Facile Synthesis and Thermal Stability Enhancement of Sulfurized Polyacrylonitrile for Thermal Battery Cathodes

Article information

Abstract

1. INTRODUCTION

Thermal batteries, designed to operate at high temperatures (400–500 °C), are ideal for military applications due to their thermal stability, long shelf life, high output power, and robust environmental adaptability [1–3]. They are used in advanced weaponry and aerospace systems, including guided bombs, missiles, and torpedoes [4–7]. Recently, the increasing demand for high-power thermal batteries in advanced technology weapon systems has highlighted the need for cathode materials capable of superior output [8]. Although Pyrite FeS2 has been used as a conventional cathode material because of its abundance, stable discharge performance and good compatibility with electrolytes, its low capacity, low operating voltage, and poor thermal stability during the discharge has hampered its application in next-generation thermal batteries [9–11].

Sulfide-polyacrylonitrile (SPAN) is a promising cathode material, offering excellent cycling stability and efficient sulfur utilization [12]. In Li-S batteries [13], SPAN is notable for its lower polysulfide dissolution, higher conductivity, and superior compatibility with carbonate electrolytes compared to other sulfur-based cathodes [14]. Moreover, carbonized SPAN (c-SPAN), produced by thermally annealing sulfur and PAN in a nitrogen atmosphere, features a stable molecular structure due to covalent bonding between sulfur and the pyrolyzed PAN backbone[14,15]. This stability makes c-SPAN suitable for high-temperature applications exceeding 300 °C, like Na-S batteries [15,16]. While sulfide-polyacrylonitrile (SPAN) has been recognized as a promising cathode material for various battery applications, its potential has yet to be revealed in the context of thermal batteries. Thus, despite its potential, the application of c-SPAN in thermal batteries remains unexplored.

In this study, we successfully fabricated carbonized SPAN (c-SPAN) under optimized synthesis conditions, demonstrating its suitability as a cathode material for thermal batteries. The c-SPAN was synthesized using a straightforward approach involving ball milling and annealing. Its efficiency was confirmed through comprehensive structural analysis and thermal stability evaluation. The results showed that the c-SPAN had excellent thermal stability, remaining stable up to 500 °C under optimal conditions. We revealed that c-SPAN exhibits significantly improved thermal stability compared to conventional PAN, remaining stable up to 500°C with optimized sulfur content. Additionally, we demonstrated that controlling sulfur content is crucial to achieve enhanced thermal stability. The thermal battery incorporating c-SPAN operated successfully, indicating its potential applicability in thermal batteries. These results suggest that c-SPAN could serve as a promising cathode material for future thermal battery technologies.

2. EXPERIMENTAL

2.1. Materials Synthesis

Figure 1(a) illustrates the chemical structure of the raw materials[14] and the fabrication process of the c-SPAN materials. The c-SPAN materials were synthesized via a ball-milling process[17]. Initially, pure polyacrylonitrile (Sigma-Aldrich) and sulfur (99.5%, Alfa Aesar) were milled using either a mortar and pestle or a planetary mill (Pulverisette 7, FRITSCH, Germany) under a nitrogen (N₂) atmosphere. The resulting milled powder was then subjected to thermal treatment in a tubular furnace at 155 °C for 2 hours under an argon (Ar) atmosphere. Following this, the temperature was increased to 400 °C for 6 hours for a secondary thermal treatment to obtain c-SPAN. Additionally, a synthetic procedure identical to the one described above was used to synthesize c-SPAN treated at 350 °C under atmospheric pressure, and 400 °C under semi-atmospheric pressure, respectively. Here, atmospheric pressure and semi-atmospheric pressure refer to 1 atm and 0.5 atm, respectively.

Fig. 1.

(a) Schematic representation of the c-SPAN fabrication process. SEM images and EDS mapping data of c-SPAN samples prepared using (b) planetary ball milling and (c) mortar/pestle before annealing, and (d) planetary ball milling and (e) mortar/pestle after annealing.

