), and CMC:SBR (3 wt %:7 wt %) as a binder. The electrode film was prepared by coating this mixture onto a carbon-coated Al foil current collector. The battery performance was evaluated using a Maccor 4300 test system. CR2032 cointype cells were assembled with Li metal foil as the counter electrode and a polypropylene membrane positioned between the two electrodes; the cells were assembled in a dry room with controlled humidity (