3. RESULTS AND DISCUSSION

Figure 1(b)–(e) shows the SEM images and energy-dispersive X-ray spectroscopy (EDS) elemental mapping results of various samples prepared under different milling conditions (mortar/pestle and planetary ball mill). The SPAN synthesized via planetary ball milling exhibited particle sizes in the micrometer (μm) scale before thermal treatment, and EDS mapping revealed that the sulfur and carbon elements were agglomerated separately (Figure 1(b)). The SPAN synthesized using a mortar/pestle exhibited smaller particle sizes compared to the SPAN prepared via planetary ball milling, with an uneven elemental distribution similar to that of the planetary ball milling sample (Figure 1(c)). After thermal treatment at 400 °C, the particle size of the c-SPAN synthesized via planetary ball milling increased overall, and EDS mapping showed a uniform distribution of sulfur and carbon elements (Figure 1(d)). In contrast, the c-SPAN prepared using a mortar/pestle exhibited similar SEM morphology but relatively lower uniformity in sulfur and carbon distribution compared to the planetary ball milling sample (Figure 1(e)).

To assess the thermal stability of the various samples, thermogravimetric analysis (TGA) was conducted over a temperature range of 25–1000 °C (Figure 2). As shown in Figure 2(a), the untreated SPAN, regardless of the milling method, exhibited a significant weight loss of 77.5% around 200 °C, which is attributed to the sublimation of unreacted sulfur [16]. After thermal treatment, the c-SPAN synthesized via planetary ball milling demonstrated enhanced thermal stability up to 500 °C, with only about a 3.3% weight loss at 100 °C, primarily due to moisture evaporation (Figure 2(b)). This improved thermal stability can be attributed to the stable structure resulting from the chemical bonding of sulfur to the PAN-derived turbostratic carbon matrix [14,15]. Additionally, the bond dissociation energy of the C–S bond (272 kJ/mol) is higher than that of the S–S bond (251 kJ/mol), further contributing to the enhanced thermal stability [16].

Fig. 2.

(a) TGA results of c-SPAN samples before thermal treatment. TGA results of c-SPAN samples prepared using (b) planetary ball milling and (c) mortar/pestle after annealing.

The c-SPAN annealed at 400 °C showed a weight loss of 74.1% in the higher temperature range of 500–1000 °C, which can be attributed to the desorption of sulfur chemically bonded to the PAN-derived carbon matrix[16,18]. In contrast, the thermal stability of c-SPAN samples prepared using a mortar/pestle was relatively poor compared to those synthesized via planetary ball milling (Figure 2(c)). This could be due to the sublimation of sulfur that is unevenly distributed, or agglomerated within the carbon matrix, as indicated by the EDS mapping results.

Previous studies have indicated that increasing the sulfur content covalently bonded within the carbon framework is essential to enhance the electrode's capacity [14]. To achieve higher sulfur content in the samples, we adjusted the pressure and thermal treatment temperature of the c-SPAN samples. Figure 3(a) shows the SEM image of c-SPAN treated at 400 °C and atmospheric pressure (referred to as SPAN-400(AP)), revealing particle sizes in the micrometer range. Figure 3(b) and Figure 3(c) display SEM images of c-SPAN treated at 350 °C under atmospheric pressure (SPAN-350(AP)) and 400 °C under semi-atmospheric pressure (SPAN-400(SP)), respectively. The particle morphology of SPAN-350(AP) and SPAN-400(SP) was similar to SPAN-400(AP), suggesting that pressure and temperature variations do not significantly impact c-SPAN morphology.

Fig. 3.

SEM images of c-SPAN samples thermally treated at (a) SPAN-400(AP), (b) SPAN-350(AP), and (c) SPAN-400(SP). (d) XRD patterns, (e) Raman spectra, and (f) composition analysis results for the various samples (SPAN-400(AP), SPAN-350(AP), and SPAN-400(SP))

XRD and Raman spectroscopy were employed to determine the crystallinity and physical characteristics of the samples (SPAN-400(AP), SPAN-350(AP), and SPAN-400(SP)). Figure 3(d) shows the XRD patterns, where all samples exhibited a peak around 25°, corresponding to the (002) plane of the graphite structure [19]. Differences were minimal, primarily in peak intensity. Figure 3(e) displays Raman spectra of the samples, with typical S-S bond peaks at 469 and 929 cm?¹ observed in all, confirming the presence of C-S-S bonding in the c-SPAN [19]. Additionally, the D (1348 cm?¹) and G (1590 cm?¹) bands, associated with the carbon matrix, appeared in all samples, indicating successful c-SPAN synthesis regardless of conditions [19,20]. Composition analysis (Figure 3(f) and Table 1) showed similar carbon content across the three samples. However, SPAN-350(AP) and SPAN-400(SP) had about 10% more sulfur than SPAN-400(AP) (41.7%), with 46.4% and 45.6%, indicating sulfur content increases with adjusted pressure and temperature.

Table 1.

Composition ratios of various c-SPAN samples: SPAN-400(AP) (thermally treated at 400°C under atmospheric pressure), SPAN-350(AP) (thermally treated at 350°C under atmospheric pressure), and SPAN-400(SP) (thermally treated at 400°C under semi-atmospheric pressure).

Figure 4 presents the thermal stability of SPAN-400(AP), SPAN-350(AP), and SPAN-400(SP). Despite the higher sulfur content, the TGA results for SPAN-350(AP) and SPAN-400(SP) (weight loss of 78.5% and 82.1%, respectively) showed inferior thermal stability compared to SPAN-400(AP) (weight loss of 77.4%), likely due to the sublimation of unreacted sulfur. This finding suggests that optimizing sulfur content is critical to improving thermal stability.

Fig. 4.

Thermogravimetric analysis (TGA) results of various c-SPAN samples: SPAN-400(AP) (thermally treated at 400°C under atmospheric pressure), SPAN-350(AP) (thermally treated at 350°C under atmospheric pressure), and SPAN-400(SP) (thermally treated at 400°C under semi-atmospheric pressure).

To evaluate c-SPAN's potential in thermal batteries, we fabricated a thermal battery using the optimized SPAN-400(AP) sample and investigated its properties. Figure 5(a) shows the structure and a photograph of the fabricated thermal battery, where a single cell consists of a Li anode, electrolyte, cathode, and current collector in sequence. As demonstrated in Figure 5(b), the thermal battery fabricated with SPAN-400(AP) operated effectively under pulse current, indicating the suitability of the thermally stable c-SPAN for use in thermal batteries. Additionally, the internal resistance of the single cells was calculated using the pulse discharge data, as per Equation R = (V₁ – V₂)/ (I₁ – I₂)[8]. Here, R (Ω) represents the internal resistance of a single cell, V₁ (V) and V₂ (V) are the working voltage and pulse voltage, respectively, and I₁ (A) and I₂ (A) are the working current and pulse current. Using this equation, the internal resistance of SPAN-400(AP) was determined, as shown in Figure 5(c). The average internal resistance was calculated to be 72.7 mΩ, which is sufficiently low for thermal battery applications. These findings indicate that the c-SPAN material developed in this study holds promise for use in thermal batteries.

Fig. 5.

(a) Schematic and photograph of the thermal battery fabricated with SPAN-400(AP). (b) Pulse discharge curve and (c) internal resistance of the thermal battery using SPAN-400(AP).

4. CONCLUSIONS

This study demonstrates that c-SPAN, synthesized under appropriate conditions, is a viable cathode material for thermal batteries. The c-SPAN was produced using a straightforward ball milling and annealing method. Its properties were thoroughly characterized through microscopic evaluation and TGA analysis. The SPAN-400(AP) sample exhibited enhanced thermal stability, maintaining stability up to 500°C, highlighting the importance of appropriate sulfur content. The thermal battery with c-SPAN operated effectively, showing a minimal voltage drop of 0.17 V and a low internal resistance of 72.7 mΩ. These findings suggest that the c-SPAN developed here is a promising candidate for thermal batteries.

Article information Continued